Session QP1: Poster Session VI: NSTX; C-Mod; Non-Neutral Plasmas; Acceleration and Heating in Space; Mini-Conf. on Interface Between Fluid and Kinectic Processes

An Overview of NSTX Research Facility and Recent Experimental Results

The 2006 NSTX experimental campaign yielded significant new experimental results in many areas. Improved plasma control achieved the highest elongation of 2.9 and plasma shape factor q$_{95}$I$_{p}$/aB$_{T}$ = 42 MA/m$\cdot $T. Active feedback correction of error fields sustained the plasma rotation and increased the pulse length of high beta discharges. Active feedback stabilization of the resistive wall mode in high-beta, low-rotation plasmas was demonstrated for $\sim $100 resistive wall times. Operation at higher toroidal field showed favorable plasma confinement and HHFW heating efficiency trends with the field. A broader current profile, measured by the 12-channel MSE diagnostic in high beta discharges revealed an outward anomalous diffusivity of energetic ions due to the n=1 MHD modes. A tangential microwave scattering diagnostic measured localized electron gyro-scale fluctuations in L-mode, H-mode and reversed-shear plasmas. Evaporation of lithium onto plasma facing surfaces yielded lower density, higher temperature and improved confinement. A strong dependence of the divertor heat load and ELM behavior on the plasma triangularity was observed. Coaxial helicity injection produced a start-up current of 160 kA on closed flux surfaces.   

NSTX Progress and Plan of Interest to Component Test Facility (CTF)

Continued rapid progress on NSTX and more broadly in Tokamak and ST plasma science has indicated relatively robust physics conditions in a broad number of topical areas for a compact CTF [1], which is included in the DOE Office of Science Strategic Plan [2]. This progress has enabled an updated projection of the practical CTF plasma conditions. The results indicate appropriate designs with R$_{0}$ = 1.1-1.2 m, A = 1.5, elongation $\sim $ 3, B$_{T} \quad \sim $ 1.5-2.5 T, and a range of I$_{p}$ = 6-12 MA to deliver a fusion neutron flux of 0.5-4.0 MW/m$^{2}$, requiring a range of 30-70 MW of combined neutral beam and RF heating and current drive power. Database is evaluated to be adequate in Macroscopic Plasma Physics; Multi-scale Plasma Physics; Waves and Energetic Particles; and Physics Integration; but not yet adequate in Plasma Boundary Interface (high divertor heat flux) and Solenoid-Free Operations (current initiation and ramp-up). Near-term ST research to strengthen and fill in the needed database will be described, including a discussion on how the CTF testing program could begin with plasma facing component testing in D-D at low currents followed by fusion component testing in D-T at higher currents. [1] Plasma Phys. Contol. Fusion \textbf{47} (2005) B263. [2] http://www.sc.doe.gov/bes/archives/plans/SCSP{\_}12FEB04.pdf.   

NSTX Plasma Start-up using Transient CHI

Non-solenoid based start-up is essential for development of a fusion reactor based on the spherical torus concept. A method of non-inductive startup, referred to as transient coaxial helicity injection (Transient CHI), was successfully developed on the Helicity Injected Torus (HIT-II) experiment and employed on the National Spherical Torus Experiment (NSTX) to produce up to 160 kA of toroidal plasma current on closed flux surfaces without use of the central solenoid. In this method, plasma current is produced by discharging a capacitor bank between coaxial electrodes in the presence of toroidal and poloidal magnetic fields chosen such that the plasma rapidly expands into the chamber. When the injected current is rapidly decreased, magnetic reconnection occurs near the injection electrodes with the toroidal plasma current forming closed flux surfaces. Detailed experimental measurements made on NSTX include fast time-scale visible imaging of the entire process.   

The Development of LITER - a Lithium Evaporator for Use in Fusion Devices

An evaporator to be used for the deposition of elemental lithium onto the plasma facing components of fusion devices has undergone several stages of design, evaluation and development. A 150 Watt prototype evaporator with a modest 10 gram reservoir was first tested in the laboratory before it was successfully employed on the CDX-U device in a horizontally-pointing geometry. Drawing on that experience, the prototype evaporator design then evolved through several stages as its reservoir capacity was increased to 50 to 100 grams, its power consumption increased to about 300 Watts and its pointing-orientation changed from horizontal to vertically-downward so that it could be used to coat the lower divertor in the NSTX device. These design changes were dictated from both laboratory experience and from practical experience on NSTX. The most challenging problems associated with the evaporator have been the need to achieve reasonable rates of evaporation (1 to 100 mg/ min) at operating temperatures in the range of 500 C to 700 C while avoiding the uncontrolled wetting of the lithium onto external regions of the evaporator. The design and performance of the evaporator at each stage of its development will be summarized and discussed in detail.   

Effect of Evaporated Lithium PFC Coatings in NSTX

Two versions of a lithium evaporator (LITER-1) were installed in succession on an upper port aimed toward the graphite tiles of the lower center stack and divertor. Lithium temperatures in the range 450 to 680$^{\circ}$C produced evaporation rates of 0.08 to 35 mg/ min with a gaussian-like angular distribution with a 1/e full width of about 22$^{\circ}$. A quartz crystal micro-balance in the gap between two tiles on the lower divertor was used to measure the lithium deposited. Twelve separate depositions, ranging from about 10 mg to 5 g of lithium were performed. Lower single-null L-mode and H-mode, and double-null reversed-shear plasmas were studied. After lithium coating, the reference ratio of oxygen to carbon emission was lower than achieved with boronization. Other effects of lithium coating were variable, but under some conditions, there were decreases in the density and increases in electron and ion temperature, and neutron rate, and reductions in H-mode ELM frequency. These results and future plans will be discussed.   

Results from the NSTX MSE-CIF diagnostic

A twelve channel Collisionally-Induced Fluorescence Motional Stark Effect diagnostic, covering the plasma minor radius on the outboard side, is routinely operated for the primary purpose of providing internal magnetic pitch angle measurements for equilibrium reconstruction. While the NSTX MSE-CIF requires the use of novel high-throughput, narrow bandpass (0.07 nm) Lyot filters to operate at low magnetic field (0.3-0.55T), a traditional PEM-based polarimeter is used to make the angular measurement. The polarimeter is calibrated by reconciling MSE measured angles with magnetic reconstructions during beam injection into gas-fill-torus with vacuum fields. It has recently been suggested that fast ions resulting from ionization of the beam neutrals with the fill gas may contaminate the measured angle by emitting additional Balmer-alpha after re-neutralizing via charge-exchange. Results from 3D simulations of this widely used MSE calibration technique from several machine geometries, including NSTX, C-Mod, and ITER will be presented. Highlights from MSE results for the 2006 NSTX run campaign will also be presented. Supported by DOE contracts DE-FG02-99ER54520 and DE-AC02-76CH03073.   

Confinement Scaling and Transport Physics in NSTX

The confinement time trends in high power NSTX neutral beam heated discharges exhibit significant differences from those at conventional aspect ratio; the scaling with toroidal magnetic field is stronger while that with plasma current is weaker. The improvement of confinement with increasing field is due primarily to a reduction of the electron transport. The electron transport is the dominant anomalous loss mechanism in NSTX, with the ion transport can be close to neoclassical levels. Predictions from linear GS2 gyrokinetic calculations and from analytic theory indicate that ETG modes may play a role at low field. Details of the parametric trends of confinement and transport, along with the relation to changes in the high-k fluctuations and methods for improving the electron confinement, will be discussed.   

Perturbative electron transport experiments using pellet injection in NSTX

After the edge perturbation from Type-I ELMs, the perturbed electron transport in NSTX transiently exhibits unusual features, such as ms time scale, global electron temperature crashes. To study this cold pulse propagation we use `multi-color' SXR imaging with fast diode and `optical' arrays and produce controlled perturbations with shallow injection of low-Z pellets. The analysis indicates stiff electron temperature profiles and large perturbed electron thermal diffusivity for both pellet and Type-I ELM perturbations. At the same time, the particle diffusivity obtained from the evolution of the pellet injected impurity is about two orders of magnitude lower than the perturbed electron diffusivity, being comparable to the neoclassical value. This seems to exclude low-k electrostatic fluctuations as a cause for the rapid perturbed electron transport in NSTX, suggesting instead high-k fluctuations, or magnetic transport. Results of experiments are presented in which perturbations were introduced as the heating power and q-profile were varied. Supported by DOE Grant DE-FG02-99ER5452 at JHU and Contract DE-AC02-76CH03073 at PPPL.   

Impurity injection studies in NSTX H-modes using a tangential ``optical'' soft X-ray array

First impurity transport experiments in high power and high density NSTX H-modes have been performed by puffing neon at the edge for several tens of ms and following the penetration of the injected ions with a newly developed, tangential multi-color “optical” soft X-ray array (OSXR). The neon penetrates the core on the hundred ms time scale, indicating low particle diffusivity. The peaking of the neon density late in the shot also indicates an inward convective velocity, while a field scan at fixed q-profile shows slower penetration at higher field. Two techniques for measuring the evolution of the impurity density with the OSXR array are compared, the Abel inversion and the forward modeling of the emission profiles. First results of determining the impurity transport coefficients will also be presented. Supported by DOE Grant DE-FG02-99ER5452 at JHU and Contract DE-AC02-76CH03073 at PPPL.   

Coherent and Turbulent Fluctuation Measurements on NSTX Using Millimeter-Wave Reflectometry

Recently the millimeter-wave reflectometers on NSTX were modified to allow several new measurements of both coherent and turbulent density fluctuations in the core plasma. 1) The 28-40 GHz (0.97$-$2.0$\times$$10^{13}$ cm$^{-3}$, O-mode) correlation reflectometer has been upgraded from homodyne to quadrature detection, which allows the dual-channel system to measure fluctuation levels and poloidal flow in addition to radial correlation lengths. 2) A fixed frequency quadrature channel at 42 GHz (2.2$\times$$10^{13}$ cm$^{-3}$, O-mode) was been modified for detection of density oscillations associated with the HHFW launched at 30 MHz. Initial measurements show strong modulation of the reflected wave by both turbulent and coherent fluctuations. 3) Improvements to the FM-CW reflectometer (13-53 GHz, 0.2$-$3.5$\times$10$^{13}$ cm$^{-3}$ for O-mode) now allow 10 $\mu$s repetition rates. Radial profile measurements of $\tilde{n}/n$ for fast ion driven modes (EPM's and possibly low-frequency TAE's) are explored.   

Investigation of core electron gyroscale fluctuations in reverse shear and monotonic-q discharges on NSTX

Electron thermal transport is the dominant loss mechanism across NSTX profiles and within tokamak internal transport barriers. Reverse shear discharges can exhibit improved electron thermal confinement compared to similar monotonic-q discharges. The improved electron confinement may be associated with reduced electron gyroscale fluctuations. With this motivation, a five-channel scattering system is employed to study core electron gyroscale fluctuations in reverse shear and monotonic-q discharges on NSTX. \ Scattering measurements and the subsequent density fluctuation spectra are localized in both real space and $k$-space. The NSTX scattering system can measure fluctuations with $k_\perp\leq20\mbox{ cm}^{-1}$ and $k_\perp\rho_e\leq0.7$ at five discrete wavenumbers. The $k$- space resolution is $\Delta k_\perp\approx0.7\mbox{ cm}^{-1}$. Steerable optics can position the scattering volume at any location throughout the plasma minor radius. In addition to fluctuation spectra, MSE q-profiles, \texttt{TRANSP} transport calculations and gyrokinetic simulations are also presented.   

GRYO simulations of turbulent transport in NSTX L-mode plasmas

We present simulations of long-wavelength, drift-kinetic plasma turbulence in reversed-shear L-mode discharges in NSTX. This work extends studies of core plasma turbulence to low aspect ratio, in plasmas where finite $\rho_\ast$ effects (i.e., profile variations, and `non-local' effects) are expected to be important. In such plasmas the fluctuations can more easily produce large scale deviations from the background density and temperature, and it is essential to remove these deviations with an `adaptive source' model. We demonstrate that the simulation results are converged with respect to the parameters of the `adaptive source' model before the source affects the overall transport. These electrostatic simulations include a fully drift-kinetic treatment of both electrons and ions. The GYRO simulations of a dozen plasma conditions consistently exhibit several characteristics: the predicted ion heat flux is strongly dominant, contrary to TRANSP analysis of the experimental data; non-adiabatic electron effects are very strongly destabilizing; and the ExB shear is strongly stabilizing (as expected) but in some discharges the residual predicted heat fluxes are comparable to the heating power. We will explore whether those simulations that are completely quenched by ExB shear are close to marginal stability.   

Current multiplication and flux amplification in magnetic helicity injection

The usefulness of magnetic helicity injection in toroidal concepts such as the spherical torus under co-axial helicity injection (ST-CHI), the spheromak, and the spherical torus with a plasma center column (ST-PCC), is measured by the concepts of current multiplication and flux amplification. The current multiplication factor is defined as the ratio of the toroidal plasma current inside the discharge chamber and the injector current that is collected on the electrodes. The flux amplification factor is defined as the ratio of the poloidal magnetic fluxes inside and outside the magnetic separatrix. One desires a high current mupltiplication factor because it implies a low engineering recirculating power requirement. The flux amplification factor gives a more direct measure of the effectiveness of the helicity injection scheme. High flux amplification implies that a higher strength plasma confining magnetic field is achieved for a given bias poloidal magnetic field. These two concepts are related, but significant difference exists for the case of ST-CHI, in constrast to the spheromak and ST-PCC cases. We will present the theory and calculation that clarify the subtle physics and its implications for the actual experiments.   

Electron Transport in NSTX

The electron heat transport in NSTX is investigated. The inferred $\chi_e$'s from TRANSP were compared to those calculated from electron temperature gradient (ETG) theory [1] for a variety of discharges. Reasonable agreement was found for electrostatic and electromagnetic ETG theory with the $\chi_e$'s calculated for a 2 MW discharge in both the L- and H-phases. H-mode discharges at low toroidal field ($0.35$T) show good agreement with the electromagnetic ETG, while for discharges at $0.45$T and $0.55$T, the inferred $\chi_e$ agrees with the gyro-Bohm scaing of $T_e^{3/2}/B^2$. For these higher field discharges, GS2 calculations show the ETG to be stable, while it is unstable at $0.35$ T with a maximum growth rate at $k_\theta\rho_e \sim 0.3$. The heat flux obtained from the simulation of the ETG 3-field model [2] will be compared. And the effect of inverted profile $\nabla n_e\cdot\nabla T_e < 0 $ on the ETG electron transport will be discussed. \newline \newline [1] W. Horton {\it et. al.}, Phys.Plasmas \textbf{11}, 2600 (2004). \newline [2] W. Horton {\it et. al.}, Nuclear Fusion \textbf{45}, 976-985 (2005).   

Investigation of microtearing mode effects on electron transport in NSTX

While ETG turbulence is a popular candidate for the explanation of anomalous electron transport in tokamaks, microtearing modes may be particularly important in spherical tokamaks due to their low magnetic field. These modes can be the fastest growing modes at low k$_{\theta }\rho _{s}$ in NSTX\footnote{M.H. Redi et al, Proc. 30$^{th}$ Conf. Control. Fusion Plasma Phys. Pp.494 (2003).} and MAST.\footnote{D.J. Applegate et al., Phys. Plasmas \textbf{11}, 5085 (2004).} Microtearing modes have odd parity electrostatic eigenfunctions, are extended along field lines and they propagate in the electron diamagnetic drift direction. Such analysis will be extended to recent NSTX data. Estimates of their effects on electron transport based on existing nonlinear theory will be presented and compared with experimental results.   

Simulation Studies of Electron Temperature Gradient Turbulence in NSTX Plasmas

Motivated by recent results of high-k measurements in NSTX experiments, we have investigated the electron-temperature-gradient (ETG) driven turbulence and the associated electron transport in NSTX plasmas using a global gyrokinetic simulation. This study contributes to the search for the key microturbulence activities responsible for energy loss through the electron channel in NSTX plasmas under various experimental conditions, including reversed magnetic shear, H- and L-mode discharges. The physical focus of the present study is on i) the dependence of ETG turbulence on the magnetic shear; ii) the effect of equilibrium EXB shear flow; and iii) the effect of collisional dissipation on ETG steady state behavior on the electron collision time scale. The comparison of the k-spectrum in nonlinear saturation with experimental measurements will be discussed.   

ELM Rotation Measurement via the NSTX FIReTIP System

The high time resolution of the Far Infrared Tangential Interferometer/Polarimeter (FIReTIP) allows measurement of the critical parameters of the small Edge Localized Modes (ELMs), also known as a type V ELM, on the National Spherical Tokamak Experiment (NSTX) including the speed of rotation and size of each ELM structure. Understanding of ELMs is extremely important since the operation mode of future fusion devices such as ITER is likely an ELM free high confinement mode (H-mode). An extensive analysis of the type V ELM data demonstrates a strong relationship between the rotation speed and interval of the ELM. This paper includes the upgrade process of the multi-channel FIReTIP system and implications of the ELM physics by its rotation measurements.   

Investigation of Fast-Ion Redistribution or Loss due to MHD Modes and Alfv\'{e}n Instabilities in NSTX

In the National Spherical Torus Experiment (NSTX), fast-ion redistribution or loss due to two types of instabilities has been observed: low-n kink/tearing-like modes and fast-ion driven instabilities such as chirping Energetic Particle Modes (frequency range 0--120 kHz) and Toroidal Alfv\'{e}n Eigenmodes (frequency range 50--200 kHz). Both the volume-integrated neutron and the line-integrated charge exchange neutral particle diagnostics show fast-ion depletion due to such instabilities, but cannot distinguish between redistribution or loss. Two recently implemented diagnostics on NSTX, the Motional Stark Effect (MSE) and scintillator Fast Lost Ion Probe (sFLIP), facilitate separation of redistribution and loss effects. Outward redistribution of the core-peaked energetic beam ions modifies the beam-driven current profile and hence the core q-profile. MSE-constrained q-profiles are used to assess this effect. sFLIP measures the pitch and energy of fast ions that are ejected from the plasma and intercept the wall-mounted probe thus identifying fast-ion loss. Observations and TRANSP simulations of a range of fast ion instability-induced redistribution/loss phenomena in NSTX will be presented.   

Study of beam-ion confinement in MHD-quiescent NSTX plasmas

The NSTX with its extensive set of fast-ion diagnostic (neutron detectors, E$\vert \vert $B-type neutral particle analyzer, scintillator-based beam ion loss probe and newly developed solid state neutral particle array) provides a good test-bed to study the confinement of beam ions in quiet spherical tokamak plasmas. Ten ms pulses (``beam blips'') of 90keV deuterium neutrals are injected into helium plasmas with plasma current between 0.5 and 1.0 MA, and toroidal fields between 3.0 and 4.5 kG. Pitch angle scattering and slowing down of beam ions are studied by measuring the decay of the neutron and charge-exchange neutral particle signals following the ``beam blip'' and they are in good agreement with the expectations of TRANSP simulation, which includes beam deposition calculations, Coulomb collisions and charge exchange loss. Examples of deviations from classical behavior during instabilities are also given.   

MHD-induced Fast Ion Loss from NSTX Plasmas

The Scintillator Fast Lost Ion Probe (sFLIP) on NSTX measures the loss of neutral beam ions lost to the wall near the outer midplane, resolved in gyroradius and pitch angle. The diagnostic has recently been upgraded to record the characteristics of the loss at up to 40,500 frames/s, allowing resolution of losses arising from MHD activity. That loss is often at high pitch angle. Fast ion loss correlated with field error correction and resistive wall mode control fields has also been observed.   

New MHD modes below the geodesic acoustic frequency and beam driven instabilities in NSTX

We report on new global MHD eigenmode solutions which are found associated with the gap in the Alfv\'en/acoustic continuum. The modes are at frequencies below the geodesic acoustic modes (GAM). In contrast to the mostly electrostatic polarization of GAMs the new modes contain electromagnetic component due to the interaction with the Alfv\'en branch. Indeed a new continuum branch emerges with a modified shear Alfv\'en wave dispersion relation. The new modes associated with this shear Alfv\'en branch have phase velocity significantly above the background ion thermal velocity and thus are expected to be insensitive to kinetic effects. These new modes may be associated with so-called BAE modes that have been found (but not explained theoretically) in numerical MHD codes [A. Turnbull, et.al. Phys. Fluids {\bf B 5} 2546 (1993)] and kinetic codes [N.N. Gorelenkov,et.al., Nucl.Fusion {\bf 42} (2002) 150]. We speculate that new modes are being driven by beam ions in NSTX at roughly half of the TAE frequency.   

Investigation of fast ion mode nonlinear dynamics and spatial structure in NSTX

Neutral beam heated plasmas in NSTX exhibit a rich spectrum of fast-ion driven coherent modes that includes fishbones as well as toroidicity-induced and compressional Alfv\'en eigenmodes (TAE and CAE). These modes are of significant interest because they can induce fast ion transport and channel fast ion energy into the plasma. In recent experiments, the spatial structure of fishbone density perturbations has been investigated through the simultaneous application of a 288 GHz radial interferometer and three fixed-frequency microwave reflectometers operated by UCLA, three tangential far-infrared interferometers operated by UCD and an array of magnetic sensing coils external to the plasma. The coils and the UCLA diagnostics have also been utilized in a similar investigation of TAEs. The results are compared with predictions from NOVA-K. Nonlinear three-wave interactions between fishbones, TAEs and CAEs are also studied. These interactions transfer energy in space and time and can significantly influence the effect of the modes on fast ions.   

Full Wave Modeling of Wave -- Plasma Interactions in NSTX.

Wave plasma interactions play an important role in the dynamics of NSTX plasmas in a wide range of frequencies. High harmonic fast waves (HHFW), with frequencies significantly above the fundamental ion cyclotron frequency, are used to heat and drive noninductive currents in NSTX plasmas. Fast ions from neutral beam injection can excite compressional and / or global Alfven eigenmodes (CAE/GAE) with frequencies near the fundamental ion cyclotron frequency. Simulations of power deposition profiles obtained with the full wave code, TORIC, will be compared to the observations from recent HHFW experiments that show that the wave propagation and absorption depend strongly on the antenna phasing and plasma conditions [i]. The issue of mode conversion of the HHFWs to shorter wavelength modes will be revisited. Initial simulations of driven eigenmodes in the CAE / GAE frequency range will also be discussed. [i] See contributed Oral Talk by J. C. Hosea et al this conference   

Electron Bernstein Wave Coupling and Emission Measurements on NSTX

Electron Bernstein waves (EBWs) offer the potential for T$_{e}$(R,t) measurements and local current drive in overdense NSTX plasmas. However, these applications require resilient and efficient coupling between EBWs and electromagnetic waves outside the plasma. Two new remotely steered, obliquely viewing, quad-ridged horn antennas connected to absolutely calibrated dual-channel radiometers have simultaneously measured 8-18 GHz (fundamental) and 18-40 GHz (second and third harmonic) B-X-O emission. Emission data combined with EBW ray-tracing and Thomson scattering T$_{e}$(R,t) data allow the EBW coupling efficiency to be calculated. The EBW coupling efficiency and emission polarization have been mapped as a function of poloidal and toroidal angles for L- and H-mode plasmas and the results are compared to theoretical predictions. Coupling efficiencies of 50-100{\%} have been measured in L-mode discharges while much lower values have been measured in H-mode discharges.   

Relativistic description of the interaction of electron Bernstein waves with electrons

In the electron cyclotron range of frequencies (ECRF), Bernstein waves (EBW) are useful in spherical tokamaks (NSTX) and stellarators for imparting external wave energy and momentum to electrons. Previous theoretical studies on ECRF X and O modes have noted the importance of including relativistic effects in the propagation and damping of these waves. We find that, in order to properly describe EBWs and their interaction with electrons, a relativistic treatment is necessary. Relativistic effects are found to be important for temperatures which are routinely attained in present day laboratory fusion plasmas. A description of the changes in wave propagation characteristics and in the exchange of momentum and energy between EBWs and electrons will be discussed. For steady state, high-$\beta$ operation of NSTX type spherical tokamaks non-inductive current drive is needed. ECRF waves can generate currents either by the Ohkawa or by the Fisch-Boozer schemes. We will elucidate the two current drive schemes as applied to NSTX. Results from two coupled codes, one which solves the fully relativistic plasma dielectric tensor (R2D2) and the other which solves the drift kinetic equation for electrons (DKE), will be presented.   

Effect of Various EFIT NSTX Equilibria on EBW Simulations

In overdense NSTX plasmas the standard O and X mode radiation emitted from the first several electron cyclotron (EC) harmonics cannot be observed. An explanation for any such measured EC emission is the mode conversion of Electron Bernstein Waves (EBW) near the upper hybrid resonance region near the plasma edge. EBW is an electrostatic wave that is not subject to a density limit and is strongly emitted/absorbed near the EC harmonics. Thus EBWE yields information on the local plasma temperature near the EC harmonics. However EBW emission is strongly coupled to the equilibrium fields in NSTX as reconstructed from EFIT. We examine some models of EFIT ranging from the basic model that uses only the external magnetic measurements with simple models for plasma current and pressure, to those models that include effects of edge currents, MSE measurements {\ldots} . and their effect on EBW simulations. The frequency dependence of EBWE provides optimization parameters for proposed EBW heating and current drive.   

Improved detection algorithms for low-frequency MHD activity in NSTX

Unstable MHD modes can pose a serious threat to equilibrium stability and lead to a loss of plasma confinement. The resistive wall mode (RWM) and locked mode are of particular concern on NSTX. To counteract these modes, an active feedback system has been installed on the device. In order to detect and correct these modes in real-time, the device relies on an array of 48 poloidal and radial magnetic field sensors, which are evenly divided between the upper and lower hemispheres. Due to the conditions of extreme heat during device ‘bake-out’ as well as the harsh conditions inside the machine, some magnetic sensors are routinely inoperable during a given shot. Missing sensor data can often lead to a failure to detect a given mode, resulting in the inability to provide feedback to suppress it. The algorithm used to detect these modes was modified to provide an up/down average of the respective sensor sets in the hope that more accurate data could be obtained even in the case of several missing sensors. The results of this adapted algorithm will be discussed, as well as the general characteristics of the mode detection capabilities of the present sensor system.   

Internal magnetic field structure of perturbed tokamak equilibria

The 3D magnetic field structure of perturbed tokamak equilibria is important for understanding phenomena such as error field correction and plasma flow damping. Perturbed quantities such as plasma displacement and magnetic field can be obtained by minimizing the total potential energy numerically using codes such as the DCON ideal stability code. For error field correction applications, initial work is focusing on determining which external magnetic perturbations tend to drive magnetic islands. Jumps on the rational surfaces in the radial derivative of the normal magnetic field perturbation give the singular currents that arise in an ideal plasma to prevent an island from opening[C. N\"{u}hrenberg and A. H. Boozer, Phys. Plasmas \underline {10}, 2840 (2003)]. These currents serve as a measure of the tendency of a perturbation to open a magnetic island. Additionally as a first step benchmarking exercise, we compare the 3D structure of NSTX instabilities as predicted by the DCON and MARS-F codes in the absence of external error fields. Future applications include examining the interplay between external error fields, plasma rotation, and plasma resistivity using the MARS-F code.   

Edge Turbulence in High Density Ohmic Plasmas on NSTX

The Gas Puff Imaging Diagnostic (GPI) consists of a controlled neutral gas puff of deuterium, which increases the brightness of the turbulent emission fluctuations at the plasma edge. A 2-D measurement of the edge turbulence from D$_{\alpha }$ light emission is obtained using a high-speed camera sharing the same viewing area with 13 discrete chords, arranged in a radial-poloidal cross pattern. Light collected from the viewing area 20cm above the outer mid-plane is gathered by an array of 13 discrete optical fibers, each connected to a photomultiplier tube/preamplifier providing time series data at a rate of 500kHz over 120ms. Using the GPI, edge turbulence is studied in high-density, Ohmic, lower single null deuterium plasmas near the Greenwald density. Densities up to 0.8 times the Greenwald density were achieved using a combination of low and high field side gas injection as well as super sonic gas injection for fueling. The data analysis as well as the characteristic features of the turbulence in both high-density and low-density regimes will also be discussed in this poster.   

Studies of T$_{e}$, n$_{e}$, and Particle Flux at the Divertor Surface during NBI Heating and Various Divertor Configurations in H-mode Plasmas in NSTX

Interaction of the plasma with the divertor surfaces has been studied using an array of Langmuir probes and other divertor diagnostics in NSTX NBI heated H-mode plasmas[1]. These studies cover a range of plasma conditions including NBI injection power scans and different divertor configurations. Studies were done mainly with a toroidal field of B$_{t}$ = 0.45 Tesla, plasma current, I$_{p}$, in the range of 0.6 to 1.0 MA, and NBI power in the range 1\underline {$<$} P$_{b}$ \underline {$<$} 6MW. An array of Langmuir probes, flush mounted with the surfaces of the divertor tiles, was used to obtain divertor T$_{e}$, n$_{e}$ and particle flux. The particle flux to the probes increased with P$_{b}$, however, the variation in T$_{e}$ was more complex. Results were obtained for lower single null (LSN) and double null (DN) divertor configurations and these will be compared. Correlations between probe T$_{e}$, n$_{e}$, and particle flux, and divertor heat flux, D$_{\alpha }$ signals and other divertor diagnostics will also be discussed. [1] C. E. Bush, et al., Physics of Plasmas \textbf{10} (2003) 1755. *This work was supported by US DoE contract DE-AC05-00OR22725.   

Characterisation of intermittent turbulence in the edge of the NSTX experiment with the Gas Puff Imaging

The edge turbulence of the National Spherical Torus Experiment (NSTX) fusion device is studied with the Gas Puff Imaging (GPI) diagnostic that measures the visible light emitted from a locally puffed neutral gas. The analysis of the fluctuations is carried out in terms of poloidal and radial propagation velocity, power spectra, frequency-wavenumber spectra and correlation. Moreover the statistical properties of the fluctuations at different time-scales are analysed using the Continuous Wavelet Transform; the probability distribution function is computed for different plasma discharges and for different radial positions. The poloidal and radial characteristic lengths of the intermittent structures at different time scales have been determined by applying the conditional average technique. The dependence of the number of structures on the radial position is studied, and the characterisation of turbulence for ohmic, NBI and RF heated plasma is carried out.   

Particle and Energy Transport in the SOL of DIII-D and NSTX

While intermittent transport is the only SOL radial transport vehicle in L-mode, knowing the relative importance of inter-ELM vs ELM particle flux in H-mode is crucial. Density scans in DIII-D show that ELMs account for $\sim $90{\%} of the wall particle flux at low Greenwald fraction (f$_{g}\sim $0.4), decreasing to $\sim $30{\%} at f$_{g}\sim $1.0. Both intermittent transport and ELMs are comprised of filaments of hot, dense plasma (n$_{e}$ $\sim $ 1x10$^{13}$ cm$^{-3}$, T$_{e}\sim $100 eV) originating at the pedestal and convective in nature, leaving the pedestal region at speeds of $\sim $0.5-1 Km/s and losing heat and particles by parallel transport as they travel through the SOL. The intermittency and ELM heat is quickly lost, resulting in temperature radial decay lengths $\sim $1-2 cm, but the particles are not, resulting in radial density decay lengths $\sim $4-13 cm that increase inversely with SOL colissionality. In DIII-D the intermittency decays in both intensity and frequency in H-mode while it only decays in frequency in NSTX.   

Validation of DEGAS 2 with NSTX Gas Puff Imaging Experiments

The Gas Puff Imaging (GPI) diagnostic operated on NSTX uses a small puff of neutral gas at the edge of the torus to allow plasma fluctuations to be visualized and recorded with a fast camera. We will describe progress made towards validation of the \urllink{DEGAS 2}{http://w3.pppl.gov/degas2} neutral transport code against GPI experiments carried out during the 2004 NSTX run campaign. Rigorous geometric calibration of the GPI camera prior to and during these experiments resolved a previously noted misalignment of the simulated and observed clouds\footnote{D.\ P.\ Stotler et al., Contrib. Plasma Phys. {\bf 44}, 294 (2004).}. A discrepancy in the width of the simulated and observed clouds was eliminated once the nonlinear response of the GPI camera was taken into account. The resulting simulation cloud widths and peak locations then agreed to within the error bars associated with the GPI camera's geometric calibration and the Thomson scattering data used to provide the plasma density and temperature.   

MARFE stability analysis in ELMy H-mode NSTX plasmas

The temperature and density dependence of plasma and atomic processes have been previously identified as the cause of many abrupt transition phenomena in tokamaks, e.g. Multi-faceted Axisymmetric Radiation From the Edge (MARFE). In the National Spherical Torus Experiment (NSTX), edge-localized modes (ELMs) are observed using a fast-framing camera to interact with an inner-wall MARFE, leading to partial burn-through of the MARFE during the ELM cycle[1]. After the ELM pulse, the light pattern subsequently transitions from a helical pattern (a residual from the previous partially burned-through MARFE) to the classic MARFE axisymmetric pattern, with the cycle being repeated at each ELM. We use thermal instability theory to attempt an explanation of the MARFE/ELM dynamics in the NSTX. In particular a single discharge provides many examples of plasma profiles which are thermally stable, and also many profiles which are thermally unstable. Details of the analysis will be presented. \newline [1] R. Maqueda, e.t. al., this conference.   

Mobile dust particle trajectories in NSTX

The transport of dust in plasmas may play a significant role in the performance of next-step fusion devices. [1]. Highly mobile incandescent dust particles are routinely observed on NSTX using fast visible cameras. The dynamics of the dust trajectories can be quite complex exhibiting a large variation in both speed (10-200m/s) and direction. Particles may have constant velocities or exhibit various degrees of acceleration or deceleration. Abrupt reversals in direction are sometimes observed while some of the larger particles are seen to break apart during mid-flight. Measurements of dust trajectories taken simultaneously from two observations points on NSTX with two different fast cameras are being used to derive 3D trajectories of the dust particles which are being compared to the dust transport code DUSTT[2]. Experimental results will be compared to the DUSTT predictions and implications for next-step devices will be discussed. [1] S.I. Krasheninnikov, Y. Tomita, R.D. Smirnov, R. K. Janev, Phys. Plasmas 11 (2004) 3141 [2] A. Yu. Pigarov et al., to be published in Physics of Plasmas (December 2005)   

Testing of a Plasmadynamic Hypervelocity Dust Accelerator

A plasmadynamic accelerator for microparticles (or dust grains) has been designed, built and tested at Los Alamos National laboratory. The dust grains are expected to be accelerated to hypervelocities on the order of 1-30 km/s, depending on their size. The key components of the plasmadynamic accelerator are a coaxial plasma gun operated at 10 kV, a dust dispenser activated by a piezoelectric transducer, and power and remote-control systems. The coaxial plasma gun produces a high density (10$^{18}$ cm$^{-3}$) and low temperature ($\sim$ 1 eV) plasma in deuterium ejected by {\bf J} $\times$ {\bf B} forces, which provides drag on the dust particles in its path. Carbon dust particles will be used, with diameters from 1 to 50 $\mu$m. The plasma parameters produced in the coaxial gun are presented and their implication to dust acceleration is discussed. High speed dust will be injected in the National Spherical Torus Experiment to measure the pitch angle of magnetic field lines.   

The NSTX Plasma Control System Operating Experience

The National Spherical Torus Experiment (NSTX) project has developed extensive operating experience using the Plasma Control System (PCS) with rtEFIT control as well as with the original shape control algorithm. While both algorithms were in use during the past run period, rtEFIT shots became the more frequent choice. Control with rtEFIT allowed access to plasma elongation (kappa) to a record level above 3, improved stability, produced longer pulse lengths and more consistent edge control, and permitted experiments to make shape comparisons with other machines. To support experiments, the PCS was upgraded to control the Resistive Wall Mode (RWM) coils using active feedback control on resistive wall mode sensors, error field sensors, and ohmic and toroidal field coil currents alone or in combination with preprogrammed current control. Use of the RWM coils to correct error fields was successful in delaying the onset of discharge-ending MHD by about 700 ms in the best cases. Planned improvements for the 2007 run include upgrading to a faster computer that will reduce the latency in the control system and provide faster control of vertical instability and RWM instabilities.   

Development and Measurements of the Edge Neutral Density Diagnostic

Edge power balance is not well understood in a spherical torus, where large orbits of lost fast particles on the edge can drastically affect the center of the plasma. In order to gain a more precise understanding of power balance, it is important to have an accurate measure of the profile of neutral particles. The development and preliminary demonstration of the Edge Neutral Density Diagnostic are presented. A 2-D CCD with a maximum frame rate of 500 fps was used to image the outboard edge of the National Spherical Torus Experiment (NSTX), using a D$_{\beta}$ filter to select an atomic transition. The spatial calibration and absolute photometric calibration are described. The image was Abel inverted assuming toroidal symmetry to obtain a radial profile of the emission intensity. A collisional-radiative model was used in conjunction with electron density and temperature obtained from Thomson Scattering to obtain an absolute radial density profile. Future development of this diagnostic is proposed, including using a 12 bit CCD camera to increase sensitivity and time resolution. Comparison is made between beam heated shots and RF shots.   

Ion velocity measurements on NSTX using the SWIFT diagnostic (Shifted Wavelength/lnterference Filter Technique)

Using a split-image interference filter technique, the measurement of ion flows is being extended from individual analog channels (as implemented on Columbia's HBT-EP tokamak) to a 2-D view on NSTX. A beam-splitter duplicates the image, each of which is then filtered with separate interference filters whose passbands have opposite linear slopes. A high frame-rate Photron Ltd. U1tima SE CMOS digital camera views He II line emission at 468.6 nm at the edge of the center stack. A white-plate calibration of two 64 x 64 pixel regions of the detector in necessary to measure relative gain and linearity of the corresponding pixes. A spectral high-resolution wavelength calibration is done for each pixel-pair, probably requiring the use of a tunable, narrow-band, bright light source such as a dye-laser. But once completed, the ion velocity can be calculated very simply from the ratio of the intensities from the two images. From data taken when viewing through a He II interference filter, the light level in helium discharges in NSTX is adequate to make observations at 1,150 frames per second. A preliminary optical system was bench-tested and the lens used can image a 20 x 20 cm region of the plasma with 0.3 mm spatial resolution.   

Measurements of impurity line emission in the EUV on NSTX

The X-ray and Extreme Ultraviolet Spectrometer (XEUS) has been used to monitor the line emission from various impurity ions on NSTX, in particular the K-shell emission of B, C, N, and O. The wavelength range of the instrument has recently been extended from 65 \AA\ to 135 \AA\ allowing measurements of the K-shell emission of the Lyman lines of hydrogenlike Li ions following lithium deposition on the plasma-facing components as well as of the 2-2 transitions of L-shell Fe ions. The latter transitions are of special interest as diagnostics of stellar coronae and other astrophysical plasmas. Our measurements provide information on impurity behavior in NSTX plasmas under various discharge conditions. Moreover, they calibrate the intensities of astrophysical lines in an intermediate density regime not accessible by other laboratory sources.   

Overview of the Alcator C-Mod Research Program

Alcator C-Mod has compared plasma performance with all-metal plasma facing components (PFCs) to PFCs coated with boron. Low-Z coatings reduce metallic impurity influx leading to higher H-mode pedestal pressure that increases temperature and pressure globally. Profile measurements in L-mode Ohmic plasmas suggest that the controlling mechanism regulating pressure in the near scrape-off (SOL) layer is a critical gradient phenomenon, similar to the scaling of edge pressure gradient in H-mode. Cross-field transport decreases markedly as Ip is increased, corresponding to both higher pedestal density and gradient. Disruption mitigation through massive gas-jet impurity puffing has been extended to significantly higher plasma pressures and shorter disruption times. Using mixed gases (He + Ar) yields the fastest mitigation with high radiated energy fraction. Halo currents are reduced by $\sim $50{\%} and divertor surface heating is substantially decreased. Fast framing camera images of intermittent turbulent structures show they travel coherently through the entire SOL. The peak of the radial velocity distribution is about 1{\%} of the ion sound speed.The `chirping' evolution of Alfven cascades during current ramps with intense ICRF heating measured by PCI agrees well with calculations of the NOVA-K code that includes geodesic deformation of the Alfven continuum.   

H-mode pedestal behavior in low collisionality regimes on Alcator C-Mod

H-mode experiments on the Alcator C-Mod tokamak have exploited a wide range of operational parameters, in order to make expanded studies of pedestal structure, edge relaxation mechanisms and the related level of global confinement. Three specific regimes have received increased attention in recent studies: high triangularity ($\delta_{\mbox{lower}}\approx 0.75$), high toroidal field ($B_T\approx 8\mbox{T}$), and reversed $B_T$. Operation in these regimes has yielded pedestal temperatures of nearly 1keV, giving relatively low values for edge collisionality, $\nu^\star<1$. At low $\nu^\star$, H-modes do not favor the steady state enhanced-$\mbox{D}_\alpha$ (EDA) regime, but tend either to exhibit edge-localized modes (ELMs) or to be transiently ELM-free. In these regimes we study the temporal behavior of pedestal density and temperature profiles (especially their widths and gradients) and compare to that in the well-characterized EDA. Of particular interest is whether a ballooning-like scaling for pressure gradient observed in the EDA H-mode edge persists in less collisional pedestals.   

Parametric Survey of ITB Plasmas in Alcator C-Mod

Internal transport barrier plasmas have been created in Alcator C-Mod using off-axis ICRF heating in target plasmas using ICRF frequencies of 50, 70 and 80 MHz at magnetic fields ranging from 2.8 T to 6.4 T and with plasma currents between 0.7 and 1.2 MA. These ITBs are marked by highly peaked density and pressure profiles. They rely on a reduction of particle and thermal flux in the barrier region which then allows the neoclassical pinch to peak the central density while maintaining the central temperature. Although these ITBs are obtained frequently, there remains uncertainty in what qualities of the target plasma parameters will result in obtaining reliable and predictable ITB formation. The object of this study is to examine the characteristics of these target plasmas, especially with respect to input power, density, impurity content, and shaping. The results from the parametric examination will be presented along with results of recent ITB C-Mod operation.   

Inter-Machine Comparison of Intrinsic Toroidal Rotation

Toroidal plasma rotation in discharges with no external momentum input has been observed on a number of tokamaks in a variety of operating regimes. On all the machines included in this study (Alcator C-Mod, DIII-D, JET, Tore Supra, TCV {\&} JT-60U) the intrinsic rotation in high confinement regimes is in the co-current direction, increases with plasma stored energy and decreases with increasing plasma current. Although all machines show positive correlation between stored energy and intrinsic rotation, the constant of proportionality for each machine differ considerably. These constants of proportionality can be brought into much better agreement by recasting the relationship in the non-dimensional form: M$_{A} \propto $ $\beta _{N}$ (some other mach numbers also work reasonably well). Extrapolation of this scaling to ITER relevant $\beta _{N}$ suggests that toroidal rotation on ITER may be much higher than predicted based on neutral beam torque alone. For an ITER discharge with $\beta _{N}$ = 2.6, the scaling predicts an intrinsic rotation of approximately M$_{A}$ = 0.02 (i.e. 2{\%} of the Alfven Speed), which may be enough to suppress resistive walls modes.   

Application of a PILATUS II Detector to an X-Ray Imaging Crystal Spectrometer for fast Measurement of Ti and Rotation-Velocity Profiles on Alcator C-Mod

A high resolution imaging x-ray crystal spectrometer (XICS) is being developed for Doppler measurement of radial profiles of ion temperature, T$_{i}$, and rotation velocity on Alcator C-mod. The XICS consists of a spherically bent crystal and a 2D imaging x-ray detector, and provides x-ray spectra from highly charged ions from multiple plasma sightlines. The proof of principle of the XICS was demonstrated by measurement of Ar XVII K$\alpha $ spectral images from +/- 8 cm of the plasma height in Alcator C-Mod and +/- 40 cm in NSTX. However, the time resolution was limited to values $>$100 ms by the $\sim $400 kHz global count-rate limit of the available 2D gas detector. A silicon pixel array detector, PILATUS II, with a count-rate capability of 1 MHz PER PIXEL, was tested on C-Mod by recording spectra of ArXVII at 3.1 keV, and should enable XICS time resolution $<$ 10 ms. The detector test results and C-Mod XICS design and expected performance will be presented.   

Edge Ion Velocity Measurements with a Novel Doppler Spectrometer at the Alcator C-Mod Tokamak

A high throughput, f/\# $\sim$3.1, transmission grating Doppler spectrometer for visible light (3500-6700 \AA) is currently measuring ion or neutral velocities and temperatures at the Alcator C-Mod tokamak. The ion velocities are measured through the Doppler shift of impurities that are present in the plasma. A line width of as small as 0.4 \AA (velocity sensitivity of $\sim$10$^5$ cm/s) has been measured using calibration lamps. The spectrometer is fiber optically coupled and has access to toroidal and poloidal views. A spectral survey has been done with various views of the C-Mod plasma identifying various intrinsic impurities. The first Doppler measurements of B II were recorded with $\sim$15 ms per frame. Additional Doppler velocity and temperature measurements in both poloidal and toroidal directions for some of the brighter impurities (e.g. He II and N III), will be given. This work is supported was performed under the auspices of the DoE by UC LLNL under contract W-7405-ENG-48 and also under DoE Coop. Agreement DE-FC02-99ER54512.   

Lower Hybrid Coupling Studies on Alcator C-Mod

The Alcator C-Mod Lower Hybrid Current Drive experiment launches RF waves at 4.6 GHz into the plasma via 4 rows of 24 phased waveguides. 156 directional couplers in the launcher system measure forward and reflected power in the waveguides. Langmuir probes mounted to the front of the grill monitor density at the plasma edge and act as RF probes for the observation of parametric decay. Measurements of the coupling of lower hybrid waves have been performed at power levels up to 600 kW. Edge density, launched n$_{\vert \vert }$ spectrum, and forward power have been scanned. Experimentally observed coupling results will be compared to theoretical predictions from the Brambilla code(M. Brambilla. \textit{Nuc. Fus.}, \textbf{16}:47-54, 1976.).   

Measurements and Modeling of X-Ray and ECE Spectra During C-Mod Lower Hybrid Current Drive Experiments

Lower Hybrid Current Drive (LHCD) has been demonstrated on Alcator C-Mod. The lower hybrid waves generate non-thermal electrons, which can be detected through hard x-ray Bremsstrahlung emission as well as relativistically downshifted electron cyclotron emission (ECE). C-Mod has a hard x-ray diagnostic with 32 chords, designed to detect Bremsstrahlung radiation from LH-driven nonthermal electrons\footnote{J.E. Liptac, to be published in RSI}, as well as several outboard midplane ECE diagnostics, normally used to measure electron temperature in Maxwellian plasmas. CQL3D/GENRAY\footnote{R. W. Harvey and M. G. McCoy, General Atomics Report GA-A20978, www.compxco.com} is a modeling package that employs a 3-D Fokker-Planck solver to compute steady-state distribution functions for a given LH N$_{\vert \vert }$ spectrum and plasma. It also can perform self-consistent synthetic diagnostic calculations. LH phase and power scans have been carried out. Experimental measurements will be compared with synthetic diagnostic modeling to see how changes in phasing and power affect current profile control and to benchmark CQL3D in ITER-relevant regimes. This work is supported by the US DOE grants \#DE-FC02-99ER54512 and \#DE-AC02-76CH03073.   

Recent ICRF Results in Alcator C-Mod

Up to 6 MW of ICRF power (8 MW source) is available in Alcator C-Mod to heat the plasma in the D(H) or D(3He) minority regimes. The compatibility of high RF power with all metal plasma facing components (Molybdenum tiles in C-Mod) is a critical issue for present and future tokamaks (ie, ITER). We have achieved world record volume averaged plasma pressures (1.8 atm) and record stored energy (250 kJ) in C-Mod with 5.25 MW of net injected ICRF power (5.5 MWm-3) in a freshly boronized machine. The key to high plasma performance was determined to be the control of impurities, particularly Mo.\footnote{B. Lipschultz et al, Phys. Plasmas 13 (2006) 056117} Since the boronization eroded more quickly for ICRF heated (as opposed to Ohmic) H-modes, enhanced sputtering due to RF sheaths in specific locations is suspect. Experimental results will be presented which explore potential source(s) of Mo impurities due to rapid erosion of the boron layer with intense ICRF power. The main candidate for impurity generation with an impact on central plasma performance is the divertor tile region that is magnetically connected to an active ICRF antenna which then could impact sputtering and erosion of boron by RF sheaths.   

Mode conversion current drive studies on Alcator C-Mod

Current drive from mode converted waves in the ion cyclotron range of frequencies is being investigated on the Alcator C-Mod tokamak. In experiments where the mode conversion layer was swept through the q = 1 surface, significant variations in the sawtooth period were obtained. The results are consistent with localized mode conversion current drive and are in qualitative agreement with full wave simulations using the TORIC code [M. Brambilla, \textit{Plasma. Phys. Cont. Fusion} 41 (1999)]. Measurements of net current drive based on loop voltage differences have been attempted during the flattop and ramp-up phases of C-Mod discharges. The experimental results and analysis using TORIC and TRANSP simulations will be presented. Numerical calculations have also been improved by coupling TORIC with the Fokker-Planck code DKE [J. Decker, PhD Thesis, MIT], which allows for a better treatment of magnetic trapping effects when calculating the local current drive efficiency.   

Fast Ferrite ICRF Matching System in Alcator C-Mod

The loading of an ICRF antenna can vary significantly in a plasma discharge due to L-H mode transitions, the occurrence of edge localized modes and changes in plasma shape. The matching systems currently used in Alcator C-Mod have fixed positions during a discharge, which are often inadequate to accommodate all the loading variations. To avoid excessive VSWR to the transmitters, the RF power has to be shut off when mis-match happens. This imposes limitations on the range of plasma parameters that the antennas can couple effectively. In order to make ICRF operation in C-Mod more robust and flexible, a double-stub fast ferrite matching system has recently been developed. The magnetic permeability of the ferrite material can be changed by varying a magnetic field, generated by coils. As a result, the equivalent electrical length of the stubs can be controlled dynamically in a plasma discharge. The coil current control is carried out through a digital feedback system, which computes the outputs from real-time impedance signals. Simulations have shown that this fast ferrite system will allow a good loading match under most plasma conditions in Alcator C-Mod. Bench test results of the system will be presented.   

Real-time sensing and gas jet mitigation of VDEs on Alcator C-Mod

Experiments have been carried out in Alcator C-Mod to test the effectiveness of gas jet disruption mitigation of VDEs with real-time detection and triggering by the C-Mod digital plasma control system (DPCS). The DPCS continuously computes the error in the plasma vertical position from the magnetics diagnostics. When this error exceeds an adjustable preset value, the DPCS triggers the gas jet valve (with a negligible latency time). The high-pressure gas (argon) only takes a few milliseconds to enter the vacuum chamber and begin affecting the plasma, but this is comparable to the VDE timescale on C-Mod. Nevertheless, gas jet injection reduced the halo current, increased the radiated power fraction, and reduced the heating of the divertor compared to unmitigated disruptions, but not quite as well as in earlier mitigation experiments with vertically stable plasmas. Presumably a faster overall response time would be beneficial, and several ways to achieve this will also be discussed.   

Study of the error associated with background variations during active CXRS analysis

Analysis of active charge-exchange spectroscopy requires a knowledge of background emission from times for which the beam source is inactive. The intensity of this background emission can vary during a shot, even for 'constant' plasma conditions, leading to uncertainties in its subtraction during beam-on periods. We present a study wherein we use an incorrect background emission during analysis to see the effect on derived velocity and temperature. Assuming an error of up to 10{\%} in the background total counts and a signal to background ratio of 1.5, we obtain up to 4 km/s error in ion velocity and 20 eV in temperature for a test signal that was originally 30 km/s and 400 eV. The error is consistent with expectations due to the relative positions of the two line shapes. As expected, the error (in both temperature and velocity) decreases with increased signal to background ratio. These errors will combine with the uncertainty introduced by photon noise.   

Study of B$^{+1}$, B$^{+4}$ and B$^{+5}$ impurity poloidal rotation in Alcator C-Mod plasmas for 0.75 $< \quad \rho \quad <$ 1.0.

Poloidal and toroidal velocities of B$^{+1}$, B$^{+4}$ and B$^{+5}$ impurity ions are measured in Alcator C-Mod tokamak plasmas using charge exchange recombination spectroscopy (CXRS) for B$^{+5}$ and ambient emission for B$^{+1}$ and B$^{+4}$. The set of 25 poloidal optical channels, 10 toroidal optical channels, modulated diagnostic neutral beam and fast Roper CCD camera allow 2mm poloidal spatial resolution in the region 0.75 $<$ r/a $<$ 1.0 and 10 mm toroidal spatial resolution with 13 msec temporal resolution at all times during the 1.5 sec plasma pulse. The variation in the poloidal rotation as the plasma transitions from ohmic to L- to H- mode will be described. Implications for E$_{r}$ will be discussed. Data for ELMing and EDA H-modes will also be presented. The emphasis of this work is on comparing the poloidal rotation of B$^{+1}$, B$^{+4}$ and B$^{+5}$ impurities, cataloging the effects of difference plasma modes and finally on attempting to understand the poloidal rotation based on neoclassical theories.   

Upgrade of the Edge Charge Exchange Diagnostic on Alcator C-Mod

The current edge Charge Exchange Spectroscopy system is being upgraded to include both a beam viewing and a background viewing toroidal periscope. The beam viewing periscope will be focused on the center of the DNB and will cover a 4cm radial region at the edge of the plasma starting a few centimeters in from and extending up to 1centimeter outside of the last closed flux surface. The background periscope will view the same radial region but will be displaced toroidally by 36 degrees. Each periscope has 20 chordal views with a radial resolution of 2.5-3mm. The presence of a background periscope obviates the need for a chopped DNB by providing time synchronized background B$^{+4}$ spectral data that can be subtracted directly from the active beam-derived B$^{+4}$ line-shapes. This system has been designed to work in conjunction with the current poloidal Charge Exchange periscope which has 25 fibers focused in the same region with equivalent radial resolution. The new toroidal system will enable concurrent measurements of the poloidal and toroidal velocity as well as the temperature and density of the B$^{+5}$ ions in the edge pedestal region; a measurement that currently does not exist on C-Mod. This information will then be used to calculate radial electric field profiles and study edge physics phenomena.   

Quantitative estimation of the beam fast ion effects on MSE measurement in Alcator C-Mod

The MSE diagnostic in Alcator C-Mod observes anomalous behavior during beam-into-gas calibrations and discrepancies between measured pitch angles versus pitch angles calculated by EFIT. We conjecture that these phenomena are caused by beam neutrals that ionize following collisions with the torus gas or plasma and then re-charge exchange at a random gyro-angle. Motional Stark emission from such `secondary' beam neutrals will have a Doppler shift and a polarization direction that differs from that emitted by primary beam neutrals. This effect is particularly large in Alcator C-Mod because the diagnostic beam injects perpendicularly and thus the residence time of the beam ions in the MSE field of view is limited only by the grad-B drift. We present calculations of this effect on the MSE spectrum and polarization angle and evaluate the consequences of rotating the beam several degrees in the toroidal direction.   

Results from the Alcator C-Mod Polarimeter Prototype and Plans for an FIR System

A single chord CO$_{2}$ polarimeter channel has been installed on the existing Two-Color Interferometer (TCI) on Alcator C-Mod. The measured rotation ranged from 0.1-0.3 degrees in magnitude depending on plasma density, current, and configuration.~ Empirical results are comparable to predicted values from model density profiles and Efit reconstructions.~ Experimental techniques developed to optimize signal quality and minimize drifts will be discussed.~ This work is in support of an FIR system with poloidally viewing chords to be installed on C-Mod in FY08.~ Some aspects of the FIR design will also be discussed.   

Two-dimensional radiated power diagnostics for Alcator C-Mod

Upgrades to Alcator C-Mod's radiated power diagnostics are presented. These tools, currently under construction, will be deployed in FY2007 to study the radial and poloidal distribution of photon emissivity in a variety of C-Mod plasmas. Commonly utilized AXUV diodes arrays will view from the inner to outer limiter on five different horizontal planes using a total of 132 channels. In order to study radiation in the divertor/x-point region, data from toroidally and poloidally viewing arrays will be tomographically inverted. To compensate for the AXUV's $>$ 50 {\%} drop in sensitivity for $\sim $10 eV photons, 50 Lyman-$\alpha $ filtered diodes and 40 unfiltered diodes will view the divertor and be independently inverted. Signal to noise modeling for typical plasmas is presented and its implications for the inversion are discussed. Estimations of expected flux surface asymmetries in impurity density, and thus emissivity, are summarized using both existing theory and previous Alcator C-Mod experimental results.   

Results from Active Excitation of Toroidal Alfv\'{e}n Eigenmodes in Alcator C-Mod

Toroidal Alfv\'{e}n Eigenmodes (TAEs) are weakly damped MHD waves in tokamak plasmas. Interaction with fast particles such as fusion-born alphas can overcome the damping and lead to the spontaneous appearance of unstable TAEs. The Active MHD diagnostic on Alcator C-Mod is used to investigate the relationship between the TAE margin to instability and controllable plasma parameters. The diagnostic identifies the frequency response of the plasma in the TAE frequency range, $f_{TAE}=v_A/4\pi qR$. It perturbs the magnetic field with two antennas and detects the plasma response with an array of pick-up coils. The total damping rate and toroidal mode number of the TAE are extracted from a parametric model fitted to the frequency response. Particular attention is paid to signal processing techniques for minimizing uncertainty. The relationship between the TAE damping rate and ICRF heating is investigated with the aid of a neutral particle analyzer to quantify the fast ion population. Other parameters investigated for their effect on damping rate are collisionality, normalized ion gyro-radius, beta, triangularity, and the direction of the $\nabla B$ drift with respect to the x-point in diverted plasmas.   

Alfv\'{e}n eigenmode activity during the sawtooth phase in Alcator C-Mod

Recent current ramp experiments in Alcator C-mod have utilized early ICRF heating (P$_{RF}$ $\sim$ 3 MW) and low densities (n$_{e}$l $\sim $ 0.6x10$^{20}$ m$^{-2})$ to destabilize core localized Alfv\'{e}n eigenmodes [1]. This results in creation of Alfv\'{e}n cascades, chirping modes highly sensitive to small changes in the minimum of the q-profile [2,3]. This great sensitivity makes Alfv\'{e}n cascades particularly useful as a diagnostic and MHD monitor [3]. The phase contrast imaging (PCI) system has observed modes similar to the cascades, but preceding sawteeth in sufficiently low density plasmas. The hybrid kinetic-MHD code NOVA-K [4] has provided results showing the theoretical existence of q=1 cascade modes under plasma conditions taken from actual discharges. Implications for post- crash current profiles will be presented. \newline \newline [1] M. Porkolab \textit{et al}., IEEE Trans. on Plasma Science \textbf{34}, 229 (2006). \newline [2] H. L. Berk \textit{et al.,} Physical Review Letters \textbf {87}, 185002 (2001). \newline [3] S.E. Sharapov \textit{et al}., Physics Letters A \textbf {289}, 127 (2001). \newline [4] C.Z. Cheng and M.S. Chance, J. of Comp. Physics \textbf{71}, 124 (1987).   

Turbulence Measurements with the Upgraded Phase Contrast Imaging Diagnostic in Alcator C-Mod

The Phase Contrast Imaging diagnostic (PCI) in Alcator C-Mod has been upgraded to improve the frequency and wavenumber sensitivity and provide localized measurements along the beam. Optional optics configurations are available for optimizing turbulence measurements in different wavenumber ranges. With the low-k setup (6 cm$^{-1})$ and the widened beam, new features of the Quasi-Coherent (QC) modes were observed. The coherence of the QC modes is reduced toward the X-point. Furthermore, the spatial location of the coherence reduction moves with the X-point. With the medium-k setup (12 cm$^{-1})$, three coupled (f$_{3}\sim $f$_{2}$+f$_{1}$ and k$_{R3}\sim $k$_{R2}$+k$_{R1})$ semi-coherent modes have been detected. The initial analysis of these two measurements will be presented. For the high-k setup (up to 40 cm$^{-1})$, initial tests with sound burst transducers have shown capability to measure modes with wavenumbers up to 40 cm$^{-1}$. Currently, the PCI is operating with the high-k setup to search for the short wavelength turbulence in the ETG range and results will be presented as available.   

Reflectometry Analysis of Density Fluctuations in Alcator C-Mod

Electron density fluctuation measurements on Alcator C-Mod are presented for various density profiles using O-mode reflectometry. Seven discrete frequency channels ranging from 50GHz to 140GHz are radiated into the plasma from the low field side at the midplane of the tokamak, probing the cutoff density layers (n$_{e}$=3x10$^{19}$m$^{-3}$ to n$_{e}$=2.4x10$^{20}$m$^{-3}$ respectively) for electron density fluctuations. The system's output is provided by an I/Q detector, enabling the comparison of the measured signals with results from reflectometry modeling of density fluctuations in C-Mod plasmas. Turbulent phenomena are highlighted at different radial locations of the plasma and compared with other diagnostics such as Phase Contrast Imaging and magnetic pick-up coils. Reflectometry measurements for low density Internal Transport Barrier (ITB) plasmas are presented where the signal traverses the pedestal and probes the foot of the ITB.   

ECE Temperature Fluctuations associated with EDA H-Mode discharges in Alcator C-Mod

Alcator C-Mod exhibits an ELM-free H-mode with ``$enhanced\, D\, alpha$'' emission accompanied by a quasi-coherent mode (QCM) edge relaxation mechanism. This steady state H-mode lowers the peak heat load to the diverters which is advantageous for reactor operations. A high-resolution heterodyne electron-cyclotron-emission (ECE) radiometer with 32 channels ($\Delta R\sim7mm$) and a bandwidth up to $1MHz$ covering the full radius of C-Mod has observed spatial resolved temperature fluctuations that are highly correlated with the edge QCM mode. The QCM mode is also directly observed by the edge ECE channels though the changes in optical depth due to the large density fluctuations in the QCM ($\sim30\%$). Details of these measurements will be presented in this poster.   

Langmuir-Mach Probe Geometries for Measurement of Total Flow Vector on Alcator C-Mod

Strong plasma flows are observed in the scrape-off layer in fusion devices. Recent data from Alcator C-Mod indicate that plasma flow along field lines is particularly strong on the high-field side, which appears to be driven by particle transport asymmetries [1]. However, owing to uncertainties in probe-sheath derived measurements of the plasma potential profile, the contribution of cross-field flows to the total flow vector has been difficult to unfold. For direct measurement of parallel and cross-field flows, we have begun to explore the use of a ``Gundestrup''-type probe [2], adapted for the high-heat flux environment of C-Mod. Two different variations on a four-faceted tungsten electrode geometry have recently been operated, with facets tilted by 30 or 45 degrees with respect to magnetic flux surfaces in order to minimize surface heating. We report parallel and cross-field flow measurements, compared to standard Mach probes measurements and ExB flow estimates. \newline [1] B. LaBombard et al., Nucl. Fusion \textbf{44} (2004) 1047. \newline [2] MacLatchy, et al., Rev. Sci. Instrum. \textbf{63} (1992) 3923.   

Fast-switching Langmuir probe bias electronics for Alcator C-Mod

In order to resolve fast-changes in edge plasma density, temperature and plasma potentials with a single Langmuir electrode, the I-V characteristic must be generated and sampled at high frequency. To this end, a custom-designed package of fast-switching electronics is being assembled for use in Alcator C-Mod, employing three principal components: (1) a master TTL ‘waveform generator’, (2) fast-switching MOSFET drive circuits ($\sim$ 30 ns rise time), and (3) current-voltage monitor circuits. Three voltage bias states are capacitively coupled to up to 6 Langmuir probes in a sequence that samples portions of the I-V characteristic: ‘ion saturation ($\ge$ -234V), ‘electron collection’ ($\le$ +64V), and ‘near floating’ ($\sim$ 0V). Up to 2 amps of peak current can be supplied to each probe with waveform durations of $\sim$ 2 seconds. Resultant I-V characteristics are digitally sampled ($\le$ 50MHz) by cPCI transient recorders. Three additional TTL waveforms, synchronized to the data-sampling times of the different bias states, are also supplied. These may be used to report plasma conditions in real-time using a ‘mirror Langmuir probe’ technique [1].\par {\noindent [1] B. LaBombard and L. Lyons, manuscript in preparation.}   

Magnetic topology effects on ion flows

We consider the effect of magnetic topology on the Pfirsch-Schl\"uter flows inside the separatrix of a tokamak for reversal in the direction of (i) the plasma current or poloidal magnetic field, (ii) the toroidal magnetic field, or (iii) both; and for (iv) a switch between lower and upper X-point operation. We find that the results of the magnetic topology changes on the flows observed in Alcator C-Mod are consistent with the predictions of neoclassical theory when both the up-down symmetric and up-down asymmetric drives for toroidal rotation are retained.   

Progress on Stereoscopic Imaging of Ablating Lithium Pellets in Alcator C-mod

Ultra high speed imaging (max frame rate = 500 kHz) of injected Li pellets into Alcator C-Mod during previous campaigns identified the presence of poloidally moving striations in the pellet's ablation cloud. The behavior of these striations is remarkably different in L-mode and H-mode plasmas. In L-mode plasmas, the velocity of the striations tend to change direction on a length scale of the order of tens of gyro-radii, ranging from +/- 3000 m/s. In H-mode plasmas the striations currently have only been observed to move in the direction of the ion diamagnetic drift, with velocities that range up to 4000 m/s. During the 2005-2006 campaign a stereoscopic imaging system was installed and tested on Alcator C-mod to observe the ultra-short time scale ($<$ 2 $\mu $s) ablation dynamics in three dimensions, to determine the cause for both the formation and evolution of these striations. Presented here are experimental observations of the ablation process and possible explanations in terms of ablation dynamics and radial electric fields in the tokamak plasma.   

Dust Measuring Diagnostics on Alcator C-Mod

We present initial results from a laser scattering diagnostic viewing scattered light from large particles present in the C-Mod vessel. The light source consists of pulsed Nd:YAG lasers principally used for Thomson scattering. In order to view the YAG laser line (1064 nm) collection fibers and detectors were installed alongside the existing Thomson scattering diagnostic. Results from this diagnostic are presented with comparisons to data collected from similar diagnostics on other machines. Additionally, we present designs for a new Dust Scattering Diagnostic to view the dynamics of injected dust in the scrape off layer. Boron dust particles will be injected into the vessel via a gas puff during a shot. The particles will be illuminated by 532 nm laser light and the images will be captured with a fast-framing CCD camera. This diagnostic is scheduled to be operational in the next campaign.   

Alcasim Axisymmetric Simulation Code for Noise and Stability Analysis on Alcator C-Mod

Tokamak control is an active research area as the next generation reactors will need active suppression of local and global instabilities within actuator constraints, as well as adaptive handling of off-normal scenarios. The complexity of the control issues and the risk of direct testing on the hardware call for accurate and fast simulators targeting the control problem at hand. Alcasim is a simulator developed for C-Mod with the specific purpose of studying the axisymmetric stability of elongated plasmas in the presence of power supply saturation. The code is highly optimized so that Alcasim is suitable for parametric investigation of the controller gains. Alcasim has been used to study the vertical oscillations seen early in C-Mod discharges when plasmas are fairly elongated. Analysis of real data supported by extensive simulation show that plasmas occasionally travel through regions of instability for which the controller gains need further optimization. Work is ongoing for a systematic and efficient approach to optimal gain calculation, to be possibly employed in real time gain scheduling. USDoE award DE-FC02-99ER545512.   

Error field and locked mode threshold studies on Alcator C-Mod

Non-axisymmetric fields of order $\tilde B/B\sim10^{-4}$ are observed to destabilize non-rotating tearing modes (locked modes) in present-day tokamaks. Prediction of the error-field sensitivity and determination of the requirements for corrective measures is an important issue for ITER. At Alcator C-Mod a set of external coils (A-coils), producing predominantly n=1 fields with adjustable toroidal phase and poloidal ($m$)spectra, is routinely used to compensate the intrinsic error fields, resulting in substantial expansion of the accessible operating space. Recent experiments have investigated the scaling of the locked mode threshold, with particular attention to the dependence on toroidal field and verification of non-dimensional scaling constraints, in order to provide an extrapolation to ITER. Additional experiments were directed to improving characterization of the intrinsic error field on C-Mod.   

Warp simulations of non-neutral plasmas in traps with nonuniform magnetic fields

The three-dimensional particle-in-cell simulation Warp [1] is used to study a variety of nonneutral plasma systems. The equilibrium density in a trap with a spatially varying axial field is studied. Numerical results indicate that, as anticipated, equilibrium is reached along field lines. Plasma confinement and equilibrium when combined multipole fields added to a conventional Penning-Malmberg trap, as is being done in antihydrogen experiments, is investigated---some preliminary results indicate that in some cases high density regions are produced which may be caused by potential variations along field lines. Such variation is expected and demonstrated in the case of a trap with a mirror [2]. Numerical and computational issues associated with this modeling are discussed. [1] D. P. Grote, A. Friedman, I. Haber, J-L Vay, 2004 ECRIS Workshop, AIP Conf. Proc. \textbf{749}, 55, (2005). [2] K. Gomberoff, J. Fajans, J. Wurtele, A. Friedman, D. P. Grote and R. H. Cohen, ``Simulation studies of non-neutral plasma equilibria in an electrostatic trap with a magnetic mirror,'' in preparation (2006).   

Wide-range control of ion temperature in a linear rf ion trap

Ion trap is a device which confines charged particles using an electromagnetic field. The characteristics of the confinement by ion trap are long confinement time and weak interaction with the surrounding environment. Thus ion trap is suitable for the research of statistical character of ion cloud. The aim of our study is to clarify the dynamics of the phase transition between liquid and solid state of ion cloud. The technique of accurate ion temperature control is important for this purpose. As the first step of our ion trap experiment, we tried to expand the controllability of ion temperature in a linear rf ion trap using buffer-gas cooling, laser cooling, and rf heating. We developed a linear ion trap and tunable diode laser systems. The laser systems were used for both of laser cooling and the laser induced fluorescence (LIF) measurements of Ca ion. The ion temperature was estimated from Doppler LIF spectrum. The ion temperature could be controlled from several thousand Kelvin to less than room temperature using buffer gas cooling and laser cooling. We are trying to decrease the minimum temperature by decreasing rf heating effect. We will discuss about accurate temperature control around the phase transition between liquid and solid state of ion cloud.   

Space Charge Neutralization and Plasma Compression in the Periodically Oscillating Plasma Sphere (POPS)

Theoretical works by Barnes and Nebel [R. A. Nebel, D. C. Barnes, Fusion Technology {\bf 38}, 28 (1998)] [D. C. Barnes, R. A. Nebel, Phys. Plasmas {\bf 5}, 2498 (1998)] have suggested that a tiny oscillating ion cloud may undergo a self-similar collapse in a harmonic oscillator potential formed by a uniform electron background. One issue for this concept is how much plasma compression can be achieved by the POPS oscillations. Recent work has shown that by properly programming the distribution function of the injected electrons it is possible to significantly improve the space charge neutralization and the plasma compression. This paper extends that previous work in a systematic fashion by developing a formalism that determines the required velocity distribution of the injected electrons so space charge neutralization can be achieved. This formalism is then included as a boundary condition in a griddles particle code. Results indicate that although the formalism works well during the early phases of compression, when the compression gets large the solution bifurcates and becomes unphysical. Two-dimensional simulations with the PIC code CELESTE2D [G. Lapenta, Phys. Plasmas {\bf 6}, 1442 (1996)] will be presented.   

Periodically Oscillating Plasma Sphere (POPS) and Neutron Source Development at LANL

The inertial electrostatic confinement (IEC) device provides a favorable development path for fusion applications. It is a compact device that is easy to construct and operate. It is also straightforward to accelerate the ions to fusion relevant energy. Though the existing IEC device typically operates with relatively low efficiency, there are several promising near-term fusion applications such as nuclear assay. We will discuss about the on-going LANL research efforts using IEC based neutron generators for nuclear assay applications. Neutron yields of $\sim $10$^{6}$ neutrons/s per kw input power have been demonstrated using D-D fuels. An inductively coupled plasma source has been installed to enhance the neutron yield. Separately, we will discuss about the current status of periodically oscillating plasma sphere (POPS) experiment. The POPS research is aimed at achieving fusion energy production based on the novel plasma heating concept proposed by Nebel and Barnes$^{1}$. POPS oscillation has been verified experimentally and an electron beam probe is being developed in order to measure the POPS compression and heating. 1. R. A. Nebel, D. C. Barnes, Fusion Technology 38, 28 (1998).   

Modeling the FTICR-MS signal of a $^{7}$Be Ion Plasma using a 2D PIC code

Beryllium-7 ($^{7}$Be) decays only by electron capture into Lithium-7 ($^{7}$Li) with a half life of 53 days. As a result, changing its electronic structure will affect its decay rate. We desire to study the effect of ionization on its decay rate. We will do this by trapping a $^{7}$Be ion plasma in a Malmberg- Penning Trap and measuring its and $^{7}$Li's concentration as a function of time by using Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS). We use this ratio as a function of time to directly measure the decay rate of the confined ion plasma rather than using gamma detection. Since $^ {7}$Be and $^{7}$Li ion cyclotron frequencies are very close and these signals are not well understood in the plasma regime, we will model them in an electrostatic particle-in-cell (PIC) code. This simulation will be in two dimensions on a rectangular grid with a circular boundary. The progress of this investigation will be presented.   

A Modified MeVVA Ion Source for a Malmberg-Penning Trap

We have designed a new type of plasma gun ion source for a Malmberg-Penning trap based on Metal Vapor Vacuum Arc (MeVVA) ion source designs. Our primary intent with this MeVVA-type source is to create a confinable beryllium-7 (\textsuperscript{7}Be) plasma. The radioactivity of \textsuperscript{7}Be requires us to replace the sample inside the ion source on a regular basis. Our design makes it possible to easily remove the cathode of the ion source from an ultra-high vacuum trap and exchange \textsuperscript{7}Be samples while only needing to repressurize a small chamber rather than the entire trap. This design has an added benefit of being capable of generating plasmas from a wide variety of metals by simply exchanging the source target in the removable cathode. Because of this wide compatibility, we will be able to use our trap for studying any number of different plasmas, including other radioactive types.   

A numerical investigation of Landau damping by bounce-resonant particles

There is very little damping by particles which are velocity-resonant with an electrostatic wave in a non-neutral plasma because they remain in resonance only until they bounce off the end. The time spent in turning around dephases the particle from the wave. Particles which have a higher midplane velocity which compensates for the turn-around time remain coherent with the wave in an average sense and cause damping of the wave$^{1}$. Previous attempts to measure Landau damping in non-neutral plasma Particle-in-Cell (PIC) codes have been unsuccessful because the mode amplitudes had to be too large for the damping to be linear. This was because of the density fluctuations caused by the limited number of particles. With increases in processing speed and memory available, we have been able to simulate modes small enough that the damping is linear for a substantial period of time. Velocity space diagnostics clearly show the damping is due to the bounce-resonant particles. Cutting off the distribution function in velocity space shows the expected effect of first increasing the damping, followed by a decrease. The mode frequency also shifts by more than expected as the cutoff passes through the bounce-resonant velocity. \\ $^{1}$M. E. Koepke, Bull. Am. Phys. Soc., {\bf 49}, 40 (2004).   

Instability of the $m=1$ self-shielded mode in finite-length nonneutral plasmas

The $m=1$ self-shielded mode in a Malmberg-Penning trap is stable for a hollowed density profile in the infinite-length theory, but has been observed to be unstable in experiments. Earlier work by us and others showed theory and simulations to be a persistent factor of about 2 or more lower than experiment for the growth rate when applied to a single experimental point from measurements of Kabantsev and Driscoll (UCSD). Recently Shi, Chang, and Mitchell (University of Delaware) have measured the growth rates of the mode for a series of hollowed plasmas. We have done drift-kinetic particle-in- cell simulations of several of these experimental equilibria and have found the simulated growth rates also to be lower than experiment. We describe numerical experiments to vary the shape of the plasma ends, to vary the velocity distribution as it might result from the hollowing procedure, and to introduce resistive energy losses from the sectored confining ring to explain the discrepancy.   

Effect of Realistic End Boundaries on Particle Dynamics in Asymmetry-Induced Transport

We have added realistic end boundaries to a simple computer code\footnote{D.L. Eggleston, Bull. Am. Phys. Soc. {\bf 50}, 125 (2005).} developed as an aid to understanding asymmetry-induced transport. For the typical experimental case of a standing wave asymmetry, the code reveals dynamical behaviors not included in the analytical theory\footnote{D.L. Eggleston and T.M. O'Neil, Phys. Plasmas {\bf 6}, 2699 (1999).} of this transport. The resonances associated with the two constituent helical waves typically overlap and produce a region of stochastic motion. In addition, particles near the radius where the asymmetry frequency matches the $E\times B$ rotation frequency $\omega_R$ can be trapped in the potential of the applied asymmetry and confined to one end of the device. Both behaviors are associated with large radial excursions and mainly affect particles with low velocities $v_z < \sqrt{e\phi_1/m}$, where $\phi_1$ is the asymmetry amplitude. With realistic ends, the plasma length is no longer the same as the applied asymmetry wavelength and is a function of radius, and $\omega_R$ is a function of radius and axial position. For conditions matching our experiment, these end effects cause a small shift in the resonant velocity and produce minor secondary resonances. The two new dynamical particle behaviors are not significantly altered by these modifications.   

Observations of an Ion Driven Instability in the CNT

An instability develops when there is an appreciable ion fraction in an otherwise pure electron plasma in the CNT stellarator. Plasma oscillations with real frequencies around 12kHz are observed for a wide range of experimental parameters, with a local peak in growth rate when $w_{pe}^2 /w_{ce}^2 \approx 2.5\ast 10^{-4}$. Much stronger oscillations in the 1kHz range are also observed intermittently. Ion driven instabilities have been observed in penning traps and pure toroidal electron traps, and the theory describing them is fairly well known. However, the existence of magnetic surfaces in CNT makes the physics governing an ion driven instability in this machine very different from penning and pure-toroidal traps. This poster presents experimental and basic theoretical work that is being done to understand how this ion driven instability evolves on magnetic surfaces.   

Equilibrium Reconstructions of Pure Electron Plasmas Confined on Magnetic Surfaces

Using recently obtained experimental data from the Columbia Non-Neutral Torus (CNT), the first equilibrium reconstructions of electron plasmas confined in a stellarator have been performed. A fully three-dimensional equilibrium solver (PBS) is used to solve the non-linear Poisson-Boltzmann equation for electrostatic potential that characterizes such plasmas. In order to fully specify an equilibrium, measurements of temperature and density are required. Specifically, we require one-dimensional temperature and density profiles along any line that runs from the magnetic axis to the outer magnetic surface. These are the inputs for the PBS code, which outputs full three-dimensional density and potential information - these quantities are not flux functions. In fact, for typical CNT plasmas, the PBS code predicts a factor of four variation of electron density along the magnetic axis. Comparisons between CNT equilibrium reconstructions and measurements will be presented, along with work done in collaboration with the Compact Helical System (CHS) group. This material is based upon work supported by the National Science Foundation under Grant No. 0317359.   

Pure Electron Equilibrium Experiments in the CNT Stellarator

CNT is a simple stellarator designed to study non-neutral, partly neutral and positron-electron plasmas on magnetic surfaces. Pure electron plasmas are currently being studied. These plasmas are created by a single negatively biased, tungsten filament on the magnetic axis. Plasmas are also diagnosed by tungsten filaments --- when heated these filaments act as emissive probes and when cold, Langmuir probes. Local plasma potential, temperature and density are measured from probe filament I-V characteristics. The CNT device, diagnostics and measurement techniques will all be described. \\ Particle confinement times of $\tau_c\!\approx\!20$\,ms have been measured, limited by the presence of bulk insulating materials in the confining region. These plasmas are in macroscopically stable equilibrium. Radial equilibrium potential, temperature and density profiles have been measured. $N\!\approx\!10^{11}$ particles have been confined corresponding to an average density of $n_e\!\approx\!10^{12}\,\mbox{m}^{-3}$. The temperature profiles are flat through most of the confining region at $T_e\!\approx\!5$\,eV. These measurements of $n_e$ and $T_e$ indicate that relatively cold, small Debye length plasmas have been created. Comparisons between experimental profile measurements and numerical predictions have shown good agreement. These experiments, their results and comparisons to numerical predictions will be discussed.   

Overview of the Columbia Non-neutral Torus program

The Columbia Non-neutral Torus (CNT) is an experiment designed to study pure electron, electron-positron and partially neutralized plasmas in a stellarator. Stellarators have unique properties as charged particle traps, in particular the ability to confine plasmas at an arbitrary degree of neutralization all the way from pure electron to quasineutral plasmas. CNT has been in operation since November 2004, and experiments with non- neutral plasmas are now being conducted. An overview of the CNT program will be presented. This includes recent studies of pure electron equilibria and instabilities seen in electron plasmas with a finite ion content. Ongoing and future work on ion resonant instabilities, operation with a retractable emitter, and plans for electron-positron plasma research will also be discussed.   

First Experiments with the Retractable Electron Emitter on the Columbia Non-neutral Torus

One goal of the Columbia Non-neutral Torus research program is to investigate the possibility of enhanced confinement in stellarators due to high electric fields. Non-neutral plasmas with high electric fields are predicted to have long confinement times. Pure electron plasmas have recently been created in the CNT stellarator, however, it was seen that the factor limiting confinement was the presence of ceramic rods holding a probe array and an electron emitter in the plasma. The confinement time was measured to be 20 ms, much shorter than the theoretically possible confinement times in the device. In order to realize the possible enhanced confinement times, a retractable electron emitter has been employed. The electron emitter is able to retract from the core of the plasma to the edge in 20 ms. Using a capacitive sector probe, we will be able to measure the decay of the image charge of the plasma and hence determine the confinement time in an unperturbed manner. Results from first experiments with the retractable emitter will be presented and discussed.   

First results from the Lawrence Non-neutral Torus II: a toroidal electron plasma device

Progress towards studying the dynamics of long-lived ($\sim$1 s) non-neutral plasmas in toroidal geometry will be presented. Achieving trapping times on the order of 1 second will permit study of higher order mode dynamics and inherently toroidal transport physics [1]. This poster presents the first results from the newly constructed Lawrence Non-Neutral Torus (LNT) II apparatus. The LNT II features an improved toroidal magnetic field magnitude ($\sim$ 0.5 kG) and base vacuum pressure ($<$10$^{-9}$ Torr). A segmented Au-plated Al electrode shell contributes to the reduction in field asymmetries and enables enhanced diagnostics. Additionally, the electron source is located on a retractable bellows for study of confinement dynamics in a complete torus. Confinement times on the order of 1 second would represent more than an order of magnitude improvement over measurements made with the previous apparatus [2].\\ $[1]$ S.M. Crooks and T. M. O'Neil, \textit{Phys. Plasmas} \textbf{3}, 2533 (1996)\\ $[2]$ M.R. Stoneking \textit{et al.} \textit{Phys. Rev. Lett.} \textbf{92}, 095003 (2004)   

Experimental Measurements of Elliptical Electron Vortices

In the 2D $E \times B$ approximation, a magnetized electron column is a vortex evolving in $(r,\theta)$ according to the Euler equation. We trap pure electron plasmas within hollow conducting cylinders in a uniform axial magnetic field, and subsequently measure the density $n(r,\theta,t)$ by dumping the electrons onto a biased phosphor screen and measuring the light intensity. Diocotron/Kelvin modes, which are $\cos(m,\theta)$ surface modes with no axial dependence, can be grown with a variety of techniques. We are able to access a new regime of very large amplitude modes through the use of precisely shaped applied impulses. Vortices with large amplitude modes have been found to be susceptible to a variety of processes which contribute to axisymmetrization, including resonant wave-fluid interactions, resonant beat wave-fluid interactions, and filamentation. We have explored the stability of very elliptical vortices with aspects ratios of up to $a/b \sim 6$. We find that these vortices are additionally subject to instabilities of $m=2,3$ and 4 surface modes similar to those predicted in 1893 for Kirchhoff elliptical vortices with $a/b>3$. Interestingly, the instability is observed on vortices with aspect ratios well below 3, and the effect may play a more important role in axisymmetrization than previously thought.   

Experimental measurements of the $m=1$ unstable diocotron mode

The first experimental observation of an exponentially growing $m=1$ unstable diocotron mode on a trapped, magnetized, partially hollow electron column was in 1990\footnote{C. F. Driscoll, {\it Phys. Rev. Lett.} {\bf 64}, 645 (1990).}, and subsequent measurements examined the effect of end shape curvature\footnote{A. A. Kabantsev and C. F. Driscoll, Non-Neutral Plasmas III, 208 (1999).}. We present new experimental measurements of the $\mu$ (initial column hollowness parameter) and $\kappa$ (end shape curvature parameter) dependences of growth rates of the $m=1$ instability. Measurements of the perturbed longitudinal temperatures of the electron column have been incorporated into the present experiments. We have experimentally established a $\mu^{4/3}$ scaling of the growth rates on the column hollowness $\mu$ for $\mu<2.2$. Our results of growth-rate scaling on $\kappa/\mu$ for relatively large $\mu$ and $\kappa$ are in agreement with theoretical predictions for the instability near onset $(\kappa,\mu)\rightarrow 0$ by Finn et al.\footnote{J. M. Finn, D. del-Castillo-Negrete and D. C. Barnes, {\it Phys. Plasmas} {\bf 6}, 3744 (1999).}.   

Plasma Manipulation Techniques for Positron Storage

Described here are new plasma manipulation techniques central to the development of a multicell Penning trap\footnote{C. M. Surko and R. G. Greaves, {\it Rad. Phys. Chem.} {\bf 68}, 419 (2003).} that is designed to increase positron storage by orders of magnitude (e.g., to particle numbers N $\geq 10^{12}$). The experiments are done using test electron plasmas. A technique is described to move plasmas across the confining magnetic field and dump them at specific radial and azimuthal locations. Techniques to fill and operate two in-line plasma cells simultaneously and use of 1 kV confinement potentials are demonstrated. These experiments establish the capabilities to create, confine, and manipulate plasmas with the parameters required for a multicell trap, namely N $\geq$ $10^{10}$ in a single cell with temperatures $\leq$ 0.2 eV, plasma lengths $\sim$ 10 cm and radii $\sim$ 0.2 cm. The updated design of a multicell positron trap for $10^{12}$ particles is described. Potential applications, including prospects for a portable positron source (i.e., to replace conventional isotope and accelerator-based sources) will be discussed.   

The High-Density Fixed Point for Nonneutral Plasma Compression.

A new ``strong-drive'' regime was recently discovered for the radial compression of single component plasmas in Penning-Malmberg traps using a rotating electric field [the so-called rotating wall (RW) technique]\footnote{J. R. Danielson and C. M. Surko, {\it Phys. Rev. Lett.} {\bf 95}, 035001 (2005); and {\it Phys. Plasmas} {\bf 13}, 055706 (2006).}. The transition to this regime occurs via a bifurcation, and the steady-state density exhibits hysteresis as a function of the applied RW voltage. Plasmas can be compressed until the E x B rotation frequency, $\omega_{E}$ ($\omega_{E} \propto$ n, the plasma density) approaches the applied frequency, $\omega_{RW}$. Here, we discuss a simple nonlinear model that explains these observations as convergence to an attracting, high-density fixed point - a torque-balanced steady state. Measurements of the RW torque magnitude and dependence on drive voltage are presented. Quantitative agreement is found with a newly developed theory\footnote{M. W. Anderson and T. M. O'Neil, adjacent poster.} that calculates the torque near the fixed point produced by the Debye-shielded RW electric field. Applications of the RW technique in this high-density, strong-drive regime and factors limiting its utility will be discussed.   

An Invertible Collision Operator for Single Species Plasmas.

Inversion of the Fokker-Planck collision operator typically is not possible by analytic means. Thus, we propose a simpler alternative operator that preserves the essential properties of the true operator and yet is invertible by eigenfunction expansion. The simplified operator satisfies the following properties: it conserves particle number, momentum, and energy; it vanishes for any Maxwellian distribution function; it yields a Maxwellian distribution as the long-time solution of the Boltzmann equation $\partial f / \partial t = ( \partial f / \partial t )_{\mathrm{coll.}}$; and, being of the Fokker-Planck form, it captures the dominant role of small angle scattering in plasmas. The new operator is a modification of the Lenard-Bernstein operator,\footnote{A.~Lenard and I.B.~Bernstein, Phys.~Rev.~{\bf 112}, 1456 (1958).} which is invertible by eigenfunction expansion,\footnote{C.S.~Ng, A.~Bhattacharjee, F.~Skiff, Phys.~Rev.~Lett.~{\bf 83}, 1974 (1999).} but does not conserve momentum or energy. Of course, all three conservation properties are necessary for connection to fluid theory. As an application of the new operator, the torque exerted by a rotating field asymmetry on a nonneutral plasma is calculated. This result agrees well with measurement.\footnote{J.R.~Danielson {\it et al.}, adjacent poster.}   

Numerical Simulation of Driven Electron Acoustic Waves.

Electron-acoustic waves (EAW's) are nonlinear modes that can exist even at low amplitude.\footnote{J.P.~Holloway and J.J.~Dorning, Phys Rev A {\bf 44} 3856 (1991).} Within linear theory, EAW's would be heavily Landau damped because the wave phase velocity is comparable to the electron thermal velocity ($\omega \approx 1.3 k {\mathrm{v_{th}}}$). However, the nonlinearity (trapped particles) effectively turns off Landau damping. This paper uses Eulerian and PIC simulations to investigate the excitation and stability of EAW's.\footnote{F.~Valentini, T.M.~O'Neil, D.H.E.~Dubin, Phys Plas {\bf 13} 052303 (2006).} Successful excitation occurs when a relatively low amplitude driver field is applied resonantly for a sufficiently long time (many trapping periods). The excited EAW rings at nearly constant amplitude long after the driver is turned off, provided that the EAW has the longest wavelength that fits into the simulation domain. Otherwise, the EAW decays to a longer wavelength EAW. In phase space, this decay to a longer wavelength EAW appears as a merger of the vortex-like trapped particle distributions. In recent experiments with pure electron plasma columns (see poster by Kabantsev and Driscoll), EAW’s were successfully excited at the predicted resonant frequency, and the predicted decay to longer wavelength was observed.   

Experimental Investigation of Electron Acoustic Waves in Electron Plasmas.

Electron-acoustic wave (EAW) solutions of the linearized electrostatic Vlasov equations are usually ignored because their small phase velocity implies a huge linear damping. However, recent nonlinear theory and simulations\footnote{J.P.~Holloway and J.J.~Dorning, Phys.~Rev.~A.~{\bf 44} 3856 (1991).}$^,$\footnote{F.~Valentini, T.M.~O'Neil, D.H.E.~Dubin, Phy Plas {\bf 13}, 052303 (2006).} found that electrons trapped in wave potentials result in long-lived BGK states at the EAW mode frequency. Experimentally, the predicted modes are readily observed on pure electron plasmas, when they are excited by weak wall voltages which are resonant over $\sim$100 cycles. The modes have phase velocity $\mathrm{v}_\phi \approx 1.3 \mathrm{v_{th}}$, in close agreement with theory; and the long-wavelength BGK states exhibit only weak damping $(-\gamma / \omega \leq 0.01)$ due to electron-electron collisions. The mode frequencies are unambiguously calibrated by comparison to electron plasma wave frequencies. Discrete standing modes are observed, but modes with $m_z = 2,3...$ show a strong decay instability into $m_z =1$. This instability corresponds to a merger of vortices in $(z, {\mathrm{v}}_z )$ phase space, which can be suppressed (or enhanced) by application of potential barriers (or wells) between the high $m_z$ wavelengths.   

Trapped-Particle-Mediated Collisional Damping of Non-Axisymmetric BGK Modes in Electron Plasmas.

Weak axial variations in magnetic or electric fields in long cylindrical electron plasmas cause a small fraction of the electrons to be trapped axially. Collisional diffusion across the trapping separatrix then causes surprisingly large transport and damping effects, including the damping of $m_\theta = 1,2,...$, $k_z = \pm 1$ Trivelpiece-Gould (TG) plasma modes discussed here. These modes are Landau damped at low amplitudes, but they appear as long-lived BGK states ($- \gamma/ \omega \sim 10^{-4}$) when strongly excited. We observe that trapped-particle-mediated (TPM) collisional damping (predicted to scale as $\gamma \propto ( \nu_{ee} / \omega )^{1/2}$) generally dominates over traditional collisional damping (scaling as $\nu_{ee} / \omega$) in determining the lifetime of the BGK state. Experimentally, this TPM damping is readily enhanced by additional trapping barriers or by wiggle-induced resonant scattering. The frequencies and eigenfunctions of the BGK states show close agreement with linear theory, except for small, amplitude-dependent frequency shifts $f(A) = f_0$$[1-a \, \ln (1+bA)]$, similar to (stronger) shifts observed\footnote{J.D.~Moody \& C.F.~Driscoll, Phys.~Plas.~{\bf 2}, 4482 (1995); W.~Bertsche, J.~Fajans \& L.~Friedland, Phys.~Rev.~Lett. {\bf 91}, 265003 (2003).} for $m_\theta = 0$ BGK states.   

Diocotron Modes in Nonneutral Disc Plasmas.

We study the ${\mathbf E} \times {\mathbf B}$ dynamics of a thin disc of charge confined in a cylindrical Penning trap with wall radius $R_w$. The axial extent (thickness) of the disc is assumed to be negligible. We show that, as in other similar cases, monotonic density profiles are stable to small perturbations. We then consider the specific density profile that is a uniformly charged spheroid of radius R projected onto a plane, because for $R \ll R_w$ it has the special property of rotating without shear. An eigenvalue equation for a density perturbation can be obtained by linearizing the equations of motion and using a Green's function. We find an analytic expression for a special class of eigenmodes, which are the diocotron modes, and the corresponding eigenfrequencies. These expressions are compared to the results of a numerical computation which was done by discretizing the eigenvalue equation. These results are also compared to the 2d limit of the Dubin theory of electrostatic modes in a spheroidal plasma.\footnote{D.H.E. Dubin, Phys. Rev. Lett. {\bf 66}, 2076 (1991).}   

Collision Rate Measurements in Mildly Correlated Ion Plasmas.

In strongly magnetized single component plasmas, collisional equipartition of $T_\|$ and $T_\perp$ is strongly suppressed by an adiabatic invariant, but enhanced by correlation effects.\footnote{D.H.E.~Dubin, Phys.~Rev.~Lett.~{\bf 94}, 025002 (2005); M.J.~Jensen {\it et al.}, Phys.~Rev.~Lett.~{\bf 94}, 025001 (2005).} In essence, equipartition occurs only due to rare close collisions, and correlated particle Coulomb-screening makes them somewhat more likely. We conduct experiments with $10^6$ to $10^8$ Mg$^+$ ions in a Penning-Malmberg trap at $B = 3$T. For density $n = 10^6$, laser cooling parallel to $B$ gives $T_\| = 0.1$K and $T_\perp = 1$K, resulting in a correlation parameter $\Gamma = 3$. When the cooling is halted, $T_\|$ rises slowly ($\sim$~1K/sec) due to weak neutral collisions, then rises abruptly as the equipartition rate becomes large. The abrupt equipartition is sometimes observed to occur spontaneously, especially when the plasma is contaminated by multiple ion species. Alternately, we can trigger the equipartition with an applied local heat pulse, and measure the ``burn front'' propagation. Experimental results will be compared to theories of correlation-enhanced equipartition, including the effects of heat transport due to long-range collisions.   

Antihydrogen Relaxation from High-n to Ground State.

We explore the rate at which magnetized, high-n Rydberg pairs formed in antihydrogen experiments\footnote{G. Gabrielse, N.S. Bowden, P. Oxley, {\it et al.}, Phys. Rev. Lett. {\bf 89}, 213401 (2002); M. Amoretti, C. Amsler, G. Bonomi, {\it et al.}, Nature (London) {\bf 419}, 456 (2002).} relax to deep binding. While the theoretical three-body recombination rate scales favorably with low temperature ($\nu_{TBR} \propto nb^3 (n \bar{v} b^2 ) \propto T^{-9/2}$), pairs form with binding energies $\varepsilon$ near the (low) thermal level.\footnote{M.E. Glinsky and T.M. O'Neil, Phys. Fluids B {\bf 3}, 1279 (1991).}$^,$\footnote{R. Robicheaux and J.D. Hanson, Phys. Rev. A {\bf 69}, 010701 (2004).} Such atoms have classical drift orbits with negligible radiation. Collisions propel a cascade to deeper binding, but theory and simulation show an atom is unlikely to reach a radiating regime before it escapes the trap.\footnote{E.M. Bass and D.H.E. Dubin, Phys. Plasmas {\bf 11}, 1240 (2004).} However, simulations show that the energy-loss rate does not decrease as rapidly with increasing $\varepsilon$ as previously expected. We also discuss the mean magnetic moment of guiding-center atoms, and energy loss from adiation at deep binding, based on the classical Larmour formula and a presumption of stochastic orbits.   

X-ray emission from planet Venus. Comparison with Chandra observations.

It is a goal of the presentation to show that recently observed X-ray emission from non-magnetic planets Venus and Mars (Ref. 1) can be explained by the same mechanism that was originally proposed for explanation of X-ray emission from comets (Ref. 2), namely as combination of Bremmstrahlung and line K shell radiation produced in interaction of energetic electrons with neutral atmosphere. It is argued that fluorescence of solar X-rays in no way cannot be the source of the observed emission. Numerical simulation are presented that show how energetic electrons can be produced by modified two stream instability developing in planetary mantle as the result of interaction of counterstreaming plasma populations - solar wind and ionospheric cold plasma. Comparison with experimental data is also carried out. 1) Dennerel et al Astronomy and Astrophysics, 386, 319, 2002 2) Bingham et al Science, 275, 49, 1997 Shapiro et al J Geophys, Res, 104, 2587, 1999   

Studies of strong Langmuir turbulence effects at HAARP

High power HF transmitters induce a number of plasma instabilities in the interaction region of overdense ionospheric plasma. Radars such as SuperDARN have been used to study artificial field-aligned irregularities (AFAI) created by the high power HF radiowave at the HAARP Ionospheric Observatory, Gakona, AK. A new Modular UHF Ionospheric Radar (MUIR) sited at HAARP may now be used to monitor changes in the Langmuir plasma waves detected in the UHF backscatter. We report the results from recent campaigns using these new facilities in coordinated and comprehensive studies of strong Langmuir turbulence (SLT). Among the effects observed and studied are: SLT spectra including the outshifted plasma line or `free-mode', appearance of a short timescale ponderomotive overshoot effect, temporal evolution of SLT, dependence of SLT on growth or suppression of AFAI, dependence of AFAI and MUIR backscatter on HAARP pulselength and duty-cycle, aspect angle dependence of the intensity of the plasma line. In particular, we explore the observed `magnetic-zenith' effect of increased turbulence with the HF wave directed up the field line. Langmuir modes parallel to the geomagnetic field are proposed to explain other features in stimulated electromagnetic emissions (SEE). These plasma waves are theorized to play a key role in certain features of radio-induced aurora. Experimental results are then compared to predictions from recent modeling efforts.   

Turbulent heating of magnetospheric ions in downward Birkeland current sheets

A new fluid theory in the guiding-center and gyrotropic approximation derivable from the Vlasov-Maxwell equations that included the effect of wave-particle interactions for weakly turbulent, weakly inhomogeneous, non-uniformly magnetized plasma was recently given by Jasperse, Basu, Lund and Bouhram [Phys. Plasmas, July (2006)]. In that theory, the particles are transported in one spatial dimension (the distance along the magnetic field) but the turbulence is two-dimensional. In this paper, the above theory is used, together with satellite measurements of the levels of wave turbulence, to calculate the altitude profile of the perpendicular ion temperature, $T_{i\bot } $, for a downward Birkeland current sheet in the earth's magnetosphere. The ions are heated perpendicular to the geomagnetic field by cyclotron resonance with random electrostatic turbulence near the ion cyclotron frequency and its harmonics. Comparisons of the calculated $T_{i\bot } $ to satellite measurements of $T_{i\bot } $ show reasonable agreement. A new formula is also given for the altitude profile of the turbulent, perpendicular ion-heating rate per unit volume.   

Advanced Kinetic Techniques for Global Magnetospheric Simulations

The Solar wind-Earth interaction leads to a complex region of space demarcated by multitudes of coupled boundaries and discontinuities, which mediate the transfer of mass, momentum and energy from the solar wind into the magnetosphere and the ionosphere. \textit{The relative importance of these transfer mechanisms is still under debate}. The collisionless nature of the space plasmas necessitates their kinetic treatment. With much progress in computational techniques and increased computational power over the past several years, global magnetospheric hybrid simulations (electron fluid, kinetic ions) have become possible. Thus, discontinuities, current sheets, and boundary layers that were previously only studied in isolation, can now be investigated in their proper context: in the magnetosphere in its entirety, and within the context of solar wind -- magnetosphere interaction. However, care must be taken that the discontinuities such as the bow shock and the magnetopause are properly resolved. Furthermore, it is important that the required explicit resistivity (to enable reconnection) is highly localized and does not affect the thickness of current sheets and other discontinuities, or dissipation at the bow shock. We present an overview of novel techniques we are developing to make global hybrid simulations possible, and report results with particular emphasis on the resolution, numerical accuracy, and effect of resistivity models.   

Electron acceleration by drift wave solitons in the magnetopause: joint Cluster-Double Star observations

Joint Cluster and Double star observations at the dusk flank magnetopause have shown evidence for bi-streaming (aligned to the magnetic field direction), energised electrons in the magnetopause boundary layer. We believe this distribution is maintained by solitary structures associated with electrostatic zonal flows, which carry drift-mode solitons down the density gradient within the magnetopause boundary layer. These structures are typically elongated along the magnetic field direction. The observations uniquely provide five point measurements and sample boundary structure on multiple spatial scales. Combined electron measuremets cover both the energetic (40-1000 keV) and thermal (20 eV-20 keV) energy range. This information allows us to investigate the comparative behaviour both within and on either side of the magnetopause, and at both low and high latitude locations simultaneously.   

Whistler Scattering of Suprathermal Electrons in the Solar Wind: Particle-in-Cell Simulations

Recent solar wind observations show that superathermal electrons (100 eV $\le E_{\parallel} \le$ 1 keV) of the magnetic-field-aligned ``strahl'' component can, under conditions of enhanced, high-frequency magnetic fluctuations, exhibit pitch-angle distributions which become broader with increasing electron kinetic energy. Magnetosonic-whistler fluctuations at ${\bf k} \times {\bf B}_o =$ 0 (where ${\bf B}_o$ is the background magnetic field) have a strong cyclotron resonance with such electrons. This resonance enables strong pitch-angle scattering, typically leading to an increase in the perpendicular (to ${\bf B}_o$) energies of these electrons. Particle-in-cell simulations in a magnetized, homogeneous, collisionless plasma of electrons and protons are used to study the electron response to whistler fluctuations. The simulations confirm that scattering by these fluctuations indeed leads to broadening of strahl pitch-angle distributions. Further simulations will attempt to determine the conditions on magnetic fluctuation spectra and electron velocity distributions such that the properties of the observations mentioned above can be reproduced.   

Remotely measured magnetospheric ion temperatures during a geomagnetic storm

Previous studies have demonstrated that with the MENA instrument aboard the IMAGE spacecraft it is possible to remotely measure the ion temperature of the magnetosphere during periods of strong geomagnetic activity. However, neutral atom imaging of the magnetosphere during quiet intervals is problematic. In this work, we show that by mapping neutral atom fluxes obtained over many days of observation to an equatorial plane fixed in GSM coordinates it is possible to construct neutral atom images of the quiet time magnetosphere. Enhanced neutral fluxes in the range of 1 to 70 keV/nucleon are observed in the quiet-time pre- midnight region. A superposed-epoch analysis of multiple storm intervals also permits imaging of the ion temperature structure as a function of time through a geomagnetic storm. Using nearly forty storms to produce average ion temperature maps as a function of storm phase, we find that there are significant differences between the spatial distribution of neutral fluxes and ion heating. Pronounced ion heating (up to 12 keV) is observed on the dayside the main phase from 5 to 8 Earth radii. During early recovery, the ion temperature on the dayside drops to approximately 9 keV and a colder region of approximately 6.5 keV persists near pre-dawn. In the late recovery, the ion temperature throughout the inner magnetosphere appears to relax to a nearly uniform 8 keV.   

Heavy ion heating due to interaction with outward and inward Alfven wave packets

Study of the heavy ion simultaneous cyclotron interactions with outward and inward propagating Alfven wave packets in the solar wind was conducted. It was shown that because the lines of pitch-angle diffusion of the ions due to interaction with each separate type of waves intersect, the ions scatter not only over pitch angle but also in energy - second-order Fermi acceleration (e.g. Ref. 1). The dynamics of the ion distribution function was investigated. It was shown that even in case of large ratio of intensities of Alfven waves propagating from the sun and inward propagating waves, the distribution function of ions simultaneously interacting with both wave packets drastically differs from the one in case when ions interact only with waves propagating from the sun. In the latter case the ions acquire a shell-like distribution while in the former case a new non-shell-type distribution with much larger effective temperatures is formed. \newline \newline [1] M.A. Forman, and G.M. Webb, Acceleration of energetic particles, in \textit{Collisionless Shocks in the Heliosphere, A Tutorial Review,} edited by R. Stone and B. Tsurutani, \textit{Geophys. Monograph, 34,} 91, 1985.   

Simulation of Density Fluctuation Intermittency in Decaying Kinetic Alfv\'{e}n Wave Turbulence

We investigate the possibility that the distance-to-source scaling of the temporal width of pulsar signals is caused by intermittency in the turbulent density fluctuations of the interstellar medium. We simulate the small scales near the ion gyroradius where density fluctuations reach equipartition with the turbulent magnetic field in a kinetic Alfv\'{e}n wave (KAW) cascade$^{1}$. Examination of decaying KAW turbulence using a fluid model establishes the presence of strong intermittency in the current fluctuation field, and a profile of Gaussian curvature of magnetic field that supports the theoretical prediction that the intermittent current filaments avoid turbulent mixing by refracting turbulent KAW activity away from filaments. The theory predicts correlated structures in electron density. These are observed but their intermittency is less pronounced. The role of parameter values, including the ratio of resistivity to density diffusivity, and the initial spectrum are explored to determine the effect on intermittency.$^{ }$ $^{1}$P.W. Terry, et al., Phys. Plasmas \textbf{8}, 2707 (2001).   

Shocks and discontinuities in the Hall-MHD model

There are a number of papers already published in the literature which present fundamentally incorrect results on shocks and other discontinuities possible within the Hall-MHD model. The basic mistake is that the previous works assume that the jump conditions across a discontinuity must be local, connecting values of the field in the same location on both sides of the discontinuity. But this is incorrect. For example, ``massless'' electrons can travel along the discontinuity surface at ``infinite'' speed, enter the surface in one location and exit at another. We work out the shock-structure problem and present other explicit examples for the rather small number of discontinuities allowed within this model. As the shock strength diminishes, we show that the results converge to our recently dereived [1] dispersion relations of Hall-MHD waves. [1] E. Hameiri, A. Ishizawa, and A. Ishida, Phys. Plasmas\textbf{ 12, }072109 (2005).   

Energy Release in a Magnetized Resistive Corona Driven by Continuous Footpoint Motions

The solar corona is a highly conducting plasma (Lundquist number $S \sim 10^{10-13}$). As such, Ohmic dissipation is negligible except within thin current filaments. In his corona heating model, Parker suggests that thin current filaments can be induced in a magnetized corona via the shuffling of the field lines driven by continuous footpoint motions. We study this model with a three-dimensional resistive MHD code. A uniform and straight initial magnetic field is slowly driven by random footpoint motions, and the subsequent energy release is observed. Since the realistic parameters of the solar corona are unattainable in the simulation, one of the main objectives is to establish a scaling law of the energy release rate with respect to the Lundquist number, so that the simulation results can be extrapolated to the real system.   

Ion Scattering in the Solar Wind and Solar Corona: Particle-in-Cell Simulations

Alfv\'en-cyclotron fluctuations at sufficiently short wavelengths and at propagation approximately parallel or antiparallel to a background magnetic field ${\bf B}_o$ in a relatively uniform, collisionless plasma can interact with protons and heavy ions. A cyclotron resonance between such fluctuations and the thermal velocity distribution of an ion species enables strong pitch-angle scattering, typically leading to an increase in the perpendicular (to ${\bf B}_o$) energies of that species. Particle-in-cell simulations in a magnetized, homogeneous, collisionless plasma of electrons, protons, and one very tenuous species of heavy ions are used to study the heavy ion response as a function of the initial magnetic power spectrum, the proton $\beta$, and the heavy-ion/proton relative speed. The goal of these simulations is to obtain better understanding of how Alfv\'en-cyclotron scattering may heat heavy ions in the solar corona. Magnetosonic-whistler fluctuations at ${\bf k} \times {\bf B}_o =$ 0 and sufficiently high $\beta_p$ can also scatter ions; however, this process typically leads to an increase in parallel ion energies. PIC simulations are also used to study proton scattering by such fluctuations; the results suggest that magnetosonic fluctuations may play a role in heating solar wind protons.   

Magnetohydrodynamic Shock Heating of the Solar Corona II

Our paper in the Astrophysical Journal, vol. 596, pp. 646-655 presented the results of our one dimensional computations using an adiabatic energy equation and an FCT algorithm. Those results showed that strong plane polarized Alfv\'{e}n waves that propagate along magnetic field lines up into the solar corona can develop into MHD shocks and act as a significant mechanism for coronal heating and wind acceleration in regions of open magnetic field lines. We now present results that allow circular and elliptical polarizations, and we also include gravitational stratification.   

Surfatron acceleration in coronal mass ejections

We use a massively parallel 3D hybrid particle code, dHybrid, to simulate the corona environment and the acceleration mechanisms of Solar Energetic Particles. The hybrid model uses massless fluid electrons and kinetic ions. The parallel implementation of this model allows the study of large regions of space (e.g. hundreds of ion gyro radius) over extended periods of time (e.g. tens of ion gyro periods), ideal for space plasma studies. In our simulations a CME structure propagates at speeds of up to 1000 km/s interacting with the slower solar wind. The interactions cause the formation of a large scale quasi-parallel shock structure due to the flowing CME. The acceleration mechanisms of high energy particles are studied in this scenario. In the early acceleration phase, our results show that particles crossing the shock front accelerate perpendicularly to the shock front while maintaining their parallel velocity, supporting a surfatron-like acceleration model. The importance of this acceleration model as a means of providing a seed particle population for further acceleration is studied.   

Gyrokinetic Studies of Astrophysical Turbulence

Magnetized turbulence plays an important role in a wide variety of astrophysical plasmas, including accretion disks, the solar wind, and the interstellar and intracluster medium. Theories of magnetohydrodynamic (MHD) turbulence suggest that, at small scales, fluctuations become highly elongated along the direction of the magnetic field with frequencies small compared to the ion cyclotron frequency. In some astrophysical settings, the dissipation of this turbulence occurs under collisionless conditions, requiring the inclusion of the kinematic dynamics of the Landau resonance. Such turbulence is well described by a low-frequency expansion of kinetic theory called gyrokinetics. Here we describe efforts to study astrophysical turbulence using gyrokinetics, employing both analytical and numerical approaches, including a detailed understanding of the dissipation of the turbulence and consequent particle heating. Implications of this work for particle heating in radiatively inefficient accretion flows, heating and power spectra in the solar wind, and density fluctuations in the interstellar medium are discussed.   

Effect of Interplanetary Shocks on AL and Dst Indices and Real-Time WINDMI

Analytic solar wind signals are constructed using data from the ACE satellite for the 3-6 October 2000 and 15-24 April 2002 geomagnetic storm events, which included interplanetary shocks and magnetic clouds. The response of the WINDMI model, an eight dimensional model of the solar wind driven magnetosphere-ionosphere system, to the analytic signals was examined for these events. The role of the shocks are examined by using analytic signals in which the shock feature in the density, solar wind velocity, and magnetic field magnitude are tested individually. WINDMI results from the analytic inputs show that the shock events strongly impacted the $AL$ index values but changed the $Dst$ very little. Analysis of both the October 2000 and April 2002 storms showed that $AL$ peaks associated with the shocks during these storms significantly depend on the jump in $B_{\perp}$. Real-time WINDMI downloads real-time data from ACE to predict the $AL$ and $Dst$ about 1-2 hours before the data is available for these indices from the Quicklook WDC-Kyoto website. Every ten minutes real-time $AL$ and $Dst$ data and WINDMI predictions are shown on this website: http://orion.ph.utexas.edu/$\sim$windmi/realtime/. We show WINDMI real-time predictions that were captured for a recent storm on 14-15 April 2006. The work is supported by NSF-ATM grant 0539099.   

Relativistic Plasmas: A New Interface between High Energy Astrophysics and Intense Laser Experiments

In the relativistic regime, the MHD approximation often breaks down even when the gyroradii and plasma skin depths are much smaller than the relevant plasma scale height. Examples include electromagnetic-dominated plasmas where the formal alfven speed v{\_}A $>$ c or the dimensionless vector potential a{\_}o $>$ 1, and plasmas containing species with highly anisotropic relativistic momentum distributions. In such cases conventional intuition based on MHD concepts fails, and we discover novel unexpected behaviors in shock structure, particle acceleration and radiation mechanisms. In this talk we highlight several examples that lie at the interface between high energy astrophysics phenomena such as gamma-ray bursts and pulsar winds, and intense laser experiments in the laboratory.   

Coupling of fluid-scale and kinetic-scale processes in momentum transport in space and laboratory devices

Two sets of experiments and observations suggest that an important factor in the coupling of microinstabilities with large scale processes is momentum transport. In astrophysical systems, there is mounting evidence that microinstabilites can lead to large-scale plasma flows. In the Earth’s magnetotail, satellite observations have shown microinstabilites driven by pressure gradients at the edge of plasma sheet [1]. We have predicted that such instabilities can lead to momentum transport that penetrates into the current layer and lead to velocity shears on macroscopic scales [2]. In laboratory experiments, for example on Alcator C-Mod, researchers have shown that toroidal angular momentum is generated spontaneously in the plasma edge and propagates inward [3]. The analogy with the space observations is striking as in both cases the microinstabilities develop in the edge of the plasma but momentum propagates inward and angular momentum is generated in the toroidal direction (dawn-dusk direction). Also, at least some of the theories [4] assume that the microinstabilites are due to pressure gradients in both cases. We report our approach to study the problem: a full kinetic implicit simulation code CELESTE [5]. [1] I. Shinohara et al., J. Geophys. Res., 103, 20,365, 1998 [2] G. Lapenta, et al., Phys. Plasmas, 10, 1577, 2003. [3] J.E. Rice et al., Phys. Plasmas, 11, 2427, 2004. [4] B. Coppi, Nuclear Fusion, 42, 1, 2002. [5] G. Lapenta et al., Phys. Plasmas, 13, 055904,2006   

Vector Transport Processes in a Magnetized Dilute Plasma

In this work we present a formal derivation of transport phenomena in a dilute plasma based on the solution to the complete Boltzmann equation which, in contrast to its diffusive approximation (Fokker-Planck), allows for cross-effects and the derivation of exact transport coefficients in all directions (parallel, perpendicular and oblique to the magnetic field). It also permits, by appropriately defining thermodynamic fluxes and forces, the consistent placement of the theory within the framework of non-equilibrium thermodynamics. We show an explicit calculation of all transport coefficients including the Soret and Dufour cross-effects which are usually neglected in plasma transport theories. We here show that the latter is not only relevant but also the most important source of heat conduction for weak magnetic fields in both parallel and perpendicular directions, leading to a new effective conductivity coefficient. Other aspects of this work, as the Righi-Leduc effect, are also emphasized and progress in the visco-magnetic effects is presented.   

Analysis and Simulation of Plasma Wave Damping and Viscous Heating Due to the Generalized Ion Stress Tensor

Details of the analysis and simulation of two key effects of the generalized parallel ion stress tensor ($\Pi _{\vert \vert })$, in magnetized plasmas is presented in this work. Kinetic-based derivation of the $\Pi _{\vert \vert }$, employing a Chapman-Enskog-like (CEL) expansion of the particle distribution function [1], forms the theoretical basis of this study. The ultimate goal of this research is to incorporate adequate kinetic and non-collisional physics into the modeling of tenuous, high-temperature (fusion-grade) plasmas. Systematic application of hybrid fluid/kinetic models in the simulation of plasma systems gives accurate results. Sound wave propagation (parallel to the magnetic field), dissipation and the associated stress heating, are modeled as being generated by single scale-length flow perturbations. NIMROD simulation runs are , incorporating finite physical effects of the generalized $\Pi _{\vert \vert }$, in slab geometry. These runs involve various scans of plasma parameters corresponding to various degrees of plasma collisionality. Results show that parallel viscosity, damping rate and viscous heating obtained with the generalized $\Pi _{\vert \vert }$ agree with those based on Braginskii $\Pi _{\vert \vert }$ (at high collisionality). Also, generalized $\Pi _{\vert \vert }$ predicts more realistic values at low collisionality. Future research would focus on incorporating time-dependent effects, and also on using multiple scale-length flow perturbations.   

Test particle acceleration in three-dimensional Hall MHD turbulence

Numerical experiments of test particle acceleration are performed using turbulent magnetic and electric fields obtained from direct numerical solutions of the compressible three-dimensional Hall magnetohydrodynamic (MHD) equations. Comparisons are made of the results for the test particles momentum distribution function with and without the Hall term in the MHD solution. Electrons and protons are considered for the test particles at short times before particles leave the simulation box (with length size of the order of a few turbulent correlation lengths). The particles momentum distribution functions develop long tails in a short time. No substantial difference it is found between results with and without the Hall term in the MHD solution. When a background uniform magnetic field (guide field) is added, the particles acceleration is anisotropic. Electrons develop large parallel momentum, while protons develop large perpendicular momentum. A discussion of the basic particle acceleration mechanisms in this system is made.   

Microscopic Processes Involved in the Transition from a Disk to a Ring Sequence

In a differentially rotating plasma structure, in the prevailing gravity of a central object, it is realistic to consider $\epsilon_{m}=D_{m}/{(Hv_{A})}<< 1$, where $D_{m}= \eta\ c^2/4 {\pi}$, $\eta$ is the plasma resistivity, $v_{A}$ is the Alfv\'en velocity, and H is the height of the structure. Therefore, the frozen-in- law can be applied when dealing with global scale distances. However, the transition from a disk configuration\footnote{B. Coppi\textit{ Phys. Plasmas} \textbf{12}, 057301, (2005)} to a ring sequence configuration\footnote{B. Coppi and F. Rousseau, \textit{Ap. J.} \textbf{641} (1), 458 (2006).} in a structure imbedded in an external magnetic field with a ``seed'' vertical component involves the formation of a periodic sequence of new closed magnetic surfaces$^1$ similar to those characteristic of Field Reverse Configurations. Pairs of counter-streaming (around the axis of rotation) current ``filaments'' form these very large aspect ratio toroidal configurations. The transition occurs when the field produced by the internal currents is equal (and of opposite direction) to that in which the structure is immersed. Then, magnetic reconnection has to be considered. The region in which the poloidal field is null can be viewed as a coalescence of two X-points and two O-points which, after the transition has occurred, tend to separate. The scale distances involved are much smaller than the characteristic global scale distances, such as H, the characteristic values of the collisional mean free paths need to be considered, and relevant microscopic reconnection processes are analyzed. *Sponsored in part by the U.S. DOE.