Student Abstracts: Physics at PPPL

Effect of Nearby Conducting Structure on the Macroscopic Stability of NSTX Plasmas. ANNIE AHNERT (The University of Arizona, Tucson, AZ 85721) DR. JON MENARD (Princeton Plasma Physics Laboratory, Princeton, NJ 08543) .
The ability to predict the macroscopic stability of plasma is an important part of current magnetic fusion research. There are many different programs to predict the stability of plasma, but these programs also require approximations, such as idealized shapes for the closest conducting surface adjacent to the plasma. A program that was able to generate the approximate shape of the conducting surface was written with parameters to control the smoothness of the approximation. The program was used to generate conducting wall shapes accurately simulating the effects of the National Spherical Torus Experiment (NSTX) vacuum vessel and passive plates on the n=1 kink mode stability of a high beta=40% NSTX target plasma. Future studies will investigate the effect of the inner and outer conducting wall on lower beta plasmas.

BPST Coil Design and Optimization. ERIC HARKLEROAD (Princeton University, Princeton, NJ 08544) NEIL POMPHREY (Princeton Plasma Physics Laboratory, Princeton, NJ 08543) .
We seek an optimal configuration of Ohmic Heating (OH) and Equilibrium Field (EF) coils for the proposed Burning Plasma Spherical Tokamak (BPST) at PPPL. To determine the EF coil positions for the new device, we scale the design of the National Spherical Torus eXperiment (NSTX), and determine a discrete set of candidate coil positions. We then employ the Tokamak Simulation Code (TSC) to isolate which small subset of these coils experience the greatest changes in current as plasma shape and current profile vary. We select these as our design. Upon fixing the EF coil positions, we optimize the height of the center stack by varying the height and noting variations in the ohmic field error, selecting a height at which these errors are acceptably small. We next seek an ohmic current distribution-a set of EF and OH coil currents which produces little or no field within the plasma. The field produced by an arbitrary current distribution is invariant under addition of a scalar multiple of an ohmic distribution. We outline two methods of approximating an ohmic distribution and implement them in Fortran 90, calling on subroutine E04UNF from the Numerical Algorithms Group (NAG) Library, a specialized routine designed to minimize a function of many variables. We also include provisions for efficient optimization of a center stack of ohmic heating coils. Coupled with other studies and simulation codes, our coil design and optimization codes have the potential to make significant contributions to the design of BPST.

Development and Analysis of an Electrically Tunable Optical Filter. KRISTI HULTMAN (Harvey Mudd College, CLaremont, CA 91711) FRED LEVINTON (Princeton Plasma Physics Laboratory, Princeton, NJ 08543) .
The focus of my research experience was the development of an electrically tunable optical filter using a He-Ne laser and a combination of polarizers (P), LN crystals (C), and retarders (R). A variable high voltage power supply was attached to the LN crystals, allowing us to optimize transmission of specific wavelengths. The retarder used was made from 3 polarizing lenses, with the middle lens rotated 59.0° off axis. For analysis of the filter transmission, a LabView program was written to display the image of the beam captured using a CCD camera, as well as peak and average intensities, and save the data. We were successful in tuning the laser to a minimum and maximum in both the PCP and PCRCP configurations, and can resolve wavelength variations of less than 0.1nm. A stable, easily tunable optical filter would allow for a cleaner signal when looking at a specific wavelength or allow you to block out a certain wavelength, thus the noise from unwanted wavelengths would be reduced.

Electron Bernstein Wave Polarization Measurements on CDX-U. THOMAS KRAMER (Brown University, Providence, RI 02912) PHILIP EFTHIMION (Princeton Plasma Physics Laboratory, Princeton, NJ 08543) .
Mode-converted (MC) EBWs offer an attractive path for electron temperature measurement, heating, and current drive in overdense plasmas (Plasma frequency >> Cyclotron frequency). 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 optimises 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 ~1 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. *Work Supported by U.S. DOE Contract DE-AC02-76CH03073

TSC Plasma Simulations for NSTX Center Stack Upgrade. ANDREW OSGOOD (Muhlenberg College, Allentown, PA 18104) DR. STANLEY KAYE (Princeton Plasma Physics Laboratory, Princeton, NJ 08543) .
The National Spherical Torus Experiment, or NSTX, is the primary fusion device at PPPL. In an effort to continue efficient use of the machine, a planned center stack upgrade needs numerous computer simulations to determine its practicality. Using specially designed programs called Tokamak Simulation Code, or TSC, numerous qualities and quantities can be accurately simulated. Initially, a shape range had to be determined using the static version of TSC, since the shape of the plasma is integral in many other practical aspects such as stability. Once a stable static shape range was determined, the pf coil currents could be used in the dynamic TSC version to develop a more plausible plasma that evolved through time. For both 1.5MA and 3.0MA plasmas (the only two plasma currents simulated) the TSC produced acceptably wide shape ranges. The 1.5MA plasmas found a wider range, since much higher values of li and a lower current allowed control from pf coils through lower currents. (The pf coils have maximum current limits that restrained most 3.0MA runs.) These runs produced values that were used as input for the dynamic TSC runs, but also illustrated that an adequate shape range could be produced using the new center stack upgrade parameters. The coil currents produced in static simulations will be used in continuing dynamic simulations that strive for specific plasma properties according to future needs. Starting the dynamic runs has shown that this is possible and can produce viable results, and many more simulations will follow.

Study of Magnetic Damping in Liquid Metal Surface Waves. DAVID PACE (University of the Pacific, Stockton, CA 95211) DR. HANTAO JI (Princeton Plasma Physics Laboratory, Princeton, NJ 08543) .
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. More recent findings have demonstrated that a magnetic field aligned parallel to the direction of wave propagation causes significant damping of the waves which follows a gaussian dependence, and confirmed that the wave number varies in the presence of a magnetic field.

Stability Tests of Hydrodynamic Flows in Water for Laboratory Study of Magnetorotational Instability. ETHAN SHOSHAN (Rutgers University, New Brunswick, NJ 07751) HANTAO JI (Princeton Plasma Physics Laboratory, Princeton, NJ 08543) .
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.