Session GP1 - Poster Session IV.
POSTER session, Tuesday afternoon, October 30
Exhibit Hall B,

[GP1.001] Education and Outreach

[GP1.002] Coalition for Plasma Science (CPS) Education and Outreach Activities

The Coalition for Plasma Science (http://plasmacoalition.org/, CPS@plasmacoalition.org) is a group of ``\ldots institutions, organizations, and companies joining forces to increase awareness and understanding of plasma science and its many applications and benefits for society.'' Some examples of CPS educational activities to be presented are: (1) Construction and maintenance of a web page. The web page includes a ``Plasma Page,'' a compilation of brief, clear summaries of plasma-related news; a section (under development) with demonstrations for teachers and students; and a ``A Teacher's Guide to Plasma Science on the Web,'' a page that provides links to a wide range of plasma-related education sites. Most of the educational sites are analyzed for consistency with national science standards. (2) Luncheon presentations on plasma topics of broad interest to Members of Congress and their staffs. Speakers for these luncheons have addressed such topics as the value of plasmas in K-12 science education, and the use of plasma propulsion for space travel. (3) The organization of media panels at professional society meetings.

[GP1.003] Efforts in Public Relations on Fusion in Europe

An overview will be given of different published materials currently in use in Europe for public relations on fusion. We will also present a CD-ROM for individual and classroom use, containing (i) a general background on different energy forms, (ii) general principles of fusion, (iii) current research efforts and (iv) future prospects of fusion. This CD-ROM is currently in English, German, French, Spanish, Portuguese and Italian. Fusion posters developed in collaboration with CPEP in Dutch, French, German, Italian, Spanish and Portuguese will be shown. Several new brochures and leaflets intended to increase the public awareness on fusion in Europe will be on display.

[GP1.004] CPEP Fusion/Plasma Physics Education Materials/Activities

The Contemporary Physics Education Project (CPEP) is a not-for-profit organization of teachers, educators, and physicists (see http://cpepweb.org). The goals in its charter include the development and implementation of teaching materials about contemporary physics topics for use in the introductory courses. To this end, the CPEP Fusion/plasma group has produced the teaching chart, ``FUSION-Physics of a Fundamental Energy Source''. Ancillary materials including an Instructors Guide and a packet of classroom activities are under development. In order to promote effective classroom use of its educational materials, CPEP presents workshops for high school and college teachers. These workshops have been sponsored by or held in conjunction with a variety of organizations including; the APS/DPP, the AAPT, the Space Science Institute, the American Nuclear Science Teachers Association, The Allegheny Intermediate Unit, and the University of Pittsburgh at Greensburg. The chart is available in wall-size, poster-size and student notebook-size, and has been translated into six languages. Laminated versions of the poster and notebook size charts are also available as well as an overhead transparency of the chart. For more information, visit the CPEP/Fusion website (http://FusEdWeb.llnl.gov/CPEP/Chart.html)

[GP1.005] Promoting Pre-college Science Education

The Fusion Education Program, with support from DOE, continues to promote pre-college science education with educators and students. Projects pursued this year included an onsite laboratory-based 7-day educator workshop on plasma and fusion science; expansion of scientist visits to classrooms; and enhancements to the tours of the DIII-D Facility. In the workshop, each teacher built for their own classroom use a full wave rectifier power supply, a half-coated fluorescent light plasma display, an interactive vacuum demonstration unit, and a bench-top device to explore plasma physics. In addition, the teachers carried out Langmuir probe measurements of a simple air or argon discharge, performed simulations of particle trajectories using Excel, and visited a University Physics lab and the San Diego Supercomputer Center. Our ``Scientist in a Classroom" program has reached more than 1800 students at 18 schools. Our ``Starpower" interactive CD was awarded a ``5 star" rating in Physics Education. The most unique portion of the Fusion Education program is the continuing tours of the DIII-D facility attended by students, educators, and other groups.

[GP1.006] Educational Outreach at MIT PSFC

At the MIT PSFC student and staff volunteers work together to increase the public's knowledge of fusion and plasma-related experiments. Seeking to generate excitement about science and engineering, the PSFC hosts a number of outreach activities throughout the year, including Middle and High School Outreach Days. Key to the success of these tours is the interactive ``C-Mod, Jr.," which helps students understand magnetic confinement in MIT's Alcator C-Mod tokamak. The PSFC also has an in-school science demonstration program on the theme of magnetism. As ``Mr. Magnet" Technical Supervisor Paul Thomas brings a truck-load of hands-on demonstrations to K-12 schools, challenging students to help him with experiments. While teaching fundamentals of magnetism and electricity he shows that science is fun for all, and that any student can have a career in science. This year he reached 77 schools -- 30,000 teachers and students. We have also collaborated with the MIT Museum to create an interactive plasma demonstration device which students and the general public can use to create plasmas from different gasses. Pinch and deflection magnets are moveable along the axis of the display, allowing investigation of the magnetic behavior of plasmas.

[GP1.007] Physcial Models for the Classroom

The General Atomics fusion education program ``Scientist in the Classroom" now in its fourth year, uses scientists and engineers to present ``Plasma the 4th State of Matter," to students in the classroom. Physical models developed for this presentation are used to demonstrate plasma characteristics and concepts from topical areas associated with plasma producing devices; e.g. vacuum, electricity, and light.\par Using hands-on equipment, students see how magnets, gas pressure changes, and ionization potential levels affect plasmas. A piston, sealed volume, and vacuum chamber illuminate the ideal gas laws. Liquid nitrogen is used to explore thermodynamic temperature effects and changes in states of matter, through the cooling of different gases. The physical models used in this program were developed with the tight budget of teachers in mind, using simple designs and common commercial materials. The details of these models will be presented. Three very successful ``build-it" days have been sponsored to enable teachers to build these physics models for use in their own classrooms.

[GP1.008] Science Education Outreach Activities in the Fusion Energy Division of UCSD’s Center for Energy Research*

Since 1995, the Fusion Energy Division of the Center for Energy Research at UCSD has been engaged in a variety of volunteer activities in science education outreach. FED staff have developed demonstration tools on energy and plasma science which are used effectively with middle and high school students as well as teacher/student groups at: the APS DPP Plasma Expos and the San Diego Co. Educational Technology Fair. These demonstration tools have proven effective in communicating with elementary students at community science and technology exhibits at the Reuban H. Fleet Science Center (San Diego) and in elementary school classes. UCSD scientists have also participated as team members of the GA Fusion Group’s programs: "Scientist in the Classroom" , and the two Plasma Institutes for in-service science teachers. In the coming year, we plan to: 1) expand the "Scientist in the Classroom" to home-schooled children in San Diego; 2) participate in local elementary school Family Science Nights; and 3) assist in training a new group of future San Diego Unified School District ninth grade physics teachers.

[GP1.009] HAMPTON UNIVERSITY (HU) CENTER FOR FUSION RESEARCH amp; TRAINING (CFRT) SUMMER HIGH SCHOOL FUSION SCIENCE WORKSHOPS: 1996 THROUGH 2000

The HU CFRT has conducted Summer High School Fusion Science Energy Workshops during the summers of 1996 through 2000 - a total of five summer fusion science workshops for high school students. These workshops are one of many programs and activities of the center organized and focused around the education, motivation, retention and research experiences for young scholars especially under-represented minorities and female high schoolers in fusion science and related areas. These workshops consist of a very productive curriculum and palate of exciting activities that provides the students the opportunity to experience what scientific research in fusion science is all about. These activities include an intensive program of instructions on basics of fusion energy science, modeling of tokamaks, visualization of scientific data and on how to use computers to conduct scientific research. Through experimentation and discovery based on computer simulations of chaos and other nonlinear phenomena in tokamaks, students learn and understand what are the relevant patterns and structures to look at in fusion devices, what is the underlying scientific hypothesis, and what conclusions can be reached based on results.

Here we will present the highlights and the unique successes of the five workshops conducted from the Summer of 1996 through Summer of 2000. We will also discuss the fusion science workshop model, its content and scope. These workshops are funded by DOE OFES, and NASA through its SharpPlus program administered by the Quality Education for Minority Network (QEM).

[GP1.010] Plasmas after the Big Bang and in the Laboratory

This paper attempts to illustrate to students and the general public relationships between laboratory plasma physics and cosmological plasmas. The development of fusion energy is often likened to harnessing the power of the sun and stars. Indeed, many of the physical conditions, as well as scientific processes, are akin. Thus laboratory experiments have advanced scientific understanding of our universe. The plasma temperatures in the interior of fusion experiments can exceed the hundred million degree temperature of the universe posited a few hours following the Big Bang. The Big Bang theory says that this hot plasma universe cooled during the first million year period when radiation and matter strongly interacted. Ions and electrons recombined to form atoms, from which today’s universe emerged. Inertially confined plasmas are dominated by similar radiation-matter interactions to achieve fusion. Magnetically confined edge plasmas similarly radiate energy and recombine to shield experiment walls from melting and erosion caused by impinging high temperature plasma. Understanding the structure and dynamics of laboratory plasmas can thus provide glimpses into the past of our universe.

[GP1.011] LAPTAG, Encouraging High School students to Consider Physics Related Careers

The Los Angeles Physics Teachers Alliance Group (LAPTAG) represents high school physics teachers from the entire Los Angeles area. It was formed in 1993 Over the years between twenty and thirty schools have participated. Our Website is at http://coke.physics.ucla.edu/laptag and web service is provided to schools without servers or Internet access. LAPTAG encourages communication between high school and college/university physics teachers by providing regular meetings, tours of laboratories at UCLA and other institutions, and discussion of curricular issues. LAPTAG also provides unique opportunities for student involvement in research projects. Our first project was a distributed seismometer experiment in which ten schools received seismometers. LAPTAG provided a Web based astronomy class in which studied a variable star. During the past three years, we have constructed a plasma device and developed a high school plasma curriculum. These laboratory experiences engage science students and encourage them to enter physics related careers

[GP1.012] Construction of a High School Plasma Laboratory

Members of LAPTAG have constructed a laboratory plasma device for use by high school students. Currently students are using the device to do experiments to measure the velocity of ion acoustic waves The plasma is contained in a 0.25 m^3 chamber that is initially evacuated by a turbo vacuum pump to ª 10^-7 Torr. It is produced by a Helicon source which is comprised of a 250 watt 13.6 MHz RF power supply, a double loop antenna on the outside of a glass section surrounded by solenoidal magnets. The magnetic field in the source can be as much as 100 Gauss. A steady state plasma then streams into an unmagnetized experimental. Data is taken by means of a Langmuir probe connected to a 175 MHz digital oscilloscope. Current upgrades of the device include automating the motion of Langmuir probe with a computer driven stepper motor and the use of digitizers and computers to facilitate data acquisition. The high school teachers and students are directly involved in the machine upgrade. They have already constructed the probe drive and will write LABVIEW based software to control it as well as the data acquisition. Other diagnostics such as energy analyzers and a monochrometer will be installed shortly.

[GP1.013] Characteristics of the LAPTAG High School Plasma

In 1999, a group of high school teachers in the Los Angeles Physics Teachers Alliance Group (LAPTAG) successfully constructed a plasma device for high school research. Since then groups of high school students have collected data to characterize the plasma. The plasma has a helicon source which produces an Argon plasma that streams into an unmagnetized chamber. The plasma density is obtained from the ion saturation current to a Langmuir probe. In our plasma Te>>Ti. In these experiments the plasma density is 10^9-1010 cm^-3 . The electron temperature is measured from the dispersion of ion acoustic waves, discussed in another poster in this session. The plasma potential is measured by sweeping the Langmuir characteristic curve. We will present data on the spatial distribution of the plasma potential and discuss the radial electric field in the device. Measurements of the plasma production as the RF source power and background gas pressure are changed will be presented as well. Using these measurements we have generated a list of possible future experiments the device may be used for.

[GP1.014] Ion Acoustic Waves, A High School Plasma Experiment

Over the last three the Los Angeles Physics Teachers Alliance Group (LAPTAG) has built a plasma device and designed experiments for high school students to learn about plasma properties and behavior. One of the first experiments performed by small student groups (two to three students at a time) is to create ion acoustic wave tonebursts in an Argon plasma, measure the wavelength and frequency of the wave and thereby calculate the velocity of the wave. A grid antenna immersed in the plasma, which is pulsed by a function generator, creates the waves. Measurements are made using a Langmuir probe and read out on a digital oscilloscope. From this information students calculate values such as the temperature of the plasma, the plasma density and percent ionization of the plasma. In order to do these experiments students must understand what plasma is, how plasma can be created using a helicon source, how to use an oscilloscope and many other aspects of the plasma chamber involved in the experiment. Other experiments are currently being done on the device and still others are being designed. For more information visit the LAPTAG website (http://coke.physics.ucla.edu/laptag).

[GP1.015] Using Plasma Physics to Enhance the High School Physics Curriculu

Faculty and student members of the Los Angeles Physics Teachers Alliance Group (LAPTAG) have constructed a plasma machine on the ULCA. Dr. Gekelman, the faculty advisor, provides information and materials on plasma physics via the Web and lectures to high school faculty and students. Faculty members then transfer the information to students at their respective schools and schedule time for experiments on the machine. A lab manual and curricular materials suitable for high school students is being developed using a lab based, discovery approach. The manual is available as a pdf document on the LAPTAG website (http://coke.physics.ucla.edu/laptag/plasma_exp.dir/laptag_plasma.htm). Introducing plasma physics into the high school curriculum provides a 20th century application of classical physics concepts that support and motivate student interest in physics. Students from LAPTAG schools use state-of-the-art computers, software, and equipment to perform developed labs and to design experiments of their own. Collaboration exists between students and faculty from different schools and the university. Learning physics concepts takes place in the context of a "science community" that realistically demonstrates the scientific process to students.

[GP1.016] Undergraduate Research

[GP1.017] Saturation Spectroscopy of Molecular Iodine using a CW Dye Laser

We are investigating the technique of saturation spectroscopy (two counter propagating photons incident on the same sample). This technique offers the advantage of a Doppler- free spectrum. We currently have a Doppler-broadened linewidth of 400 MHz and we expect a reduction to around 500 kHz. Potentially, this technique can result in a more accurate determination of the velocity shear in the rotation profile of an unneutralized magnesium ion plasma. We are planning to implement a setup based on a Couillaud-Holbrow[1] design. The apparatus uses two antiparallel beams incident on an I_2 cell. One strong beam is used as a pump beam and saturates the iodine molecules in its path. The second weak beam is a probe beam that measures the absorption of the iodine in the saturated area. The third beam is a weak reference beam that measures the absorption in the unsaturated iodine. When the two weak beams are differenced, they will give the absorption spectrum of the hyperfine structure of molecular iodine around \lambda = 560 nm. [1] C.H. Holbrow; B. Couillaud. Hyperfine Interactions, vol.37, p.185-208 (1987). Supported by the National Undergraduate Fellowship Program and ONR grant (N00014-96-1-0239).

[GP1.018] Modeling of Neutral Transport In the SOL of Tokamak Plasmas

Plasma refueling and sputtering of impurities from the main wall are problems which depend upon the transport of neutral particles in the scrape-off-layer of a tokamak. We have developed a 1D model of SOL neutral transport which can quickly predict the core fueling fraction and the energy distribution of the neutral flux incident on the wall. Detailed atomic physics of neutral interaction with background plasma including multiply charged impurities will be included. We will compare our calculated profiles with measured D\alpha emission data.

[GP1.019] Analytic Modeling of Interaction Between Resonant Toroidal Magnetic Field Lines

We present a description of coupling between resonant magnetic field lines, in a toroidal magnetic field. We explore a mechanism by which neighboring or harmonic magnetic islands expand, and eventually make contact along a separatrix, allowing previously independent field lines to interact. Using a Hamiltonian formulation of the magnetic field line equations to model such systems analytically, we characterize the spatial trajectories and interactive behavior of field lines coupled through this mechanism.

[GP1.020] Comparison of Experimental Measurement of Carbon Flow Velocities and Temperatures to UEDGE Predictions in the DIII-D Divertor

From the Doppler broadening of emission profiles of carbon impurity lines, the temperature and parallel velocity distributions of carbon ions in the divertor region of DIII-D have been deduced in an ELMing H-mode discharge with 9\,MW of NBI. We have analyzed the following emission lines: CI (^3P \rightarrow ^3P^0, 9087ÅCII (^2P^0 \rightarrow ^2S, 6579.7ÅCIII (^3P^0 \rightarrow ^3S, 4648.8Åand CIV (n=7 to n=6 transitions at 7726ÅIn previous work we have shown that the CIV emission results from charge exchange from C^+4 so the temperature and flow velocity inferred represent that of C^+4. The spectroscopy system provides 7 vertical and 5 tangential view chords, so it is possible to reconstruct a flow pattern of carbon ions across the divertor region using the Zeeman splitting of their \pi components for deduction of the radial position along the viewing chord. We will compare the measure carbon flow for the C^+1, C^+2 and C^+4 ions to UEDGE 2D fluid model predictions.

[GP1.021] Modeling Electric Arcs in a Force-Free Environment

The behavior of the inter-electrodal region of an electric arc in a force-free environment is modeled. The assumption of weightlessness allows for the creation of horizontal arcs which are not subject to bowing by buoyancy effects. Consequently, axial symmetry obtains and transport by convection and hydrodynamic turbulence are absent. The arc is assumed long, so that electrode effects may be ignored. Additionally, the pressure is taken to be large enough that the mean free paths of the electrons, ions and neutrals are small compared to the other dimensions of the problem and, as a result, all the particles are in local thermodynamic equilibrium. The model yields radial temperature profiles for a range in operating conditions in which energy loss is dominated by thermal conduction at one extreme, and by radiative emission at the other. Model predictions are compared with experimental results.

[GP1.022] Demonstration of Mass Production Layering of Inertial Fusion Energy Targets using a Room Temperature Surrogate for DT

An inertial fusion energy reactor requires the mass production of targets. The process includes filling the targets with DT and re-distributing the DT at cryogenic temperatures into a uniform layer inside of the spherical target capsules. Proof-of-principle, mass production layering experiments were conducted at room temperature using capsules filled with a surrogate, neopentyl alcohol, for DT. The self-heating of\, DT by beta-decay of the tritium is simulated with infrared light. The self-heating produces a preferential sublimation-condensation effect from the thick (warmer) regions of DT or alcohol to the thin (colder) regions of DT or alcohol. This creates a layer of uniform thickness when capsules are placed in an environment that uniformly removes heat from the capsule surface. A fluidized bed, which can handle a vast number of targets at once, was investigated for producing the required thermal environment for capsule layering. Capsules were filled with neopentyl alcohol by diffusion and by injection through a needle.

[GP1.023] Design Construction and Test of an FM Profile Reflectometer for the Electric Tokamak

The design, construction and test of a FM microwave reflectometer suitable for edge electron density profile measurement in the Electric Tokamak (ET) at UCLA are presented. ET is a large, low curvature tokamak (R = 5 m, a = 2 m) that will operate initially at low magnetic field, B < 3 kG, and low density, n < 8 x 10^12 cm^-3. The reflectometer will operate over the 10 – 16 GHz range in either O- or X-polarizations. The purpose of this system is to complement ongoing diagnostic development work by providing density profiles in edge regions of the plasma where magnetic field, turbulence, and RF wave measurements will be made. Preliminary plasma data may also be shown.

*Supported by US DoE under grantDE-FG03-86RE-532225, Task IV

[GP1.024] Plasma Calorimeter Calibration Using a 2 MV Van de Graaff Accelerator

The Laboratory for Laser Energetics (LLE) at the University of Rochester uses a set of plasma calorimeters to study the plasma energy of laser-irradiated targets. Seventeen of these plasma calorimeters are mounted around the implosion site on the OMEGA target chamber. Each calorimeter consists of two 25 \mum thick tantalum foils and a copper reference plate of equivalent thermal mass. A set of type-E thermocouples is used to make three differential temperature measurements (foil 1-reference, foil2-reference, and foil 1-foil 2). We are developing a technique to calibrate the response of these plasma calorimeters using proton beams of several hundred keV from the 2MV Van de Graaff accelerator located at the Geneseo Nuclear Structure Laboratory (GNSL). Using a pair of electrostatic deflector plates, the proton beam is directed toward each of the calorimeter foils. The protons deposit their energy on the calorimeter foil, causing an increase in the foil temperature. The calorimeter thermocouple output signals are sent to a LabVIEW acquisition system for analysis. The beam current is measured directly by sending the charge deposited on the calorimeter through an electrometer. A biased grid is placed immediately in front of the calorimeter foils and is used to provide electron suppression. A gold foil is also placed in front of the calorimeter foils and is used as a second monitor of the beam current via Rutherford scattering. The results of these calibrations will be discussed.

[GP1.025] Tertiary Neutron Measurements by Carbon Activation

The yield of tertiary neutrons with energies greater than 20 MeV has been proposed as a method to determine the \rhoR of ICF targets. Carbon activation is an appropriate measurement technique because of its high reaction threshold and the availability of high-purity samples. The ^12C (n, 2n) ^11C reaction has a threshold of 18.7 MeV well above the 14.1 MeV primary DT neutron energy. The isotope ^11C decays with half-life of 20.3 min and emits a positron, resulting in the production of two 511-keV gamma rays upon annihilation. The positron decay of ^11C is identical to the copper decay used in the activation measurements of 14.1 MeV primary DT yields; therefore, the present copper-activation gamma-detection system can be used to detect the tertiary-produced carbon activation. Because the tertiary neutron yield is more than six orders of magnitude lower than primary neutron yield the carbon activation diagnostic requires very pure carbon samples, free from any positron emitting contamination. In recent years we have developed carbon purification, packaging, and handling procedures that minimize the contamination signal to a level low enough to use carbon activation for tertiary neutron measurements in direct-drive implosion experiments with DT cryogenic targets on OMEGA. A concept of implementing a carbon activation system on NIF will be also discussed.

[GP1.026] LOW-FREQUENCY OSCILLATIONS IN A TOROIDAL PURE-ELECTRON PLASMA

An electron plasma with n\approx 3\times 10^6 cm^-3 is confined in a partially toroidal trap with a magnetic field of 196 G. Trapping times on the order of 100 \mu s are achieved. We actively control the horizontal electric field generated by auxiliary electrodes. The horizontal electric field functions for toroidal electron plasma equilibria as the analog of a vertical magnetic field for toroidal MHD equilibria. During trapping, oscillations are observed with frequencies on the order of 100 kHz. These oscillations have frequencies that are proportional to 1/B and also depend on the horizontal electric field strength. We will report on our investigation into the nature of these oscillations using a new set of four wall probes in addition to a five-tip insertable floating potential probe. The wall probes are spaced both toroidally and poloidally around the chamber to allow us to observe the large-scale spatial structure of the oscillations.

[GP1.027] Measuring the Langmuir Probe Characteristic in a Pure Electron Plasma

Electron plasmas of minor radius 4.5 cm, major radius 43 cm and density n\approx3\times10^6 cm^-3 are created in a toroidal magnetic field of strength 196 G. Trapping times on the order of 100 \mu s have been obtained and oscillations in trapped plasmas have been observed. A 5-tip insertable probe has been used to measure the floating potential profile in electron plasmas. The electron density is determined from these measurements. Secondary electron emission is observed, indicating the presence of energetic electrons. We will report measurements of the Langmuir probe characteristic (I-V Curve), and extract the plasma temperature from these results. This measurement represents an investigation into the thermal and secondary electron emission modifications to the Child-Langmuir Law for vacuum diodes.

[GP1.028] Study of Magnetic Damping in Liquid Metal Surface Waves

Knowledge of liquid metal surface waves and instabilities provides insight regarding turbulence in plasmas and the magnetohydrodynamic (MHD) model used to describe plasmas generally. Such work is also critical in the development of liquid lithium walls to be used in fusion reactors. The Liquid Metal Experiment (LMX) is designed to study magnetically induced damping of liquid gallium surface waves by driving such waves in the presence of a magnetic field. Previous work measured the dispersion relation and confirmed that a magnetic field aligned perpendicularly to the direction of wave propagation has no effect. Current work improves the experimental design to measure the damping effect of a magnetic field aligned parallel to the direction of wave propagation.

[GP1.029] Electron Bernstein Wave (EBW) Polarization Measurements on CDX-U

Mode-converted (MC) EBWs offer an attractive path for electron temperature measurement, heating, and current drive in overdense plasmas (ømega_pe\ggØmega_ce). A quad-ridged antenna was installed in CDX-U with a movable limiter, which shortens electron density scale length at the MC layer and hence optimizes MC efficiency. Electrostatic EBWs are expected to MC to X-mode electromagnetic waves. Measurements were made with both the X- and O-mode aligned antennas, and the X/O ratio was calculated. An X/O ratio >2 was observed with the antenna near the MC layer, in contrast to a ratio of 1.2 measured previously with an antenna outside the vessel. A ratio of \sim1 was seen with the antenna far from the MC layer, possibly due to reflections between the plasma and vessel wall causing polarization scrambling. Reduction of the X/O ratio was observed when the limiter was extended, likely due to polarization mixing caused by reflection or refraction at the limiter surface in front of the antenna.

[GP1.030] Simulated Symmetric Plasma Transport

It can be shown analytically that symmetric plasmas, which are composed of positive and negative ions of equal mass, are transported across magnetic fields due to polarization current and the resulting E x B force [A. Y. Wong et al, pre-print]. Such transport would be useful for heating in magnetic confinement fusion reactors, which are bounded by strong magnetic fields. The theoretical requirement for transport, ømega _pi/Ømega _i>>1, has been confirmed computationally using a bounded two dimensional electrostatic particle simulation code [V. K. Decyk and J. M. Dawson, J. Comput. Phys. 30, 407 (1979)]. It is shown that for B = 0.5 T and n_i \ge 10^10 cm^-3, there is little change in the beam's velocity and little spreading of the beam.

[GP1.031] Reconnection Scaling Experiment (RSX) – Engineering of the experiment and first data on kink stability of a current channel.

Major subsystems of the RSX experiment have been designed, built and brought on line by undergraduate students this year. We describe a relatively inexpensive magnet pulsed power system and a novel probe drive. We describe the engineering and construction of a 5kA, 700 Volt, critically damped, 30 msec pulse electrolytic capacitor bank and SCR switching network that drives the magnet coils. Failsafe design, ease of maintenance and operator safety were important factors. SPICE circuit simulations and performance data are compared. A novel axial probe drive that allows one probe to explore all of a cylindrical volume is described. This includes axial, angular and radial motion with high relative repeatability and absolute position accuracy. We opted for vacuum sanitary and plasma compatible position sensing technology with motion feedback to reduce the accuracy constraints on the probe movement mechanisms that remained inside the vacuum vessel. We show the first RSX data, which was taken to address the kink stability boundary of the single current channels we create with the RSX plasma guns.

[GP1.032] Stability Tests of Hydrodynamic Flows in Water for Laboratory Study of Magnetorotational Instability

Magnetorotational Instability (MRI) is a powerful candidate mechanism for the fast transport of angular momentum in magnetized accretion disks. In an accretion disk, when the mass spirals in towards the stellar object, due to gravity, the velocity increases to conserve angular momentum. When the force of gravity is balanced with the centripetal force, the viscosity pulls it in towards the central compact object, which is too small to explain the fast transport of mass, so there must be another reason. Hydrodynamic (HD) instabilities, like the Rayleigh instability, are ineffective in producing turbulence in accretion disks because it requires a negative gradient of specific angular momentum. Magnetohydrodynamics provides a better description of plasma in hot accretion flows where angular momentum has an extra degree of freedom due to the presence of the magnetic field. The radial transport of angular momentum due to MRI will hopefully explain how mass gets accreted onto a stellar object. Despite the popularity of MRI, it has never been tested in the laboratory. In an attempt to demonstrate MRI in the laboratory, a magnetized couette flow experiment using gallium is proposed. Before gallium is used, a prototype experiment using water has been constructed to study linear and nonlinear HD instability in the (omega1, omega2) space. HD stability can be monitored using particle imaging velocimeter techniques, which will serve as a reference for effects due to the MRI mechanism.

[GP1.033] Characterization of Low Frequency Magnetic Oscillations in the MRX Neutral Sheath

Magnetic reconnection is an important mechanism for converting magnetic energy into plasma thermal energy, and is thought to heat the solar corona and accelerate particles in the earth's magnetosphere. The main objective of the Magnetic Reconnection Experiment (MRX) is to study the fundamental physics of magnetic reconnection in a laboratory environment. Current sheets that are formed in the MRX contain features such as density gradients and strong cross-field currents, which can drive a variety of unstable fluctuations^1. Recent experiments studied high frequency oscillations in the electrostatic and magnetic fields associated with current sheet formation and reconnection. These oscillations were identified as lower hybrid drift instabilities^2. Low frequency oscillations were also observed near the end of reconnection, but were not fully characterized. The objective of the present study is to measure the properties of these oscillations, and to try to understand their role in the reconnection process.

^1M. Yamada et al., Phys. Plasmas, vol.7, 1781 (2000) ^2T. Carter, submitted to PRL.

[GP1.034] Plume Comparisons between Segmented Channel Hall Thrusters

Angular ion flux plume measurements were taken in several configurations of segmented channel Hall thrusters. The configurations differed by the placement of relatively short rings made from materials with different conductive and secondary electron emission properties along the boron nitride ceramic channel of the thrusters (these have been shown to affect the plume [1]). The ion fluxes are compared with ion trajectory simulations based on plasma potential data acquired with a high speed emissive probe [2]. Preliminary results indicate that in addition to the physical properties of the segments, the plume angle can be strongly affected by the placement of segmented rings relative to the external and internal walls of the channel.

[1] Y. Raitses, L. Dorf, A. Litvak and N. J. Fisch, Journal of Applied Physics 88, 1263, 2000 [2] D. Staack, Y. Raitses, N. J. Fisch, Parametric Investigations of Langmuir Probe Induced Perturbations in a Hall Thruster, DPP01 Poster Presentation

This work was supported by the U.S. DOE Contract No. DE-ACO2-76-CHO3073.

[GP1.035] Summary of the Madison Dynamo Experiment

A spherical dynamo experiment has been constructed at the University of Wisconsin's liquid sodium facility. The goals of the experiment are to observe and understand magnetic instabilities driven by flow shear in MHD systems, investigate MHD turbulence for magnetic Reynolds numbers of \sim100, and understand the role of fluid turbulence in current generation. Magnetic field generation is possible for only specific flow geometries. We have studied and achieved simple roll flow geometries in a full scale water experiment. Results from this experiment have guided the design of the sodium experiment. The experiment consists of a 1 m diameter, spherical stainless steel vessel filled with liquid sodium at 110 Celsius. Two 100 Hp motors with impellers drive flows in the liquid sodium with flow velocities \sim 15 m/s. A gaussian grid of Hall probes on the surface of the sodium vessel measure the generated external magnetic field. Hall probe feed-thru arrays measure the internal field. Preliminary investigations include measurements of the turbulent electromotive force and excitation of magnetic eigenmodes.

[GP1.036] Relationship between Impurity Dispersal Patterns and Local Edge Plasma Conditions in Alcator C-Mod

The understanding of impurity transport in the edge plasmas is crucial for the design of future fusion reactors. Impurities in the core plasma can dilute the fuel and produce a significant amount of radiation, which will lead to the loss of input power. A system has been developed for studying impurity transport behavior in the edge plasma of Alcator C-Mod by viewing impurity emission patterns (plumes) generated by local gas injection.(S. Gangadhara, et al., J. Nucl. Mater., \textbf290-293) (2001) 598. Dispersal patterns of C^+1 and C^+2 ions have been observed for a wide range of local plasma conditions, including variations in density, temperature, parallel velocity and radial electric field. A database of impurity dispersal patterns is being assembled to correlate variations in the extent and asymmetry of these patterns along and across the field lines with local plasma conditions.

[GP1.037] Adaptation of TRIPND Field Line Tracing Code to a Shaped, Poloidal Divertor Geometry

The magnetic field line tracing code TRIPND(T.E.\ Evans, Proc.\ 18th Conf.\ on Control.\ Fusion and Plasma Phys., Berlin, Germany, Vol.\ 15C, Part~II (European Physical Society, 1991) p.~65.) has been modified to use the axisymmetric equilibrium magnetic fields from an EFIT reconstruction in place of circular equilibria with multi-filament current profile expansions. This adaptation provides realistic plasma current profiles in shaped geometries. A major advantage of this modification is that it allows investigation of magnetic field line trajectories in any device for which an EFIT reconstruction is available. The TRIPND code has been used to study the structure of the magnetic field line topology in circular, limiter tokamaks, including Tore Supra and TFTR and has been benchmarked against the GOURDON code used in Europe for magnetic field line tracing. The new version of the code, called TRIP3D, is used to investigate the sensitivity of various shaped equilibria to non-axisymmetric perturbations such as a shifted F coil or error field correction coils.

[GP1.038] Measurements of rotating waves in ALEXIS

Theory predicts that in the presence of radial electric field shear, rotating waves can propagate in a plasma. We hope to find these waves in ALEXIS, the Auburn Linear Experiment for Instability Studies. A circular array of six Langmuir probes will be used to measure the fluctuating density of the plasma at a fixed axial location in the plasma. If these waves are present, we expect to observe a phase difference between each of the probes. This presentation will show the design and construction of the probe array and preliminary experimental data.

[GP1.039] JET, JT-60U, FTU and Textor

[GP1.040] High confinement high density in quasi-stationary low triangularity ELMy H-mode discharges on JET with impurity seeding

ELMy H-mode discharges are characterised by a decrease of confinement and fuelling efficiency when applying heavy gas puff to increase the density. Nevertheless quasi-stationary (up to 12 energy confinement times) discharges could be achieved with averaged values of n/n_GW\geq0.85, H98(y,2)\simeq1 and \beta_n\simeq1.8 in the phase following the D puff ("afterpuff" regime). The heavy D puff degrades the confinement but brings the density near n_GW thanks to the Ar seeding which counteracts the decrease of fuelling efficiency. After the switch-off of the strong D puff H-mode confinement quality is recovered whereas the density can be maintained by a careful refuelling in D and Ar. This regime shows peaked density profiles and moderate Z_eff (\simeq2). Best results were obtained with the X-point lying on the septum of the divertor. Effects of X-point position and plasma configuration changes as well as ELM behaviour will be discussed.

[GP1.041] Power deposition in the JET MKIIGB divertor during impurity seeded discharges

One of the unsolved problems for a next step fusion device is the handling of the power flux into the divertor in particular the rapid heat deposition caused by Edge Localized Modes (ELMs). In recent JET experiments with the MKII Gas Box divertor, ELMy H-Mode discharges with impurity seeding of argon and nitrogen have been investigated. A possible method of mitigating the power flux to the divertor during ELMs is to seed the plasma with impurities. In nitrogen seeded discharges a measured radiated power fraction of up to 85% of the heated power is measured, accompanied by a transition to Type-III ELMs and a loss of confinement (H98(Y,2)=0.7). On the other hand, in argon seeded discharges Type-I ELMs are still observed after argon injection at a radiation level of up to 65%, but without a loss of confinement (H98(Y,2)=1.0). The power load on the target are studied with high time resolution IR thermography (65microseconds) to distinguish between the power fraction deposited during and in between ELMs. For a limited time no significant power load is observed after the injection of either impurity between ELMs and a reduced power load during ELMs. The efficiency of power detachment for Type-III and Type-I ELMs will be discussed.

[GP1.042] Role of magnetic shear and ExB flow shearing rate in JET ITB discharges

A database of plasma discharges has been selected in different experimental conditions (with and without an ITB; with positive or negative magnetic shear), to study the role of the magnetic shear and of the shearing rate. In most of the discharges ion and electron temperature profiles exibit identical features concerning the location and the timing of the transport barrier, but sometime the barrier occurs on electrons only. By taking into account the "correct" dependence of the magnetic shear in the linear growth rate evaluation, it is qualitatively possible to explain the radial location, the time of formation and the time evolution of different kinds of transport barriers in terms of the accepted mechanism based on the ExB shear flow suppression of ITG-driven turbulence. This is true for almost all the discharges of the selected database; however in some discharge the picture is not completely clear, either due to experimental errors or to the presence of an additional/alternative mechanism of turbulence suppression. The point will be discussed.

[GP1.043] CONTROL OF IMPURITY ACCUMULATION BY CENTRAL HEATING

JET radiative mantle experiments in the ELMy H-mode regime have produced high confinement plasmas with densities close to the Greenwald limit. In the septum configuration, high performance plasmas (H_97*f_GWD >0.8) were obtained with two gas injection phases: continuous D_2 and Ar fuelling, followed by the "after-puff" phase when both gases injection rates are reduced. Confinement degradation is observed when Ar accumulates in the plasma core. This correlates with disappearance of sawtooth activity, when q(0) becomes greater than unity. In order to improve the q(0) stationarity, 1-3MW of ICRH power was added to the main NBI heating. The RF resonance layer was located on axis to increase the central electron temperature, keeping q(0)<1. Sawteeth were maintained and core impurity accumulation was prevented, resulting in quasi-steady state, high performance discharges with high Ar content. Sawteeth loss, followed by discharge degradation, is also a common observation in JET plasmas with peaked density profiles. The use of ICRH in these discharges, to avoid loss of confinement, is assessed.

[GP1.044] Stability of JET Discharges with Zero Core Current Density

Injection of lower hybrid heating and current drive into the current ramp-up phase of JET discharges can produce extremely reversed q-profiles characterized by a core region (r/a less than 0.2) of zero current density (within Motional Stark Effect diagnostic measurement errors) and q greater than 1 everywhere [1]. Electron temperature measurements show sawtooth-like collapses and the presence of an internal transport barrier. The core current density does not appear to go negative, although TSC/LSC code modeling indicates that the drive for this is present. A 2-D, two fluid simulation predicts that discharges with negative core current would be unstable to n=0, m=1 modes. The possibility that these modes redistribute the core current to prevent it from becoming negative is examined.

[1] N. C. Hawkes, et al., Phys. Rev. Lett. (accepted July 2001).

[GP1.045] INTERNAL FLUX SHAPING IN JET TRANSPORT BARRIER.

Starting from the evidence that in the reversed sheared discharges a deeper penetration of the plasma shaping is achievable, a set of experiments have been performed with different elongation and triangularity; this has allowed to study the effects of the internal flux surfaces topology on the Internal Transport Barrier. Although the plasma performances, in the triangularity scan (0.2–0.45), have been strongly conditioned by the presence of giant ELMs, it seems that, at the JET aspect ratio, the plasma triangularity is playing a negative role. For these configurations, an analysis of the edge ballooning stability has been carried out by using the code IDBALL. By selecting discharges with an identical time evolution for the macroscopic parameters (to get the same current profile), it comes out a positive role of the plasma elongation. A preliminary analysis of the transport properties is reported.

[GP1.046] Influence of varying aspect ratio on confinement

For the first time on JET,experiments were realized varying the aspect ratio(A) in the interval 4.09-3.65(a 12 % excursion).The aim was to study the effect of the variation of A on cinfinement of ELMy H-mode.The variation of A was achieved varying the major radius,keeping fixed the minor radius(a=0.88m), at fixed NBI power(7MW).Plasma parameters were B=1.7T,IP=1.2MA,ne=5 10^19 m^-3, upper triangularity~0.32(obtained at A=3.65)-0.4(at A=4.09), lower triangularity~0.4,elongation k=1.8,Grenwald fraction 0.8,Te=2.6-3.2keV.The discharges exhibited different ELM behaviour, most likely due to the difference in triangularity.The measured confinement time (\tau)was constant within less than 5%.Using the scaling expression of ELMy H-mode confinement time ITER98(y,2), \tau~A^1.4, an increment of at least 16% was expected.Therefore the dependence of confinement on A seems to be weaker than that estimated using the scaling expression.The possible improvement of confinement time with increased upper triangularity could only partially compensate the effect of aspect ratio.The relevance of this result in terms of dimensionless physics parameters will be analysed.

[GP1.047] Initial Pellet Spectrometer Results from JET

Initial spectroscopic measurements have been made of the Stark Broadening of the Balmer alpha and beta lines produced in the luminous cloud surrounding deuterium pellets injected into JET H-mode plasmas heated by NBI in combination with ICRH [1]. Pellets were injected in an extended sequence of 10 or more pellets launched from the high field side (HFS). One pellet data point was also captured for low field side (LFS) launch. These lineshape measurements are time integrated over the full duration of each observable pellet event, and they generally reflect the maximum broadening produced in the pellet cloud. For HFS launch the broadening is generally consistent from pellet to pellet in the sequence with a typical Balmer-alpha linewidth (FWHM) of \sim1.5nm. However, in some pellet sequences linewidths vary from \sim1.2 to 2nm. The observed broadening is suitable for the determination of electron density in the pellet ablation cloud. These density results are discussed. [1] Lang, EPS, Portugal (2001), P3.012

[GP1.048] Experimental Studies of Alfven Mode Stability and Instability Regimes on the JET Tokamak

Experiments have been performed on JET to study in detail the linear stability property and the non-linear evolution of Alfvén modes in conventional and advanced tokamak scenarios with internal transport barriers. For the first time the systematic dependence of the damping rate of low-n AEs on the plasma beta, the normalized Larmor radius, the magnetic shear and safety factor, and the plasma shape was reconstructed experimentally. The AE damping rate can be accurately measured in real-time, suggesting the development of a feedback system to prevent access to unstable domains. The dynamics of quasi-periodic, upward chirping, Alfvén cascades in reversed shear equilibria is qualitatively explained in term of excitation and non-linear saturation of Energetic Particle Modes. This work has been conducted under the European Fusion Development Agreement.

[GP1.049] Design of a High Power Prototype for the new JET-EP ICRF antenna

A high power prototype (HPP) of a new ICRF antenna for JET ("JET-EP antenna") is being designed and constructed in a collaborative effort between Oak Ridge National Laboratory, Princeton Plasma Physics Laboratory, and the European Fusion Development Agreement-Joint European Torus. The JET-EP launcher is designed for 8 MW input power (~9 MW/m^2) at 30-55 MHz. Current straps are arranged in a 4 poloidal by 2 toroidal array, minimizing voltage. A modified resonant double loop (RDL) matching circuit uses internal capacitors, and passively accommodates rapidly changing plasma loads. The HPP, consisting of one antenna quadrant, will be tested at ORNL in vaccum at \geq 33kV pk, 920A rms capacitor voltage and current. Innovative features to be tested include the modified RDL circuit, flanges which allow capacitor replacement without antenna removal, a low characteristic impedance vacuum feed line, and integral matching transformer.

[GP1.050] Influence of Ar Recycling and Divertor Configuration on Confinement in JET Radiating Mantle Discharges

Discharges have been obtained on JET in the MkII-Gas Box divertor using impurity seeding with Ar for H-mode plasmas to produce a radiating mantle and improved confinement for a variety of divertor configurations. TEXTOR Radiation Improved (RI) -mode experiments suggest the important effect of recycling conditions on confinement. Similarly, control of impurity recycling in the MkII-GB configuration is found to be important in obtaining improved confinement in JET. Argon concentrations are measured in the JET sub-divertor with a Penning gauge coupled to a photo-multiplier system. Spectrometer and bolometer measurements in the plasma edge and strike point region are used as constraints on core and divertor modeling to advance our picture of the argon recycling pathway under strong and weak recycling conditions, and for strong and weak shaping configurations.

[GP1.051] Influence of isotopic composition of fuelling gas on the performance of plasmas with a radiating mantle in TEXTOR-94

The RI-Mode on TEXTOR-94 is a well-established and robust plasma regime, combining simultaneously high confinement (with a quality between ELMy and ELM-free H-Mode confinement), high density (close to or even above the Greenwald limit) and high radiation (up to P_rad/P_tot=95%) in a mantle around the plasma. Most RI-Mode experiments have been performed in D plasmas with D fuelling and a strong correlation is found between the edge neutral pressure and the degradation in confinement with respect to the RI-Mode scaling (\tau_RI=n*P^-2/3). D plasmas fuelled with H, however, show a confinement degradation even when the plasma composition is nearly unchanged and the fuelling rate is much lower compared to their pure D counterparts. In addition with varying H content in the plasma, the confinement degradation observed is given by A_i^0.7, i.e. stronger than the usual A_i^0.5 dependence of the usual scaling expressions.

[GP1.052] Electron Cyclotron Current Drive by the Fundamental X-mode Waves Launched from the Low Field Side in JT-60U

The first measurement of the current profile driven by the fundamental X-mode electron cyclotron (EC) waves launched from the low field side is presented. Such X-mode waves can couple to electrons with large velocity in a high temperature plasma due to the Doppler shift effect, in a properly low density condition. The 110GHz EC waves of about 1MW are launched into the JT-60U tokamak, with changing power fraction of the O/X-mode component. The driven current profile is evaluated, considering transient toroidal electric field profile measured by the motional Stark polarimetry. Preliminary analysis shows that the current drive layer of the X-mode waves is narrower and that the current drive efficiency is larger (by a factor of 1.5) than that of the O-mode waves, where line averaged electron density was n_e = 0.4 \times 10^19 m^-3 and local electron temperature was T_e = 4-5 keV. The X-mode waves could not be absorbed in a higher density plasma of n_e = 1 \times 10^19 m^-3. The results are consistent with those of the ray-tracing and the Fokker-Planck calculations.

[GP1.053] Electron Cyclotron Heating Assisted Start-up Experiment in JT-60U

We have started to study the EC assisted start-up in JT-60U, where the plasma volumes at the breakdown is much smaller than the volume of the vacuum vessel. The breakdown voltage is successfully reduced by EC injection of 160\,\mathrmkW on center resonance from 30\,\mathrmV to 4\,\mathrmV which corresponds to the electric field of 0.26\,\mathrmV/m. This result satisfies the ITER requirement of 0.3\,\mathrmV/m. The study of the parameter scan, such as EC power, prefill pressure and resonance position, were done under the condition of low electric field of 0.26\,\mathrmV/m. The Ip-ramp up rate was not changed by increasing EC power up to \sim1\,\mathrmMW. The Ip ramp-up rate decreases with the prefilling pressure, while the plasma start-up at the low electric field was succeeded in the experimental pressure range of 2.7\times10^-3\,\mathrmPa to 1.7\times10^-2\,\mathrmPa. When the resonance position was shifted 71\,\mathrmcm(\sim0.7a) inward from center, the Ip ramp-up rate clearly decreases. Moreover, the Ip ramp-up was obtained only at the center resonance in the case of 2nd harmonic resonance condition. These results may indicate that the optimize of EC plasma formation is key to obtain the robust Ip ramp-up at low electric field. In conclusion, the experiments describes demonstrate that low voltage startup (E<0.3\,\mathrmV/m) is feasible in large tokamak.

[GP1.054] Full Wave Simulation of Reflectometer Measurements in the Internal Transport Barrier of JT-60U Plasmas

A two dimensional wave propagation code, developed specifically to simulate X-mode correlation reflectometer measurements in the ITB of JT-60U plasmas is described. The code makes use of separate computational methods in the vacuum, underdense and reflection regions of the plasma in order to obtain the high computational efficiency and robust solution necessary for correlation analysis. The simulations are applied to the interpretation of density fluctuation levels and radial correlation lengths measured in the ITB of reverse shear plasmas in JT-60U. Experimentally, a dramatic reduction of the radial correlation length is observed from \approx 20 cm in the core of the plasma before ITB formation to \approx 4mm in the fully developed ITB. Throughout the formation of the ITB the coherent reflection is maintained. Simulation of X-mode reflectometry in these plasmas reveals that the inferred correlation length in the ITB is close to the instrumental resolution of the diagnostic, setting a lower limit on the radial wavenumber and fluctuation level which can be inferred.

[GP1.055] Study of Momentum and Heat Transport Properties in JT-60U

Favorable energy confinement property has been obtained in JT-60U plasmas with ITBs (Internal Transport Barriers). In the ITB region, not only the temperature and the density profiles, but also the plasma rotation profile drastically changes. Especially in the ``box'' type ITB, the toroidal rotation profile of impurity shows ``notched'' feature. It is regarded that the profiles of temperature and density have relation with the profile of plasma rotation and the trial of the pressure gradient control at ITB has been given by the control of plasma rotation in JT-60U. However, the relation between the momentum transport and the heat transport has not been fully clarified in a wide range of plasma parameters. In this study, the momentum transport coefficient of bulk plasma is calculated by the rotation velocity of deuterium, which is estimated by using the measured impurity rotation velocity and the momentum balance equation parallel to the magnetic field. By comparing the momentum transport coefficient with the heat transport coefficient, the relation between the momentum transport and the heat transport is clarified in the L-mode plasmas, the hot ion mode plasmas and the reversed shear plasmas.

[GP1.056] Fast Plasma termination without runaway electron generation in JT-60U Tokamak Using Intense Gas Puffing

When an accident in a tokamak discharge occurs, it is necessary to terminate plasma as quickly as possible. In this fast plasma termination, to avoid the generation of runaway electrons is one of the most important issues. Injecting a mixture of small amounts of argon (\sim 2.2 Pa m^3 /s) and large amounts of hydrogen (\sim 120 Pa m^3 /s) into the JT-60U plasma by gas puffing, it is possible to terminate a plasma in a short time with avoiding the generation of runaway electrons. The energy of the plasma was dissipated mainly via the radiation of argon because the radiation of argon was amplified by a high electron density. The increased electron density by the intense hydrogen gas puffing avoids the generation of runaway electrons during the current quench. On the other hand, using large amounts of argon or small amounts of hydrogen in mixture leads to runaway electron generation.

[GP1.057] Impurity control by boronization and optimization of the wall temperature in JT-60U

In JT-60U, boronization using deuterated decaborane has been applied to suppress oxygen production. Boronization decreased the oxygen content in the core plasma from \sim 3% to \sim 0.5%. The oxygen content of \sim 0.5% was kept by additional boronization about every 100 shots. By lowering the vessel temperature from 540 K to 420 K, chemical sputtering yield decreased by \sim 40% at the carbon divertor plates. For hydrogen plasmas, the carbon content reduced from 3.1% to 1.8% in L-mode discharges with NB heating power of 13 MW ,and from 2.4% to 1.7% in reversed shear discharges. For deuterium plasmas, however, the carbon content did not depend on the vessel temperature. The different dependence of the carbon content on the vessel temperature is considered to be due to isotopic difference between chemical and physical sputtering yield. The effects of boronization and the vessel temperature will be discussed based on correlation between impurity content and impurity influx from the divertor plates and the first walls.

[GP1.058] High density operation with Lower Hybrid waves in FTU tokamak

Since April 2001 the lower hybrid (LH) radiofrequency system in FTU (6 gyrotrons @ f=8 GHz) can deliver to the plasma about 2 MW through two equal launchers with a reflection coefficient = 10%. This value is close to the target value of 2.2 MW (net power density of 6.2 kW/cm2 on the waveguides mouth) which could be reached after further conditioning of the grill and of the transmission lines. In high density plasmas (line density *1*1020 m-3), high magnetic field (BT=7.2 T), with PLH=2 MW we drive about 75% of the total current (Ip=500 kA) and stabilise fully the sawteeth activity. The central electron temperature Te0 increases from 1.6 to 3.3 keV (steady), and the neutron rate by about 10 times. Analysis of these pulses with effective electronic heating will be presented. In post-pellet plasmas ( *6*1020 m-3), good coupling of the LH is achieved with the launcher almost flush to the walls, due to the very dense scrape off-layer. The perturbation here induced by the pellet imposes a delay to the LH of only 20 ms. The exact location of the launcher is critical in these regimes, because the high N|| (parallel index of refraction) requested (N||>2.3) for a good penetration of the waves makes more problematic a good coupling all along the poloidal extension of the grill.

[GP1.059] The Stiffness of Electron Temperature Profile in ECRH current Ramp-up Scenario on FTU

The highly localized ECRH in the current ramp-up phase has produced high electron temperature plasma (15 keV) in the on-axis heating experiments on FTU. Off-axis ECRH has been used to study the electron energy transport in the ramp-up scenario, characterized by a variety of current density radial profiles j(r), due to the details the plasma startup, gas feed and impurity content. Previous analysis have already shown that the electron temperature radial profile T_e(r) is consistent with standard diffusive transport models in the region where j(r) is flat or hollow. The focus has now shifted to the experiments with very peaked profiles j(r) and early onset of sawtooth activity in the pre-ECRH phase. Off-axis heating of these discharges results in peaked T_e(r) inside the deposition layer due to a decrease of the electron thermal diffusivity in the plasma core, as shown by the local energy balance. This experiment can be interpreted as the result of a modification of the electron thermal transport that depends on the proximity to a "critical" temperature gradient, as shown by other recent ECRH experiments both on ASDEX-Upgrade and FTU tokamaks.

[GP1.060] Confinement enhancement and sawteeth control by pellet injection in FTU high field plasmas

Enhanced plasma performance has been obtained on the Frascati Tokamak Upgrade at high field (7.2 - 8 T) and high current (0.8-1.25 MA) by multiple deuterium pellet injection. Improved confinement and enhanced neutron rate were associated with an increase of the sawtooth period occurring when a significant portion of the ablated deuterium was deposited near the inversion radius. Reproducible access to the enhanced regime was achieved by adjusting the delay between the first two pellets, as plasma cooling by the first pellet increased the deposition depth of the second one. The enhanced regime was sustained for several resistive diffusion times by injecting more pellets with optimised timing. Density profile peaking keeps increasing for more than one confinement time after pellet ablation. The net particle flux is well below the neoclassical prediction, so that an anomalous particle pinch is needed to explain the observed profile evolution. Results of energy and particle transport analysis will be presented.

[GP1.061] m=1 mode dynamics in FTU

Stable equilibria with a large, saturated m=1 island are observed on FTU after pellet injection. The island lifetime may exceed the resistive diffusion time The mode structure can be analysed in detail when a bright X-ray emission spot forms at the island o-point. When the island width is within 20% of the q=1 radius, its contours closely resemble the linear eigenfunction of the m=1 mode (although the amplitude is well within the non-linear range), i.e. the o-point is displaced towards the plasma center and is surrounded by nested crescents, with the tips of the outer crescent connected by a long ribbon. Larger islands have a smaller poloidal extent and nearly circular contours. The island survives sawtooth crashes, and its evolution during the sawtooth cycle will be used to analyse the trigger mechanism. Island rotation is very slow just after pellet injection, and then it increases along with the electron diamagnetic velocity. Island braking is observed when coupling with higher m modes occurs. These results will be used to discuss the role of diamagnetic stabilisation in the island saturation amplitude.

[GP1.062] Impurity content of pellet fuelled high performance discharges in FTU

High field pellet fuelled discharges in FTU have achieved good energy and particle confinement properties in a quasi-steady regime. Peak densities close to 1e21 m-3 have been obtained, showing a post pellet decay time of several energy confinement times. The neutron yield, in excess of 1e13 neutrons/s, is consistent with high plasma purity (Zeff<1.3), good e-i coupling and neoclassical ion transport. This paper analyses the role of impurities in the evolution of these discharges. A very clean target plasma is required in order to avoid a negative central balance between ohmic input and radiation losses.Even if the analysis is difficult on FTU due to the contemporary presence of metallic (Mo, Fe, Ni) and light impurities (0, C), two regimes of pellet fuelled plasma are clearly well identified and discussed. In the first case, impurity accumulation has been observed, when the sawtooth activity is suppressed by pellet injection, in which the impurity density profile follows roughly the peaking of the electron density profile. In the second one, the accumulation is avoided if a slow sawtooth activity persists during the good confinement phase.

[GP1.063] Disruptions generated runaways in the FTU high field tokamak

Disruptions in FTU are usually accompanied by the generation of a strong pulse of photoneutrons (Y_N 10^12n/s), resulting from photonuclear reactions induced by the bremsstrahlung radiation emitted when runaway electrons (REs) strike the plasma facing components. Measurements of Y_N during major disruptions on TS [1] showed variations of three orders of magnitude when the toroidal field B_t increases from 1.8T to 3.9T. Similar results were found on JT-60 [2], where no REs are produced for low B_t (<2.2T) and a large Y_N was measured for higher fields (up to 4T). The range of B_t available in FTU (4T-8T) allows to extend such analysis so that useful predictions can be obtained for operation in next-step high field tokamaks (IGNITOR, ITER). The dependence of Y_N on B_t is investigated in several FTU disruptions. Y_N increases with B_t for B_t=4T-6T, while no variation is found for B_t=6T-8T: the role played by n_e and I_p on such trend is discussed. [1]P.Joyer,G.Martin,Contr.Fusion and Plasma Heating,Proc.17^thEPS Conf.Amsterdam(1990) [2]R.Yoshino et al.,Nucl.Fus.39 151 (1999)

[GP1.064] Field Reversed Configuration and Spheromaks

[GP1.065] Interaction of Accelerated SCT with an Axial Vacuum Magnetic Field in the Coaxial Acceleration Section

The accelerated spheromak-like compact toroid (SCT) has the potential to centrally fuel a magnetically confined plasma. The zeroth order penetration criterion for the SCT into a magnetized plasma is the balance between its kinetic energy density and the field energy density. On the acceleration time scale, the SCT behaves as a super-conducting plasmoid. Its interaction with surrounding vacuum magnetic fields is important in determining its final parameters as it enters the target plasma. Most magnetic confinement devices have a peripheral vacuum field region where the SCT accelerator resides. We have studied the interaction of an SCT with a transverse magnetic field relative to its direction of motion . In order to compare results with our 2-D MHD simulation, we are studying the interaction of the SCT with an axial magnetic field generated in the coaxial acceleration section. With the externally wound axial coils, both solenoidal and cusp field pattern can be produced. The interaction of the SCT with the field is investigated using surface magnetic probes, 2-D optical imaging, and He-Ne interferometer in the acceleration region. The final SCT parameters after the interaction are determined using time of flight measurements, Langmuir probes and gridded energy analyzer. Work supported by DOE grant DE-FG03-99ER54558

[GP1.066] Electron Energy Distribution Measurements in a Compact Toroid Plasma Before and After Collision with a Transverse Vacuum Magnetic Field

Previous studies on CTIX of the thermalization of high velocity SCT during collision with a transverse magnetic field using a gridded electrostatic energy analyzer probe had resulted in the observation of substantial average electron energy of up to 85 eV \pm 4.25. The electron energy distribution function was obtained on three data sets each with 200 shots. Before the interaction, it is estimated from its L/R time that the accelerated SCT has an electron temperature of less than 10 eV. The acceleration process imparts equal velocity to the ions and electrons; thus producing a high ion kinetic energy of 200 eV while the lower mass electrons obtain a drift kinetic energy of .1eV. We are investigating the SCT electron energy distribution before and after it interacts with the target magnetic field using a pair of electrostatic energy analyzers. The pre-collision electron energy distribution should provide needed information on the thermalization energy balance.

[GP1.067] Optimization of coaxial plasma source for spheromak magnetic helicity injection

Utilization of dc coaxial plasma gun – Marshall gun – to inject magnetic helicity is a standard method to create and sustain a spheromak. Magnetic helicity injection rate is proportional to the gun voltage and gun bias flux. An ideal MHD flow model was developed and successfully applied to the Los Alamos CTX experiment in the early 1990s. [C. W. Barnes, T. R. Jarboe, G. J. Marklin, S. O. Knox and I. Henins, Phys. Fluids B 2, 1871 (1990)] One of the fundamental theoretical predictions is that the gun voltage V_g is proportional to the gun current to the second power, V_g \propto I_g^2, when the mass flow rate is proportional to I_g. For the Sustained Spheromak Physics Experiment (SSPX), however, although the gun current is about the same as CTX, the gun voltage is lower by about a factor of 5. Correspondingly, the maximum observed magnetic field is limited to about 3 kG in the flux conserver region. Certainly there are geometrical differences between the two experiments, however, it is not clear that the geometrical differences are solely responsible for the low gun voltages observed in SSPX. It was proposed recently by Moses, Gerwin and Schoenberg (MGS) that gun voltage is related to electron temperature. We revisit the coaxial plasma source physics in order to explain the gun voltages observed in the SSPX. Comparison will be made to the predictions by the ideal MHD model and the MGS model. The possibility of raising gun voltage for SSPX helicity injection will be discussed.

[GP1.068] Experimental Study of Magnetic Helicity Injection Using a Planar Coaxial Gun

Magnetic helicity injection is invoked as a mechanism for tokamak current drive as well as spheromak sustainment; it is also increasingly recognized as a useful concept in solar, space, and astrophysical plasmas. However, the details of helicity injection are not well understood. We have designed and constructed a simple experiment in order to study helicity injection as a basic plasma physics process, using multi-frame digital photography, electrostatic and magnetic probes, soft X-ray detectors, and both passive and active (LIF) spectroscopy to diagnose the merging of discrete twisted flux tubes into an axisymmetric plasma configuration. Discharge parameters can be adjusted so that different plasma configurations are formed, including (i)~attached plasmas with a stable central column, (ii)~marginally detached plasmas with a kinked central column, and (iii)~detached spheromak plasmas. Configuration (i) is being considered as an analog to astrophysical jets associated with accretion disks. Configuration (ii), likely a current-driven kink instability, offers an opportunity to study a beautiful nonlinear instability in detail and the possible role it plays in energy and helicity flow. And finally, configuration (iii) will be studied in order to answer questions both about gun efficiency and helicity injection micro-physics, e.g.\ how modes overlap so that energy and helicity can penetrate from open field lines into the time-averaged closed field configuration of the spheromak.

[GP1.069] Laser induced fluorescence measurements of spheromak plasmas

A laser induced fluorescence (LIF) diagnostic is being developed for use with plasmas generated by a planar spheromak gun. LIF line broadening corresponding to T_i=19.9 eV Doppler broadening has been demonstrated in a prior experiment in an argon plasma generated by a coaxial spheromak gun. Light from an Nd:YAG pumped, tunable dye laser is directed through an optical fiber to probe the plasma. The laser pulse width is 15 ns, and the argon discharge duration is 20-30 \mus. Synchronization between the pulsed laser (15 Hz repetition rate) and the single-shot plasma device is achieved with CAMAC time delay modules. The fluorescent signal light is measured by a fast photomultiplier after passing through a narrow bandwidth (10 Åfilter centered on the emission wavelength. A critical requirement in any LIF experiment is reproducibility of the plasma. We present images from a gated, intensified CCD camera demonstrating that we have achieved good reproducibility in argon plasmas produced by the planar gun configuration. We expect to present preliminary LIF measurements from these plasmas.

[GP1.070] Imaging of spheromak formation at SSPX

An optical system to provide high-resolution images of spheromak formation in SSPX at LLNL is being designed at Caltech. The design includes a special re-entry port that fits through a 5 cm slot in the side of the flux conserver. The re-entry port will be equipped with a custom-built shutter to prevent coating of the viewport during gettering. The optics, which include a wide angle lens, will be mounted on a removable platform for serviceability and to prevent damage during chamber baking. The image will be relayed to a 1280 x 1024 pixel, cooled, intensified CCD camera with a fast shutter (3 ns) and dual exposure capability (500 ns between frames, minimum). Since the magnetic environment is ~10^-2 Tesla, magnetic shielding is required for the camera electronics. The design may also use a port located at the bottom of the flux conserver to obtain axial images of the discharge, relaying the image by means of a fiber optic bundle.

[GP1.071] Energy balance for sustained spheromak plasmas in SSPX

Formation of self-organized spheromak plasmas requires, at a minimum, that the input power exceed the loss power in order to increase magnetic field strength. Other factors, such as injector geometry or low-order MHD modes, are also thought to affect the formation process. In SSPX we measure both the input power at the coaxial source, and the loss power to the flux conserver (radiation and plasma conduction) to obtain the global power balance which we can relate to the field buildup. The radiation loss is determined by wide field-of-view bolometers, both time-integrated and time-resolved. Radiation losses are dominated by low-Z impurities and are typically less than 20energy input for clean discharges. We use edge magnetic probe measurements as input to the CORSICA code to determine the total stored magnetic energy and ohmic heating power, which then allows us to compute the energy confinement time from density and temperature profiles obtained by Thomson scattering. This work was performed under the auspices of US DOE by the University of California Lawrence Livermore National Laboratory under Contract No. W-7405-ENG-48.

[GP1.072] Particle Balance and Density Control on SSPX

As in other magnetic fusion devices, particle control in the spheromak is important for achieving best operation with good confinement and low radiation loss. In the Sustained Spheromak Physics eXperiment (SSPX), extensive wall conditioning (helium discharges, DC glow discharge cleaning, high temperature bakeout, and titanium gettering) has allowed us to operate at low density (j/n>1E-14 A-m), with most the plasma fueling provided by plasma flow from the coaxial helicity injector. Consequently, there is a strong increase in plasma density with injector current; the external gas fueling has only a weak effect on plasma density. Thomson scattering profile measurements show a direct correlation between plasma density and temperature (nT~constant) on all flux surfaces. Results from a new double-pulse gas injection system will be discussed. This work was performed under the auspices of US DOE by the University of California Lawrence Livermore National Laboratory under Contract No. W-7405-ENG-48.

[GP1.073] Spheromak Equilibria and Buildup – Hyper-resistive Modeling and Comparison with SSPX

Current and flux amplification in a coaxial helicity-injected spheromak are determined by a balance between transport of current (measured by j/B) from the plasma edge to the plasma core and losses to resistivity inside the separatrix. We model current drive in the approximations of nearly-closed flux surfaces and hyper-resistive (HR) transport in Ohm’s law using several approximate and/or ad hoc HR coefficients. Current on open fieldlines and transfer of helicity due to the n=1 kinking of the current column are parameterized. Flux and current amplification are limited if the boundary between edge plasma and closed surfaces is fixed, so the separatrix is determined by the free-boundary model in the Corsica code. Results are compared with SSPX results to evaluate the reality of the model and to estimate coefficients. The model is used for a preliminary optimization of the coaxial electrode geometry for possible upgrades.

[GP1.074] Toroidal Field Profile Measurements of SSPX Spheromaks Using the Transient Internal Probe

The Sustained Spheromak Physics Experiment has been producing temperatures in excess of 100 eV which often have a peaked pressure profile. This occurs while the coaxial gun continues to feed current at the edge to hold the field roughly constant or in a slow, controlled decay. The Transient Internal Probe (TIP) diagnostic is now installed on SSPX and is being used to make field profile measurements during this hot driven phase. The diagnostic consists of a cylindrical verdet glass that is launched through SSPX at over 1.5 km/s. While in transit, it is illuminated from the front by an argon laser. After passing through the probe the light is retro-reflected to an ellipsometer where it is analyzed for polarization rotation due to the magnetic field at the probe. As of this writing, we are testing the diagnostic with plasma to adjust alignment, signal levels, and system timing; internal field profile measurements are expected later this summer, with the data being incorporated into MHD reconstruction of the current profile to help determine the beta and stability of the spheromak plasma.

^a University of Washington, Seattle WA, 98195

This work was performed under the auspices of US DOE by the University of California Lawrence Livermore National Laboratory under Contract No. W-7405-ENG-48.

[GP1.075] FLIRT: A magnetic field topology diagnostic for self-organized magnetically confined plasmas.

We are developing a new technique for measuring the magnetic field-line topology in magnetically confined plasmas. The basic idea behind FLIRT (Field LIne TRacing Diagnostic) is to use a high-power short-pulse laser to launch a burst of energetic (~100keV) electrons from a target passing through the plasma of interest. These electrons then generally follow field lines until they strike a solid surface, where a burst of x-rays is produced. Field line connection length can be determined from the time delay between the laser pulse and the burst of x-rays. The topology of the field lines can be inferred by measuring the connection length as a function of initial target location inside the plasma. The spatial distribution of the x-ray production may provide information on magnetic field fluctuations and stochasticity. We will measure electron yield with a high power laser striking suitable targets. We will test electron propagation in the SSPX spheromak plasma edge by injecting electrons with an electron gun. This work was performed under the auspices of the USDOE by UCLLNL under contract No. W-7405-Eng-48

[GP1.076] Coupling of helicity and power for spheromak buildup and sustainment in SSPX

Coaxial gun-injected power provides buildup and sustainment in SSPX. Current flowing along the open field lines in the spheromak edge dissipates power in electrode sheaths, the resistive edge, and coupling to plasma turbulence. In this paper we compare both power and helicity balance models to calculate plasma buildup. Because both power and helicity are rapidly dissipated in the sheaths and edge resistance, the power and helicity input providing for buildup is modeled from the observed increase in gun impedance above a current threshold. Measurements of the gun voltage and current provide information on the gun impedance. From Thomson scattering measurements of the electron temperature and the CORSICA mhd equilibrium, estimates can be made for the edge dissipation and \etaJ^2 resistive losses within the spheromak core. This work was performed under the auspices of US DOE by the University of California Lawrence Livermore National Laboratory under Contract No. W-7405-ENG-48.

[GP1.077] Formation of spheromaks in SSPX and with the NIMROD code

We compare the results from the Sustained Spheromak Physics eXperiment, SSPX, with NIMROD simulations in order to better understand formation and to improve efficiency in the experiment. SSPX routinely forms spheromaks with n=1 present during formation, and n=2 during decay. NIMROD simulations, with SSPX-similar geometry, reproduce this mode-behavior and suggest that the field strength of the spheromak will grow until the n=1 mode reaches saturation. Such a saturated state is not anticipated by simple helicity considerations which, instead, point to the continued rise in field-strength of the spheromak, limited ultimately by dissipation. We explore the mechanisms that lead to the saturation, and closely compare the SSPX/NIMROD mode behavior to determine discrepancies/similarities that could lead to more efficient experimental operating regimes. This work was performed under the auspices of the US DOE by the University of California under Contract No. W-7405-ENG-48.

[GP1.078] The dynamics of fast electrons in a spheromak

We consider two related problems: 1) Generation of fast runaway electrons (which are sometimes observed in spheromaks, [1]) and, 2) The dynamics of sub-relativistic particles injected to the spheromak in order to get an information on the structure of the magnetic field [2]. In the first problem we identify two factors whose simultaneous presence is required for the appearance of relativistic electrons: a reasonable quality of the confining magnetic field, and the presence of a sufficient loop voltage. We quantify constraints imposed by these factors and relate them to current operation regimes of the SSPX spheromak at Livermore. In the second problem, we consider effects of the magnetic mirroring and finite electron gyroradius on the temporal characteristics of the flux of fast electrons returning to the injection point. This work was performed under the auspices of the U.S. DOE by University of California Lawrence Livermore National Laboratory under contract W-7405-Eng-48. [1] R.E. Chrien et al. Nucl. Fus. v.31, p.1390 (1991). [2] H. Mc Lean et al. FLIRT: A magnetic field topology diagnostic for self-organized magnetically confined plasmas. This Conference.

[GP1.079] Calculation of Free Boundary SSX Doublet Equilibria Using the FEM

A finite element method (FEM) is presented for computation of the free boundary, axisymmetric equilibrium of a compact torus (CT) for use in guiding modifications to the Swarthmore Spheromak Experiment (SSX). This experiment will investigate creation and stability of a doublet shaped, field reversed configuration (FRC) formed from the merger of two counter-helicity spheromaks. The Grad-Shafranov equation is solved within a closed, cylindrical flux conserving boundary with a boundary shaping flux defined by a set of external coils. The free plasma boundary (separatrix) makes the problem non-linear. Adaptive mesh techniques and a non-linear Newton iteration method are used to solve the partial differential equations. The algorithm is validated against several analytic models. Several doublet CT configurations typical of those expected in SSX are presented.

[GP1.080] Experimental studies of FRC stabilization through controlled spheromak merging in SSX

An experimental program at the Swarthmore Spheromak Experiment (SSX) in collaboration with GA is underway to examine the mechanisms and criteria for FRC stability at large s. Although such FRCs are expected to be tilt unstable, a recent hybrid calculation, which traces FRC evolution beginning with a cold, unreversed plasma, predicts stability arising from the spontaneous generation of weak, oppositely oriented bundles of internal toroidal field, or poloidal ion flow vortices, or both. This prediction will be tested in the upcoming program at SSX. FRCs will be produced through the controlled merging of two counter helicity spheromaks in a moderate aspect ratio flux conserver. A midplane reconnection control coil (RCC) will limit the merging and permit study of the FRC stability on the amount of annihilated toroidal flux. A set of 600 fast (800 ns) magnetic probes will measure the internal magnetic structure, and several Mach probes will determine ion flow. A single chord He-Ne interferometer and Langmuir probe provide density and temperature diagnostics. A 2D MHD simulation for the case of full merging is already complete, and Grad-Shafranov calculations of the FRC equilibrium have been implemented for RCC design.

[GP1.081] Basic Scaling of RMF Current Drive in Flux Confined FRCs

Previous calculations of RMF current drive in FRCs were done without accounting for MHD pressure balance, and showed a basic non-linear scaling with little current drive until the magnetization parameter \gamma = ømega_ce/\nu_ei exceeded the penetration parameter \lambda = r_s/\delta where \delta = (2\eta/\mu_oømega)^1/2 is the classical penetration depth. FRCs confined in flux conservers experience similar scaling and, if the initial magnetization is high enough, will expand radially, compressing the external field and raising the density and collisionality, until \gamma/\lambda = 1/2^1/2. This scaling predicts an external field B_e proportional to B_ømega/\eta^1/2 that increases linearly with the RMF strength and inversely with the square root of the cross field resistivity. The density is predicted to scale as B_ømega/(ømega\eta)^1/2 with the added frequency effect due to the ability to carry the same reversal current at lower density for higher RMF frequencies. This scaling is seen experimentally in experiments on TCS at two RMF frequencies, and allows for a rough determination of the effective cross field resistivity.

[GP1.082] Steady-state FRC Maintenance using RMF

For the first time, steady state, flux confined FRCs have been produced at TCS using non-inductive RMF current drive. Up to 60 kA toroidal current has been generated with an RMF vacuum field strength of about 60 G. These experiments differ from steady-state rotamaks in that the FRCs are similar to those formed in theta-pinch devices, that is elongated and confined inside a flux conserver. Experimental results from initial high RMF frequency ( ømega = 10^6 rad/s) experiments have relatively low ratio of equilibrium line current, I'_rev = 2 B_e/\mu_0 to possible synchronous current, I'_sync = 0.5 n_eeømegar_s^2, and the current is primarily edge driven, resulting in relatively low current drive efficiency and unsteady behavior. When the RMF frequency was lowered to ømega = 0.5x10^6 rad/s, the parameter \zeta = I'_rev /I'_sync increased, allowing higher density (up to 3x10^19 m^-3) FRCs to be produced. The plasma temperature cannot be increased much above 50 eV in these purely RMF formed FRCs, and the cross-field resistivity is high. Future experiments will start with hot, relatively low resistivity, theta pinch formed FRCs to explore RMF current drive in a more interesting regime.

[GP1.083] Power measurements in the TCS-RMF experiment

A large tube based power supply was built by LANL to power a Rotating Magnetic Field (RMF) for the TCS FRC sustainment experiment at the U. of Washington. Two antenna sets, of about 1.8 \muH each are each connected in parallel with capacitors to form parallel tuned circuits. The antennas can be run at up to ±10kV, yielding up to ±11kA at a frequency of 80kHz. Two antennas, phased 90 deg apart, produce an RMF with a field strength up to 70G over a volume 1.5m long and 1m in diameter. The power supply consists of two legs with (6) 250A, 30kV triodes each, and is capable of supplying over 5MW of power for several msec. The power absorbed by the FRC is reflected by the phase differences between the antenna voltages and currents. Since the FRC is maintained in quasi steady-state, energy confinement times can be determined. A simple 1-D model, assuming radiative (measured with a silicon bolometer), charge exchange, and convective losses has been used to estimate particle and energy confinement times, and hence ionization rates and neutral particle fractions or effective recycling rates. Collisions with neutrals are one mechanism proposed to prevent ion spin-up by electron friction. The neutral particle fractions necessary for power balance are compared with those calculated to be necessary to account for low measured values of ion spin-up.

[GP1.084] A Detailed Comparison Between Probe Measurements and Numerical Simulation in RMF Driven FRCs.

A detailed set of magnetic and Langmuire probe measurements have been made in both the STX^1 and TCS^2 devices at the University of Washington. These probes provide a direct measure of the magnetic field, density and electron temperature as a function of radius at the axial mid-plane. These measurements are compared with the predictions of a two-dimensional (r-\theta) numerical model^3, which was developed to study RMF current drive in an FRC. Two sorts of comparison are made. In the first we compare the phase and magnitude of the B_\theta component as a function of radius with that of the code, to infer the 2-D structure of the RMF as it penetrates. We look for evidence of “magnetic structures” that rotate relative to the RMF, that were predicted by the code for some conditions. In the second, we compare equilibrium profiles to estimate the plasma resistivity, and gain information about the physical processes that are required to sustain equilibrium. ^1J.T. Slough and K. E. Miller, Phys. Rev. Lett. 85, 1444 (2000) ^2H.Y. Guo, et.al., Submitted to Phys. Plasmas. ^3Richard D. Milroy, Phys. Plasmas 7, 4135 (2000)

[GP1.085] Ion Spin-up in an FRC with RMF Current Drive and its Effect on Stability

The TCS device has been equipped with an intensified CCD spectrometer to determine ion toroidal rotation via the Doppler effect. C-III emission at 229.7 nm is viewed at six chords, allowing determination of a rotation profile. In the presence of electrons synchronous with the RMF, the ions experience a frictional drag force and will spin up in a short time \tau=(m_i/m_e)\nu _ei^-1, or ~100 \mus for TCS. This is not observed in these low-density experiments; rather, the ions spin up to a small fraction of the electron velocity (3% at ømega _RMF=1.0 MHz, 15% at ømega _RMF=0.5 MHz) and the current is maintained in steady state, even with ømega _ci/ømega _RMF=0.65. The primary braking mechanism may be charge-exchange with a background of cold, non-rotating neutrals. The inferred charge-exchange rate requires a neutral density equal or greater to the plasma density, which is higher than suggested by power balance. Even at these low spin-up fractions, the ions are rotating faster than their thermal velocity, causing a rotational instability. Tomography shows that a non-destructive rotating n=1 distortion develops at low ion velocities, while at higher velocities the n=2 mode is dominant.

[GP1.086] FRC Translation and Capture Measurements on TCS

Field Reversed Configurations (FRC) have been formed, accelerated to high velocity, and captured in the TCS device. This is necessary in order to achieve a hot (\sim500 eV), low density (\sim10^20 m^-3) FRC for Rotating Magnetic Field (RMF) current drive studies. Initial results show an approximate order of magnitude increase in T_i + T_e from thermalization of directed energy after reflection off the downstream confinement mirror. Low densities are desirable for RMF current drive, and low density FRC lifetimes are better than predictions based on high density scaling. Detailed studies of the FRC poloidal and toroidal fields, using a set of internal B probes, will be used to determine the origin of a flux enhancement observed after the first mirror reflection before any RMF is applied. FRC lifetimes have been degraded by impurities picked up during the reflection. A Tantalum shield has been installed on the wall of the conical mirror section, which should decrease impurity pickup. Hot (\sim500 eV), low density (\sim10^20 m^-3) FRCs are nearly ideal for RMF current drive, and these conditions are readily obtainable from translated FRCs.

[GP1.087] High Field RMF FRC Experiments on STX-HF

As part of the effort of the TCS program to explore the physics of RMF generated FRCs, a start-up experiment has been constructed to operate at much higher RMF and axial magnetic fields. It is hoped that with the RMF alone at higher power density , it will be possible to form plasmas with both high density, and high electron and ion temperatures. To this end the STX experiment with its 2 MW RMF supply was reconfigured to drive a smaller plasma volume (r_p reduced from 18 to 6 cm). In addition an axially remote plasma source was used to supply plasma in a manner that limits neutral contamination and RMF field penetration issues. With axial fields in the kG range, plasma temperatures and densities in a more fusion relevant regime can be studied. Initial results at low RMF power (60 kW) have produced Te = Ti for the first time (T_tot ~ 25 eV). As a result, both ions and electrons are magnetized (ømega\tau >> 1). Unlike previous experiments, the plasma is formed with a fully penetrated RMF field, producing a rigid rotor current profile. The plasma resistivity at low RMF power is consistent with classical (~ 60 \muohm-m). The particle confinement is found to be considerably enhanced by the RMF, with a confinement time several times classical. Results at higher power will be presented as well.

[GP1.088] RMF Driven FRC Fusion Reactor

There have been several experiments in the past that have formed and sustained a field reversed configuration (FRC) with a rotating magnetic field (RMF). But it is only recently that the confinement and scaling properties of these plasmas have been assessed. While this work is still ongoing, recent experiments where there has been adequate separation of the FRC from wall contact, the confinement and stability of the plasmas look promising. These RMF driven FRCs display a resistivity comparable to that predicted by past FRC experiments. Enhanced particle confinement with much better confinement than previous FRC scaling has also been observed. Even with a scaling no better than that observed in past FRC experiments, a fusion system based solely on RMF driven currents is envisioned to be feasible. The fusion system would be compact (r_p < 0.1 m), and would operate with a significant RMF field (~1 T). The power flow into the FRC during formation would reach several hundred MW and ohmically heat the FRC to fusion conditions in less than 1 ms. Although the RMF field is large, the necessary oscillating currents are well within the capability of current solid-state switching technology. Details of the final reactor size and power depend on the confinement scaling observed as RMF plasmas reach more fusion relevant conditions. Various scenarios based on different scalings will also presented

[GP1.089] Stability of flowing equilibria by Beltrami function expansion

An elegant analysis of the stability of force-free states was presented by Finn et al.^1 for a cylindrical geometry by expressing equilibria in terms of Beltrami fields. These vector fields Y are “eigenfunctions of the curl operator”, i.e. solutions to curlY = \Lambda Y (the \Lambda are eigenvalues) subject to the boundary condition Y_n = 0 at the surface of the cylinder. The same approach can be used in a flowing, two-fluid plasma. Here the extrema of the magnetofluid energy W_MF (flow + magnetic field energy) subject to constraints on K_i and K_e (helicities associated with ion and electron species) can also be expressed using single Beltrami fields. In these stationary-energy states the functional form W_MF(\Lambda, K_i, K_e) can be found. This allows the associated equilibria to be classified on a 2D helicity map of K_i vs. K_e. Two classes of equilibria appear: force-free (spheromak-like) and non-force-free (FRC-like). Familiar limiting cases, including Taylor states and Beltrami states, occupy certain lines on the helicity map. The stability of a stationary-energy state depends on its rank in magnetofluid energy: the state with lowest W_MF (given K_i, K_e) is assuredly stable. This is usually but not always the state with the lowest \Lambda. As found by Finn^1 the lowest energy state is sometimes the axisymmetric “compact toroid”-like state, and sometimes a “tilted” state, depending on the cylinder’s length-to-diameter ratio.

*Supported by US DOE Grant DE-FG03-99ER54480.

^1J.M. Finn et al., Phys. Fluids 24, 1336 (1981).

[GP1.090] Plasma Flow and Equilibrium Considerations in the STX Experiment

The STX experiment was operated during the construction phase of TCS, primarily to investigate the ability of the Rotating Magnetic Field (RMF) to directly form an FRC, without the usual theta pinch technology. STX utilized a 25G RMF at 350kHz to form 2m long by 0.2m radius FRCs. Plasmas were typically fully ionized deuterium with a temperature of 60 eV and a peak density of 5x10^18m^-3. Axial confining fields of 100G maintained a true vacuum boundary around the plasma and allowed for the study of FRC RMF equilibrium interactions. Key findings are that the RMF drove strong radial and axial flows, produced radial profiles markedly different from those of theta pinch formed FRCs, and resulted in enhanced particle and energy confinement. Although the FRCs were usually not sustained, they evolved into an interesting mirror like configuration that also exhibited enhanced particle and energy confinement. Issues discussed include the importance of the RMF driving an azimuthal current distribution consistent with that of the FRC, possible benefits of varying the average beta condition, and potential RMF antenna length limits set by the tendency of driven axial flows to screen the RMF from the plasma.

[GP1.091] FRX-L: A High-density Field Reversed Configuration

We present the progress on the high-density Field Reversed Configuration (FRC) experiment (FRX-L) presently being studied at LANL. FRX-L is a magnetized-target injector for MTF experiments; the experimental setup not only allows us to diagnose the plasma during formation and in situ but also to translate it into a hollow aluminum liner (flux conserver) where compression would take place. Subsequent heating of the FRC plasma by compressing it inside this imploding flux conserver should allow access to fusion conditions. The actual compression of the plasma to fusion conditions will be performed at the Shiva-Star facility at AFRL. Diagnosing the FRC plasma during formation, translation, and compression is challenging due to the short timescales, high energy densities, high magnetic fields, and restricted access. Our goal is an FRC with density n~10^17 cm^-3, T_i=T_e~100-300 eV, external B~5 T, and a lifetime of 10-20\mus. We will present results to date on the formation of the high-density FRC, including the Pre-Ionization (PI) phase. From our previous experience, the PI process is crucial for good FRC formation, and not well understood. The hydrogen gas fill is ionized by the ringing theta coil discharge (\theta-PI) technique, which applies a rapidly oscillating axial magnetic field superimposed upon the slower-timescale magnetic bias field, of comparable magnitude. The PI process occurs just prior to applying a much (10 X) higher reverse field that causes the frozen-in field lines to radially contract and form the closed field lines for the FRC.

[GP1.092] Self-Generation Process of Hollow Current Profile in a Field-Reversed Configuration

The purpose of this study is to investigate the profile relaxation process of the field-reversed configuration plasmas by means of the three dimensional electromagnetic particle simulation. The particle simulation starts from an MHD equilibrium with a peaked current profile. For the full kinetic case of \bars=1, the separatrix radius r_sp and the field-null radius R oscillate and the separatrix beta value \beta_sp jumps to a larger value in an initial short period. For the moderate kinetic case of \bars=3, on the other hand, r_sp and R do not oscillate and \beta_sp keeps the initial value. The radial electric field is generated in the narrow periphery region near the magnetic separatrix by the ion FLR effect, and the size of this region is shorter than the ion meandering orbit size for the full kinetic case. Because the generated E \times B drift has the same sign as the electron diamagnetic drift, the electron current density increases in the periphery and then its profile becomes a hollow one. The electric field, on the other hand, acts on ions less effectively since the ion meandering orbit size is larger than the size of a strong electron field region. That is, the modification of ion current profile becomes relatively smaller. In this way, an initial MHD equilibrium with a peaked current profile relaxes to a kinetic equilibrium with a hollow current profile through the ion FLR effect for the full kinetic case.

[GP1.093] Experimental Program for Plasma Study at Prairie View Aamp;M University.

The initial stage of establishing the plasma experiment is to be accomplished by the end of 2001. The compact device Rotamak, two 400 kW RF generators and accompanying equipment had been delivered from Flinders University of Australia and are being assembled at PVAMU. The plasma in the Rotamak is created in a 50-liter glass chamber of spherical torus shape. The unique feature of the Rotamak is that the plasma current is driven by the application of a rotating magnetic field, created by two 90^o-dephased RF generators. The Rotamak is capable of operating in field reversed configuration regime. The initial goal for experiments is to achieve T_e \sim 40 eV, n_e \sim 10^13 cm^-3, I_plasma \sim 6 kA, with a pulse duration \sim 50 ms. The next stage will include boronization of the vacuum chamber, experiments with an imposed radial electric field to study the rotation of plasma, and improvement of the diagnostic tools.

[GP1.094] The origin of an n=1 mode motion on field-reversed configuration plasmas

An n=1 mode global motion on FRC plasma can be easily controlled by a multi pole field, though the deviation of the plasma center from the coil axis reaches about 20-40% of a separatrix radius on the onset of the n=2 rotational mode. The critical field strength is about 15% of the confinement one and the n=2 mode can also be stabilized by the same strength. The origin of the motion hasn't yet been clarified. An azimuthal asymmetry of a radial pinch may be triggered by an asymmetrical magnetic field reconnection at the formation phase and transferred to the n=1 motion by the axial contraction. By the enhancement of the asymmetry, the motion occurs earlier and the amplitude of the deviation increases. To avoid the asymmetry, the non-tearing formation method is introduced. Preliminary experimental data indicate that the deviation of the n=1 motion decreases within a factor of two. We'll measure the behavior of the n=1 motion at the non-tearing formation quantitatively by optical detectors and a magnetic probe array. The behavior will be compared with simulations derived by a 2-D snowplow model.

[GP1.095] A New Magnetically-Insulated, Plasma Anode Ion Diode Geometry for Ion Ring Formation

A first set of experiments has been completed on the Cornell ion ring experiment FIREX, with a novel ion diode. The diode is coaxial with the anode at 5~cm radius and the cathode at 6~cm radius from the axis of a solenoidal magnetic field. The diode accelerates an intense proton beam radially outward. The object is to use the solenoidal field to confine this beam to form a strong ring. The field (6~kG) also provides magnetic insulation of the diode accelerating gap. The anode is composed of axially streaming hydrogen plasma, produced by an inductive, gas-breakdown source inside the anode (high-voltage) feed of the 700~kV FIREX pulser. The cathode is a drifting electron population emitted from the end of the cathode of the bi-conic MITL feed. First results showed that this diode and feed operate with good efficiency and produces 50-100~A/cm^2 proton current over its \sim40~cm axial length (\sim1500~cm^2 area), with excellent uniformity. This represents a possible factor of two increase in the proton population available for ring formation on FIREX.

[GP1.096] Dynamics of strong proton rings in the Cornell Ion Ring Experiment

In the Cornell ion ring experiment FIREX, a proton ring is formed in one ion cyclotron period by cusp injection into a solenoid. The rapid magnetic field change generated by the ring current launches large-amplitude Alfvén waves with amplitudes as large as 25% of the 6~kG field. The lowest frequency and largest amplitude components of the wave spectrum are identified as the lowest-order (nearly radial) magnetoacoustic mode of the magnetized plasma column in the solenoid. In high-density plasmas this is the only significant mode, but injection into lower density plasmas in which the Alfven speed approaches the ring ion velocity also produces non-axisymmetric, higher frequency components, near the ion cyclotron frequency. Due to the wave generation, axial momentum is transferred from the ring to the plasma, bringing the ring to stop inside the solenoid. Diamagnetic loop arrays and magnetically-insulated proton detectors diagnose the wave and ring dynamics, and spectroscopic diagnostics measure energy deposition in the background plasma. We present our developing understanding of the ring-plasma dynamics, including ring stability and particle loss, and details of the magnetic dynamics.

[GP1.097] Spectroscopic Diagnostics of Plasma Response to Ion Ring Dynamics

Two new spectroscopic diagnostic systems have been developed for the Cornell ion ring experiment, FIREX. First, a MCP-intensified, wide-dynamic range CCD is used to record streaked spectral lines from a 1~m spectrometer. With sub-Åresolution and submicrosecond time resolution, plasma ion temperature can be observed by line broadening. The ion rings being formed in FIREX produce large-amplitude Alfvén waves near the ion cyclotron frequency. These waves, carrying hundreds of joules of energy, are rapidly damped in the plasma. Evidence of rapid heating of the plasma ions will be presented. The second diagnostic is a novel large-aperture spatial Fourier spectrometer, also with MCP-CCD output and computerized data analysis, for general survey spectrometry.

[GP1.098] Simulations and Theory of Ion Beam-Alfvén Wave Dynamics in the Cornell Ion Ring Experiment

Sensitivity of Alfvén wave dynamics to changes in the ion beam and background plasma in the FIREX experiment are being studied with analytic theory and in simulations using the 2D code FIRE in both (r,z) and (r,\theta) geometry. Comparisons are made between the experimental \deltaB-probe signals and variations of the magnetic field in the simulations. Comparisons are also made of the experimental beam particle losses (resolved in space and in time) and such losses in the simulations. The results of these studies are being used to elucidate the details of the ring-plasma dynamics, and to infer plasma and ion ring parameters from the observed experimental magnetic dynamics. These quantities can be compared with other diagnostics. Factors affecting the ring stability and confinement are under investigation, and possible methods of control over these factors will be discussed.

[GP1.099] Turbulence and MHD Theory

[GP1.100] Towards a Simple Model Intermittency in Drift Wave-Zonal Flow Turbulence

Since nonlocal interaction in k-space via self-generated zonal flow shearing is the dominant self-regulation mechanism for drift wave turbulence, it follows that the latter MUST BE INTERMITTENT. We describe a simple intermittency model based upon the straightforward analogy between sandpile avalanche dynamics and nonlinear k-space refraction. The existence of a conserved adiabatic invariant (here the drift wave quanta density) and general symmetry principles allow the derivation of a generalized Burgers-type equation (in k-space!!) which describes intermittency in terms of fluctuations of the quanta density about the self-organized spectral profile. Known results from the highly developed theory of Burgerlence may then be directly exploited to characterize the pdf of wave-packet intensity fluctuations, which is both asymmetric and non-Gaussian. A condition for k-space shock formation is derived and used to determine a threshold for intermittent `shearing events'. Numerical studies and extensions to the case of streamer intermittency are in progress. The relation of this work to other reduced models of shear flow dynamics will be discussed.

[GP1.101] Dynamics of Zonal FLow Instability and Saturation in Drift Wave Turbulence

We study generalized Kelvin-Helmholtz (GKH) instability as a saturation mechanism for a collisionless zonal flow in the background of drift waves. By treating drift waves as adiabatically modified by GKH, we investigate the modulation instability of drift waves due to GKH modes as well as the linear inflection-type instability of zonal flow. In the case where zonal flows evolve on the time scale much larger than GKH mode, GKH mode is shown to become destabilized not only by the linear instability of zonal flow but also by coupling to drift waves, with a growth rate which is enhanced over the linear value. Furthermore, the nonlinear (modulational) generation of a zonal flow is estimated to dominate over that of GKH. Our results indicate that GKH may not play an important role in a collisionless saturation of zonal flow, in contrast to [1] and [2]. The effect of temperature fluctuation will be discussed.

[1] B.N. Rogers, W. Dorland, and M. Kotschenreuther, PRL, 85, 5336, (2000).

[2] Y. Idomura, M. Wakatani, and S. Tokuda, PoP, 7, 3551, (2000).

[GP1.102] Analytic Theory of the Momentum Flux PDF in Drift Wave Turbulence

An analytical theory of the tails of the PDF for the local Reynolds stress (R) is given for forced Hasegawa-Mima turbulence. The PDF tail is treated as a transition amplitude from an initial state with no fluid motion to final states with different values of R due to nonlinear coherent structures. With the modelling assumption that the nonlinear structure is a modon in space (an exact solution of a nonlinear Hasegawa-Mima equation), this transition amplitude is determined by an instanton. An instanton is localized in time and can be associated with bursty and intermittent events which are thought to be responsible for PDF tails. The instanton is found via a saddle-point method applied to PDF, represented in terms of a path integral. It implies the PDF tail for R with the specific form \exp[-c R^3/2], which is a stretched, non-Gaussian exponential. The extension of this model will be discussed.

[GP1.103] Bursting in Transport and Turbulence in Drift Wave -- Zonal Flow Systems

The predictions of of the extended predator-prey model of the coupled spectral dynamics of drift wave -- zonal flow turbulence will be presented. The model exhibits three possible types of time-dependent solutions, depending on system parameters, which are: (1) quasi-periodic bursting of the transport and turbulence intensity levels (2) oscillatory relaxation to a stationary state, and, in the collisionless limit (3) an intensity pulse followed by saturation of zonal flow. These solutions are consistent with the time dependent behaviour recently observed in the global gyro-kinetic simulations.

[GP1.104] Magnetic Secondary Instabilities in ETG Turbulence

Research in the last decade has established the importance of the ion temperature gradient (ITG) instability in determining ion particle and energy transport. However, understanding of electron transport has not kept pace. The electron temperature gradient (ETG) mode has been proposed as a primary mechanism for electron transport. The possibilities of magnetic secondary instabilities (zonal "fields" and magnetic streamers) are investigated as potential mechanisms for electron transport regulation and enhancement, respectively. The role of adiabatic ions (versus non-adiabatic electrons in ITG turbulence) is discussed, as are the effects of magnetic Reynolds stresses. Growth and damping of both secondary instabilities are discussed. Finally, the effectiveness of random magnetic shearing as a turbulence regulation mechanism, and of magnetic streamers for enhancing transport, are investigated.

[GP1.105] Suppression of Cross Phase by Flow Shear in Inhomogeneous Magnetic Field

Flux measurements in probe-induced shear layers typically show flux reversal at some point in the shear layer. To understand these observations, we examine suppression of scalar transport due to the reduction of cross phase by strongly sheared flow in a sheared magnetic field. Of interest is the interplay of spatial variations for flow shear and magnetic shear, which go as the first and second powers of radial displacement, respectively. We use previously developed asymptotic methods for evaluating the scalar mixing in a Kelvin neutral layer of an isolated fluctuation of a given helicity [1]. We determine if the propagating evanescent wave structure known to occur in magnetically localized eigenmodes in the presence of flow shear [2] leads to a reversal of flux. We then consider the effect of fluctuations on adjacent rational surfaces to evaluate the total flux and determine when the flux reversal of a quasi-coherent structure can prevail over the mixing of flux contributions from adjacent fluctuations. 1. P.W. Terry, D.E. Newman, and A.S. Ware, submitted. 2. B.A. Carreras, et al., Phys. Fluids B 4, 3115 (1992).

[GP1.106] Finite amplitude dissipation of zonally averaged flow in drift wave turbulence

The importance of the nonlinear excitation of linear eigenmodes of the stable manifold has been recently pointed out in the context of a model for collisionless trapped electron mode turbulence (D. Baver, P. Terry, E. Fernandez, Phys. Lett. A, 267), 188 (2000). In the collisionless regime, the nonlinear excitation of the stable manifold, which is caused by the electron nonlinearity, ultimately alters the energy injection rate into the turbulence. Because of this supercritical instability, the spectrum at saturation differs significantly from spectra obtained with more conventional turbulent calculations. In particular, the saturation spectrum shows a finite amplitude-induced stabilization of k_y=0, zonally averaged modes (P. Terry, D. Baver, E. Fernandez, in preparation).

Here we further investigate the properties of zonally averaged flows as described by a supercritical instability calculation. We first gain insight by numerically studying both the nonlinear growth rate dependence on the saturation energy and the collision frequency, and the energy transfer dynamics. A more detailed understanding of these features is then sought by applying statistical closure theory to the nonlinear evolution equations of all the linear eigenmodes. The latter approach will also clarify the origin of the nonlinear dissipation of zonally averaged flow, as well as the condition for its occurrence.

[GP1.107] Nonlinear instability and overstability in Ion Temperature Gradient driven turbulence

Ion temperature gradient driven turbulence (ITG) is currently believed to be the dominant source of energy transport in large Tokamak experiments. It is currently understood primarily through simulations aimed at determining growth rates and linear mode structures of unstable modes. However, simulations of a collisionless trapped electron mode (CTEM) model have previously revealed nonlinear instability and overstability due to excitation of a damped branch of the dispersion relation by nonlinearities associated with three-wave coupling (1). Whether this mechanism also applies to ITG turbulence is of interest for calculating turbulent transport. Using simulations of a three-field model of ITG turbulence in a homogenous magnetic field, we track the evolution of all linear eigenmodes, including those in the damped and marginally stable branches of the dispersion relation. By comparing energy input rates at finite amplitude to linear growth rates, we determine whether the aforementioned nonlinear corrections to the growth rates are significant and gain insight into the nature of their effects.

Work supported by USDOE.

(1)Baver, et. al., Phys. Letters A 267 (2000) 188

[GP1.108] Drift-time-scale relativistic particle dynamics in perturbed tokamak plasma

Lie-transform methods are used to derive reduced Hamilton equations for relativistic particle motion on drift time-scales, in axisymmetric tokamak plasma with low-frequency electromagnetic fluctuations. The quasistatic background fields are represented by magnetic coordinates which allow for helicity, in contrast to previous work [1]. The action-angle formalism [2] is adopted for all three degrees of freedom, including both trapped and circulating particles. Resonant (e.g. quasilinear) and nonresonant (e.g. ponderomotive) effects of drift-time-scale dynamics are obtained.

[1] R G Littlejohn, PhysScripta T2/1 (1982) 119; B H Fong amp; T S Hahm, PhysPlasmas 6 (1999) 188; A J Brizard, PhysPlasmas 7 (2000) 3238. [2] A N Kaufman, PhysFluids 15 (1972) 1063.

[GP1.109] Stabilization of Sawteeth in Tokamaks with Toroidal Flows

Sheared toroidal flows with a magnitude of the order of the sound speed are shown to stabilize sawteeth in tokamaks. In the absence of flows, tokamak equilibria in which the central safety factor q is less than unity are unstable to resistive tearing modes (resistive internal kink modes) with toroidal mode number n=1. As the ratio \beta of the plasma pressure to the magnetic field pressure increases, the growth rate of the n=1 mode rises because of the increasing pressure gradient. However, the addition of a toroidal flow to the equilibrium has a stabilizing effect. As the magnitude of the toroidal flow approaches the sound speed, the n=1 resistive tearing mode can be completely stabilized by the flow, eliminating sawteeth.

[GP1.110] Solution to resistive inner layer problem with non-uniform density.

The asymptotic matching approach for resistive tearing modes is based on matching of ideal inertia free plasma equations solution between resonance surfaces and inertial resistive equations solution around the rational surfaces. Traditionally[1,2], resistive inner layer model with uniform plasma density is used. An attempt at a non-uniform inertia correction to this model was recently made[3]. We consider a generalization of the resistive non-uniform density inner layer model. A technique to solve a singular system of differential equations[4] is applied to solve the irregular singular problem numerically. We compare our results with published earlier results[2] and find an excellent agreement as well as a wider range of parameters, where the system can be solved using that numerical method. Results are presented from a study of the influence of variable density profile on behavior of the resistive mode, growth rates and matching conditions.

[1]B.Coppi,J.M.Greene, J.L.Johnson, Nucl. Fusion, 6(1966),101; [2]A.H.Glasser, C.C.Jardin, G.Tesauro, Phys. Fluids, 27(1984), 1225; [3]J.M.Greene, R.L Miller, Phys. Plasmas, 2(1995), 1236; [4]S.A.Galkin, A.D.Turnbull, M.S.Chu, J.M.Greene, Phys. Plasmas, 7 (2000),4070;

[GP1.111] Nonlinear Behaviors of Kinetic Internal Kink Modes in the Presence of Density Gradient

To study the sawtooth crash phenomena in tokamaks, nonlinear developments of the m=1(poloidal mode number) and n=1 (toroidal mode number) kinetic internal kink mode is simulated by the gyro-reduced-MHD code. Time evolutions of the electrostatic potential, the longitudinal component of the vector potential, and the electron density are solved numerically. The cylindrical model is used neglecting higher order toroidal effects. Electron inertia, effects of perturbed pressure gradients along the magnetic field, and electron diamagnetic effects are included in the basic equations. The cases with \rho_s >> d_e (\rho_s is the ion Larmor radius estimated by the electron temperature, d_e is the collisionless electron skin depth) are treated. For the weak gradient in the equilibrium density profile, the growth rate is accelerated to a constant value independent of the linear growth rate. Full reconnection is observed and the parity of the electrostatic potential at the newly formed core region after the full reconnection is positive. For the strong gradient in the equilibrium density profile, the deceleration of the growth rate is observed due to the excitation of the Kelvin-Helmholtz type instability.

[GP1.112] Forced Reconnection Rate Corrected by Plasma Inertia in Singular Layer

In plasma confinement, a resonant boundary perturbation causes a forced reconnection in a singular layer located in the vicinity of the resonant surface. The reconnection process in this singular layer is described by an inner-reconnected flux. The initial-value problem of the inner-reconnected flux is calculated by use of the boundary layer theory. The plasma dynamics in the singular layer makes the inner-reconnected flux differ from the pseudo-outer-reconnected flux obtained by the inner limit of the quasi-static outer solution. An integral equation for the inner-reconnected flux is derived and solved numerically. The time derivative of the inner-reconnected flux gives the reconnection rate in the singular layer.

[GP1.113] The analysis of the magnetic island immersed in microscopic turbulence

Various kinds of collapse phenomena of high-temperature plasmas have been observed in toroidal devices. It is necessary to understand these physical mechanism for the achievement of high performance in fusion plasmas. To analyze the collapse phenomena in high-temperature plasmas, a hierarchical model has been derived. Using this model, we analyzed the dynamics of the magnetic island immersed in the microscopic turbulence. The evolution equation of the island width w and rotation frequency ømega are derived in the cold ion limit. The effects of the current flowing in a layer close to the island separatrix are perturbedly treated by the first-order corrections of finite Larmor effects. These corrections are important to evaluate the effects of turbulent diffusions on the island rotation. The magnetic island is driven by the bootstrap current and polarization current in this model. As the results of the analytical calculation, it is found that the rotation of the magnetic island is driven by anomalous resistivity, and suppressed by anomalous ion viscosity and anomalous thermal conductivity. The dynamics of island evolution is also analyzed numerically and will be reported in detail at the meeting.

[GP1.114] The Ideal MHD Continuum in Nonaxisymmetric Toroidal Geometry

\fontßbf=cmssbx10 \def\t#1\hboxßbf #1 \def\v#1\hbox#1

The continuous spectrum of ideal MHD is governed by a fourth-order Hamiltonian system of ordinary differential equations along the equilibrium magnetic field lines. This system can be expressed as d\vu/d\ell = \tA \cdot \vu, where \ell is arc length along the equilibrium magnetic field; \vu is a 4-vector; and \tA, which depends on the real frequency ømega, is a real, 4x4 matrix. Because the system is Hamiltonian, the propagator \tP, defined by \vu(\ell) = \tP(\ell)\cdot\vu(0), is a symplectic matrix satisfying

d\tP/d\ell = \tA\cdot\tP, \quad \tP(0) = \tI. \eqno(1)

\tA is single-valued on each magnetic surface. The frequency ømega is in the continuous spectrum if on a particular magnetic surface there exists a solution \vu which is also single-valued. We investigate this question by numerical integration of Eq. (1), followed by an application of sampling theory to the solution. Accurate numerical integration of Eq. (1) requires an appropriate treatment of the problem of exponentially growing and decaying solutions. We discuss a method for doing so.

[GP1.115] New VACUUM: towards an object oriented version of the code with additional physics capability

The VACUUM Code^a which was initially created to provide the outer boundary conditions and diagnostics to the PEST and NOVA Fourier codes has been substantially modified to be interfaced to a variety of other stability codes, including DCON, the finite element GATO code as well as the nonlinear NIMROD and M3D codes. It now also includes the ability to model the feedback stabilization of external MHD modes in tokamaks so that the effects of a thin resistive shell and the feedback circuitry together with the associated sensor loops and feedback coils are incorporated. To improve the interface to an increasing number of codes, structural changes addressing portability and memory management are under development: A Fortran 90 version of the code using dynamic memory allocation is in progress, and furthermore, VACUUM will be transformed from a standalone code using I/O files to one using a set of library calls where input and output data are communicated through "set" and "get" calls. The benefit of such an application programming interface (API) layout is to allow VACUUM to be embedded in large packages (e.g. TRANSP) , scripting environments (e.g. Python, Matlab, IDL) or wrapped into C++ code to provide object oriented features.

^aM.S. Chance, \textitPhys. Plasmas \textbf4, 2161 (1997).

[GP1.116] Free boundary ballooning mode theory

A free boundary ballooning mode theory is developed. The conventional ballooning mode representation is ineffective in this case, due to failure of the so-called translational invariance at the boundary. By examining the mode symmetry properties near the plasma edge, a modified ballooning mode representation is introduced. Unlike previous theory, where the edge physics plays a role only in the first order, the edge effect is taken into account in the ballooning mode equation of the lowest order in the present theory. The relationship with the peeling mode is also studied. Generalization of this formalism to deal with the double layer mode near the safety factor q minimum is also discussed.

[GP1.117] Two-Dimensional MHD Simulations of Tokamak Plasmas with Poloidal Flow

Recent ideal MHD calculations have shown that poloidal flow in a tokamak can result in a pedestal structure across which the velocity and pressure vary strongly. In a low-\beta tokamak the effective sound speed in the poloidal direction is the sound speed C_s scaled by the ratio of the poloidal to total magnetic field strength, C_s\theta = C_s(B_\theta/B). This is a measure of the time scale necessary for a sound wave to propagate from the outer to inner radial edge of the plasma. The poloidal sound speed goes to zero at the center of the plasma by symmetry and gets quite small near the outer radial edge. Hence even small poloidal rotational velocities can result in a supersonic flow. Owing to the finite inverse aspect ratio, the toroidal geometry of the flow behaves as a de Laval nozzle with a throat at the inner midplane. Under the right conditions, a subsonic flow can become supersonic near the inner midplane and remain supersonic as it completes a poloidal revolution. This in turn results in a shock wave that propagates from the outer to the inner midplane, irreversibly heating the plasma and generating a poloidal shear flow. We present time-dependent numerical simulations detailing the formation and evolution of this flow pattern and a comparison with analytic results. This work was supported by the U.S. Department of Energy under Contract No. DE-FG02-93ER54215.

[GP1.118] Toroidal Flows in Resistive MHD Steady States

The recognition that large-scale vortices are an inevitable part of a resistive, toroidal, MHD steady state has suggested a revision of our picture of magnetically-confined, current-carrying, magnetofluid configurations [1,2]. Not only the "instabilities," but also the "equilibria," need to be calculated using dissipative MHD. This introduces changes that are far from trivial. Previously [1], we had used perturbation methods, assuming small Hartmann number (H<<1) to correct the poloidal dipolar flow pattern that emerges in the very viscous limit. We now have numerical methods that will treat a wide range of Hartmann numbers using stress-free boundary conditions, and can follow the changes in the flow pattern as the Hartmann number is raised. The most surprising result is that the (steady-state, axisymmetric) flow exhibits a continuous transition from almost purely poloidal to dominantly toroidal as the Hartmann number H goes from small to large, compared to unity. [1] L.P.J. Kamp et al, Phys. Fluids 10, 1757 (1998). [2] J.W. Bates et al, Phys. Plasmas 5, 2649 (1998).

[GP1.119] Anomalous braking of plasma toroidal rotation

There is evidence from different tokamaks that the onset of locked modes driven by resonant helical error fields produces a dramatic bulk braking of plasma toroidal velocity that leads to a self-similar quench of the whole velocity profile on short time scales. These observations are in contrast with the expected effect of an electrodynamic torque localized at a rational surface r=rq and associated with the no-slip condition that the variation of rotation viscously damped be uniform for 0 < r < rq.Recent experiments suggest that the rotation damping can be described by some mechanism depending on the square of the poloidal field perturbation . We discuss here theoretical models based on the possible effects of mode coupling and alternatively of the force contribution of a zero average radial current perturbation that does not violate quasi-neutrality, and with high electrical conductivity may give giving a strong damping rate similar to that in a Hartmann flow.The discussion is supported by the numerical solution of a full RMHD problem for the m=2,n=1 mode including an equilibrium toroidal velocity profile with boundary conditions imposed by an external helical current Iext. Finally the implications on the effects on transport phenomena due to the induced modifications of the EXB shear are indicated.

[GP1.120] Spontaneous Plasma Rotation in Alcator C-Mod

Toroidal rotation in Alcator C-Mod experiments(J.E. Rice, et al., Nucl. Fus. \bfseries 39), 1175 (1999) with no external momentum input is analyzed, putting an emphasis on comparison with neoclassical predictions. Neoclassical theory correctly estimates the direction and the order of magnitude of the impurities' rotation in the L-regime. In the H-regime, rotation is in the opposite direction. Although the formation of steep pressure gradients favors co-current rotation, this effect alone is insufficient to explain it's magnitude in the central zone. Therefore, the presence of a substantial outward radial electric field in the plasma core is required to fit experimental observations. Another strong argument against the neoclassical interpretation is the edge transport analysis for Alcator C-Mod. It shows extremely large values of the transport coefficients that grow towards the edge, indicating highly anomalous transport. Large recycling fluxes point to the importance of momentum exchange with the wall. This makes the ``accretion'' theory(B. Coppi, 18^th) IAEA Fusion Energy Conf. (Sorrento, Italy 2000), Paper IAEA-CN-77/THLP/17 the most probable candidate for successful descriptions of angular momentum ``generation''.

[GP1.121] Linear and Nonlinear evolution of driven ideal MHD instability

Ideal MHD is quite successful in determining the equilibrium and stability regime of toroidal devices. However, many present day tokamaks operate in high performance regimes (i.e., high \beta) that are very close to the ideal MHD stability threshold. Thus, high \beta disruptions caused by ideal MHD modes are a key issue. For example, a particular disruption in DIII-D plasma has been well modeled [1] by an ideal MHD interchange like instability driven slowly (via heating) through its instability threshold.

Here, we will present the linear stability analysis of a driven ideal, long wavelength interchange mode. In a sheared slab geometry, a linear eigenvalue equation is derived and solved both analytically and numerically. The Suydam's instability criterion D_I > 1/4 is obtained along with the structure of the eigenmodes for marginal stability and slowly growing instabilities. Further, the nonlinear equations for this model are derived using the reduced MHD equation in the small k_\parallel/k_\perp limit. Progress towards the analytical solution of the early nonlinear phase of these equations will also be presented.

[1] J. D. Callen, C. C. Hegna, B. W. Rice, E. J. Strait and A. D. Turnbull, Phys. Plasmas 6, 2963 (1999).

[GP1.122] Efficiently Finding Trends in Macroscopic MHD Stability Using Perturbed Equilibria

The effects of equilibrium shaping and profiles on long wavelength ideal MHD instabilities in toroidal plasmas are traditionally studied using numerical parameter scans. Previously, we introduced a new perturbative technique to explore these dependencies: assuming small equilibrium variations, new stability properties are found using a perturbation of the energy principle rather than with a traditional stability code. With this approach, stability dependencies can be efficiently examined without numerically generating complete MHD stability results for every set of parameters (which can be time-intensive for accurate representations of several configurations). Here, we briefly expand on previous successful perturbed stability analyses for screw pinch equilibria by discussing cases where the approach fails. Next, we extend the approach to toroidal geometry using the GATO and TOQ codes, and present cases that both validate the approach and suggest caution in its application.

[GP1.123] A Kinetic Treatment of a Perpendicular Gradient in Field-aligned Flow in a Thermally Anisotropic Plasma

The linearized dispersion relation describing waves in a plasma having a uniform magnetic field, uniform density, and shear in parallel (to the field) flow [G. Ganguli, M.J. Keskinen, H. Romero, R. Heelis, T. Moore, and C. Pollock, J. Geophys. Res., 99, 8873, 1994.] is generalized to include thermal anisotropy, a key feature existing in many space and laboratory plasmas. The effects of thermal anisotropy on the ion acoustic mode and the ion cyclotron mode are examined. The growth rate of the ion acoustic mode is shown to increase with T_i/T_i, and the real frequency at which the maximum growth rate occurs is shown to upshift significantly. The angle that an ion acoustic wave propagates is also shown to depend onT_i/T_i. The growth rate for the ion cyclotron mode is shown to increase with T_i/T_iin the presence of inhomogeneous flow, and the real frequency is not significantly affected. Also presented is a generalized calculation of perturbed distribution functions [Sarfaty, M., S. DeSouza Machado, F. Skiff, Phys. Plasmas, 3, 4316, (1996); Skiff, F., IEEE Transactions of Plasma Science, 20, 701 (1992)] to include shear in field-aligned flow. Without shear, the first order perturbed distribution as a function of \nu is independent of the orientation of the wavevector in the plane perpendicular to the background magnetic field. A method of determining the wavevector components present in a plasma with shear in the parallel flow is presented.

[GP1.124] Applications, Sources and Diagnostics

[GP1.125] Experiments and Modeling of the Transportation and Re-deposition of Etched Products and their Influence on Cross Wafer Uniformity

The transport and re-deposition of etched products are modeled in an inductively coupled RF plasma reactor. To simplify the convoluted etching environment, the reactor has been equipped with a mock wafer housing an orifice regulated by a needle valve. The orifice injects a stream of gas, representing etched products leaving the wafer surface, into the operating plasma reactor with fill pressures ranging from 6 – 30mT. A Langmuir probe provides the plasma density and electron temperature information. Optical Emission Spectroscopy (OES) is used to collect radial profiles of the injected species, and these profiles are modeled by a point source with a uniformly distributed sink. This gas phase model is coupled with a surface reaction model, which is obtained by measuring the carbon film thickness on silicon samples when methane is injected through the orifice. A mass spectrometer provides addition gas phase data when OES signals are difficult to determine, as well as monitoring the precursors to the carbon deposition.

[GP1.126] Control of Dissociation by Different Dilution Gases for Plasma Processing

In plasma processing the composition of the plasma is important parameter for etch and deposition rates. The dissociation rate can be controlled by the electron temperature and density. Electron temperature and density are modeled by using simple 0-d particle and power balance modeling and measured by Langmuir probe in pure He, Ar, and Xe plasmas. Trace amounts of oxygen gas is added to each of the inert gas plasma and dissociation of oxygen gas is studied by actinometry and by mass spectroscopy with various powers and various ratios of O2/inert gas. Reasonable agreement between these results and the neutral atomic oxygen density estimated using a simple model based on the measured electron density and temperature. Preliminary dissociation results using fluorocarbon will also be shown.

[GP1.127] Diagnostics of low-frequency inductively coupled plasmas for synthesis of carbon-nitride-based materials

Diagnostics of low-frequency (\sim 500 kHz) inductively coupled plasmas in the process of synthesis of carbon-nitride-based compounds is reported. Nanograins of AlCN were formed by chemical vapor deposition in a low-pressure (20 - 70 mTorr) discharge of Ar+N_2+CH_4 gas mixture. The diagnostics include a movable RF-compensated Langmuir probe and an optical emission spectroscope (OES). The electron/ion number densities and electron temperatures in the film growth process are typically in the range 3\times 10^11-1.5\times 10^12cm^-3 and 3-7 eV, respectively. The dependence of the global plasma parameters on the RF power, the feedstock gas composition, and the pressure is studied. In particular, the electron energy distribution functions (EEDFs) in Ar+N_2+CH_4 discharges appear to be Dryuvestein-like, with the peaks shifted towards higher electron energies. This can be attributed to the effect of the DC substrate bias and the quasi-stationary magnetic fields near a DC magnetron sputtering electrode. Using the EEDF, OES, and reaction threshold data, the electron population responsible for the major dissociation/ionization reactions is computed. In~situ OES and EEDF results are related to the ex~situ chemical bonding states measured using x-ray photoelectron spectroscopy.

[GP1.128] Demonstration and Analysis of Materials Processing by Ablation Plasma Ion Implantation (APII)

Experiments have demonstrated laser-ablated Fe ion implantation into Si substrates. Baseline laser deposited films (0 kV) showed an amorphous Fe-Si film overlying the Si substrate with a top layer of nanocrystalline Fe. APII films exhibited an additional Fe ion-induced damage layer, extending 7.6 nm below the Si surface. The overlying Fe-Si layer and Fe top layer were amorphized by fast ions. Results were confirmed by XPS vs Ar ion etching time for depth profile of the deposited films. XPS showed primarily Fe (top layer), transitioning to roughly equal Fe/Si , then mostly Si with lower Fe (implanted region). These data clearly prove Fe ion implantation into Si, verifying the feasibility of APII as an ion acceleration and implantation process [1]. SRIM simulations predict about 20 percent deeper Fe ion penetration than data, due to:(a) Subsequent ions must pass through the Fe film deposited by earlier ions, and (b) the bias voltage has a slow rise and fall time. Theoretical research has developed the scaling laws for APII [2]. Recently, a model has successfully explained the shortening of the decay time in the high voltage pulse with the laser ablation plasma. This reduces the theoretical RC time constant, which agrees with the experimental data. * Research supported by National Science Foundation Grant CTS-9907106 [1] Appl. Phys. Lett. 78, 3785 (2001) [2] Appl. Phys. Lett. 78, 706 (2001)),

[GP1.129] Novel Surface Layers and Thin-Films from Intense Pulsed Ion Beams

Materials applications of ion beams are being investigated on the 700 kV RHEPP-1 facility at Sandia National Laboratories. Surface modification using melt-resolidification cycles for property improvement is possible in the fluence range 1-5 J/cm^2, with ablation and thin-film synthesis at 5-20 J/cm^2 fluences. Any number of accelerated species (H, He, C, N, Ne, Ar, and Xe) can be formed in the MAP (Magnetically Confined Anode Plasma) gas-breakdown ion source. We have compared the microstructure of Pt mixed into Ti substrates using both ion beams and high-dose ion implantation. While microstructural changes produced are different, in both cases the length scale for modification greatly exceeds the ion range. Comparison of ion beam-ablated Pt films with sputtered layers also shows significant microstructural differences. Experiments with polymer modification for hardening and composite applications are also ongoing. Thin-films are being formed from graphite, YBCO, and BaTiO_3 targets for hard-coatings, superconducting, and high-density energy storage, respectively. Rare-earth doped films are being investigated for photoluminescence applications. Surface topography, SEM/TEM, stoichiometry, and optical measurements will be shown.

*Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Co., under US DOE Contract DE-AC04-94AL85000.

[GP1.130] Atomic-scale study on plasma etching using classical molecular dynamics simulation: Si/SiO2 selective etching by halogen

Better understanding of plasma-surface interaction, especially atomic-scale reaction dynamics on the surface, is of significant importance for the progress of plasma-processing technologies. Classical molecular dynamics (MD) simulation based on pre-constructed interatomic potential database has been shown to be a powerful technique to reproduce complex physical/chemical surface reaction dynamics during plasma processing. We have developed classical interatomic potential sets for Si/O/Cl and Si/O/F systems using interatomic potential data obtained from ab-initio quantum mechanical calculations. Based on these potential functions, we have performed MD simulations of Si/SiO2 beam etching and reactive ion etching (RIE) by halogen (Cl or F), as well as Si etching by Cl in an oxygen atmosphere. The latter process is often employed in actual Si/SiO2 selective etching in the semiconductor industry. In the presentation, we shall discuss our simulation results on above-mentioned processes and report the recent progress in our attempt to develop potential functions for MD simulations of polymer (hydro or fluorocarbon) etching by argon and/or oxygen.

[GP1.131] Kinetic simulation of an inductively coupled discharge using an Eulerian Vlasov code

The physics of an inductively coupled discharge with the fully nonlinear electron and ion dynamics, is simulated with a kinetic Eulerian Vlasov code. By calculating self-consistently the electrostatic potential evolution on the wave time-scale, important insight is gained on the generation of the longitudinal electrostatic field and currents at the second harmonic of the applied RF field. The generated second harmonic currents are more pronounced at lower frequency, in good qualitative agreement with experimental observation. The first simulations were 1-D in both X and Vx, with the Vy treated in a fluid approximation. New simulations, with two fully kinetic dimensions in V space will also be presented

[GP1.132] Equation of state and electrical conductivity of a strongly correlated aluminum plasma

We present experimental measurements and theoretical estimations of the equation of state and of the electrical conductivity of a hot expanded aluminum plasma (T=1-3 eV, \rho=0.1-0.3 g/cc, P=15 kbar, \Gamma=Z^2e^2/akT \simeq1). The plasma was obtained with a new experimental facility : Enceinte à Plasma Isochore (EPI) designed to confine electrical plasma discharges up to 25kbar. Theoretically, the properties were determined by ab initio molecular dynamics simulations in the DFT/LDA-GGA approximation, from which the conductivity was computed using Kubo-Greenwood formula. Equation of state and electrical conductivity are compared with Sesame tables and with a Lee and More approach using a self consistent estimation of the ionization. As suggested by ab initio simulations, \mathrm Al_2 molecules and \mathrm Al_n clusters seem to play an important role in this regime, explaining discrepancies between models.

[GP1.133] Simulations of Ultracold Plasmas

Ultracold plasmas are formed by the laser photo-ionization of clusters of laser-cooled ions. [S. Kulin, T. C. Killian, S. D. Bergeson, and S. L. Rolston, Phys. Rev. Lett. 85, 318 (2000).] The laser-cooled atoms are localized in a small region (rms radius typically 200 \mum) with a very low temperature (as low as 10 \muK). The frequency and intensity of the ionizing laser can be varied, so that initial electron energies can range from 0.1 K to 1000 K, with peak electron densities ranging from 10^5 to 10^9 cm^-3.

Using a variety of techniques, we simulate the prompt electron loss and the plasma oscillations excited, the evaporative electron loss, and the electron thermalization via electron-ion and electron-electron collisons. We also investigate the electron temperature evolution and electron loss during the free expansion (Coulomb explosion) of the electron-ion cloud.

[GP1.134] Hollow-anode electron source

The parameters of an electron source based on hollow-anode discharge and the generated electron beam are presented. A gas-puff valve is used to provide a sharp pressure gradient between the anode cavity and the accelerating gap. The parameters of the He gas and the plasma inside the hollow anode and the accelerating gap were studied by electrical and optical diagnostics for different gas pressures and different amplitudes of the discharge current. The hollow-anode discharge is initiated by six arc pulsed sources or by a Penning discharge plasma source. The parameters of the electron beam under an accelerating pulse of 200 kV, pulse duration of 300 ns for different arc discharge currents, gas pressures and anode grid cell sizes are presented.

[GP1.135] Design of A Large Oxide Coated Cathode Plasma Source for Operation in High Magnetic Fields at the New LAPD

We use a Barium Oxide coated cathode to supply accelerated electrons as an energy source to from our plasma. Oxide coated cathodes have been used for decades in vacuum tubes and plasma research. Most of these have been small (1 cm dia.) or designed to operate in a low magnetic field where the \boldmath J\timesB \unboldmath forces on them are negligible. At the new LAPD we will have large diameter plasma sources at both ends of the machine which must operate in a 3.5 kG ambient magnetic field. We have designed and built one such source which is 72 cm in diameter. It will supply up to 20 kA of pulsed beam current and uses a 1 m by 1 m, 2.5 kA (dc), 150 kW heater. Solutions to various engineering issues will be discussed. These pertain to differential thermal expansion over 1 m distances, \boldmath J\timesB \unboldmath forces on the heater and cathode, heat containment and uniformity of the oxide coating and of plasma production. These issues are important to any experimenter who plans to build an oxide coated plasma source.

[GP1.136] Rf sheet-plasma production using permanent magnets

High-density sheet-plasmas with a rectangular cross-section of 140 mm \times 20 mm are developed by inductive rf discharge using a rectangular discharge section and a pair of permanent magnets. The stainless-steel discharge section is 200 mm wide, 20 mm high, and 100 mm long. A pair of ferrite permanent magnets (length L_mag = 20 - 140 mm in S-N direction, width W_mag = 50 - 170 mm, and height = 24 mm) is placed on top and bottom of the discharge section and a static magnetic field of B_0 \simeq 600 - 800 G is generated under the center of the magnets. Rf current (frequency = 13.56 MHz and power P_rf \leq 4.5 kW) is applied to an internal antenna covered with a quartz tube in the direction perpendicular to B_0. The antenna is located behind the magnets where B_0 is nearly zero. Plasma density n_p profile is controlled by varying W_mag and distance between the antenna and the magnets due to cusped magnetic field generated by magnets. 140 mm wide plasma (n_p \simeq 2.5 \times 10^12 cm^-3) of a uniformity variation within 90% is produced using 140 mm long antenna for L_mag = 20 mm, W_mag = 120 mm, Ar pressure = 20 mTorr, and P_rf = 3 kW.

[GP1.137] Properties of the Micro-Discharge in Connection with Application to the Plasma Display Panel

One of the most important issues in the plasma display panel (PDP) is the electrical discharge in high-pressure inert-gas, where the ultraviolet (UV) light emitted from xenon discharge plasma is converted into fluorescent light, providing a TV image. The discharge plasma is generated by the electrical breakdown. Reduction of the discharge voltage is therefore the key element in enhancing the electrical efficiency of the PDP. The electrical efficiency enhancement in turn prolongs the panel lifetime. In this context, a better understanding of the electrical discharge in the PDP cells is needed to improve panel performance. However, measurement of plasma parameters in the electrical discharge of the PDP cells is very difficult due to the small size of cells. Global properties of the electrical discharge in cells of the plasma display panel are investigated by making use of the circuit model, where the plasma properties are characterized by the plasma resistivity. The electrical charge and current flowing through the discharge space are calculated from the theoretical model and compared with experimental observations. The experimental data agree reasonably well with the theoretical predictions.

[GP1.138] Microwave Interferometer for Steady-State Plasmas

Standard single frequency, "fringe-counting," microwave interferometers are of limited use for steady-state plasma experiments. We have constructed a swept frequency microwave interferometer, similar to a classic zebra-stripe interferometer, optimized for electron density measurements in steady-state plasma experiments. The key element in the system is a frequency doubled YIG oscillator capable of sweeping from 20 to 40 GHz. As the source frequency is swept, the sum of the reference and plasma leg signals exhibits a series of beats. Both the frequency shift and phase shift of the beat pattern due to the addition of plasma in one leg of the interferometer is used to determine the line-integrated electron density.

[GP1.139] Measurement of High Frequency Perturbations to the Ion Velocity Distribution in the HELIX Helicon Plasma Source

Using lasers to measure plasma parameters has become more common in recent years. Lasers can provide information about plasma parameters without perturbing the plasma. The most common technique for ion parameter measurements is Laser Induced Fluorescence (LIF). LIF typically measures the ion velocity distribution and provides information about the ion temperatures and ion flows in the plasma. More recently, Skiff and Anderegg [1987] and Safarty et al. [1996] have shown that measurements of the perturbed ion velocity distribution can provide wave number information for waves propagating in a plasma due the non-local nature of the dielectric tensor. In the past two years, attempts have been made to measure the perturbed ion velocity distribution function at frequencies relevant to Helicon plasma sources. The objective of the measurements is to identify electrostatic oscillation associated to the slow wave or "Trivelpeice Gould modes" in helicon plasma sources. Past efforts to measure the perturbed ion velocity distribution function have been unsuccessful due to technical difficulties associated with measuring the cross correlation of the photon and reference signals. Using a high frequency SR544 Stanford Research lock-in amplifier, high frequency perturbations to the ion velocity distribution in a helicon source have been measured. Perturbed ion velocity distribution measurements, along with the related theory will be presented.

[GP1.140] Study of charge collection in the wide range of operating pressure

In recent years, the use of the Langmuir probe in the high-pressure plasma becomes more important in industrial application. However, due to the difficulties of considering the collisional effects on charge collection, the theoretical and experimental works have not been performed sufficiently yet in the high-pressure regime. To investigate the collisional effects on charge collection, we measured some I-V characteristic curves in the wide range of pressure plasma. The plasma has been generated by using a 2.45GHz magnetron in a quartz tube of 36mm diameter. In the pressure range of several torr, the ratio of electron saturation current to ion saturation current seems abnormally to be reduced. The experimental results and PIC simulation on the charge collection in high-pressure plasma regime will be shown.

[GP1.141] Generation and spectroscopic diagnostics of atmospheric microwave-induced torch plasma

A microwave induced plasma was generated at atmospheric pressure assisted by argon neutral gas introduction. To maintain the plasma stable, appropriate operating conditions such as quartz tube dimension, gas flow rate, and microwave power were required. At a specific operating condition, three distinctive plasma regions were observed, which are filament region, filament-converging region, and flame-like region. Using spectroscopic methods, plasma temperatures and electron density were measured. The electron density was on the order of 1015 cm-3 from Stark broadening of the H¥á (656.3 nm) line. Electron excitation temperature and gas temperature were obtained by the Boltzmann plot of neutral argon spectral lines and OH radical lines. The measured excitation temperature and the gas temperature were about 7500 K and 4200 K, respectively.

[GP1.142] Development of simulator of corona reactor

A simulation reactor by using a pulsed corona discharge for the simultaneous removal of NOx and SOx from the flue gas has been developed. As a simulator, plate-cylinder type reactor was used with specially designed inner electrode array. The design philosophy, structure, electronics, and diagnostics will be introduced along with the first result of NO removal experiment. The effects of system parameters such as voltage frequency, gas flow rate, gas temperature, or power input will be discussed.

[GP1.143] Numerical simulation of a pulsed corona discharge plasma

In this paper we study theoretically a pulsed corona discharge plasma in air at atmospheric pressure. A self consistent one-dimensional model, using a flux-corrected transport numerical scheme, is presented to achieve it. The charged particle kinetics in this model are described by one-dimensional continuity equations coupled with Poisson's equations. The spatio-temporal local field, charge density variations and the velocity profile of an ionizing front are calculated to describe the dynamical behavior of corona discharge plasmas. The simulation results of a wire-in-cylinder corona discharge plasma explain the physical mechanisms of discharge processes. These results may also be apply to obtain the optimizing parameters for designing the plasma reactor.

[GP1.144] Molecular Dynamics Calculation of Carbon/Hydrocarbon Reflection Coefficients on a Graphite Surface Employing Distributed Computing

Reflection coefficients of carbon and hydrocarbon molecules have been calculated with a molecular dynamics code. The code uses the Brenner hydrocarbon potential, an empirical many-body potential that can model the chemical bonding in small hydrocarbon molecules and graphite surfaces. A variety of incident energies and angles have been studied. Typical results for carbon show reflection coefficients ~0.4 at thermal energy, decreasing to a minimum of ~0.15 at 10-20 eV, and then increasing again. Distributed computing is used to distribute the work among 10-20 desktop PCs in the laboratory. The system consists of a client application run on all of the PCs and a single server machine that distributes work and compiles the results sent back from the clients. The client-server software is written in Java and requires no commercial software packages. Thus, the MD code benefits from multiprocessor-like speed-up at no additional cost by using the idle CPU cycles that would otherwise be wasted. These calculations represent an important improvement to the WBC code, which has been used to model surface erosion, core plasma contamination, and tritium codeposition in many fusion design studies and experiments.

Part G of program listing