Bulletin of the American Physical Society
49th Annual Meeting of the Division of Plasma Physics
Volume 52, Number 11
Monday–Friday, November 12–16, 2007; Orlando, Florida
Session UP8: Poster Session VIII: Gyrokinetics, Two-Fluids and Assorted; DIII-D II; Turbulence and Transport; Plasma Sources, Sheaths, and Thrusters |
Hide Abstracts |
Room: Rosen Centre Hotel Grand Ballroom, 2:00pm - 5:00pm |
|
UP8.00001: GYROKINETICS, TWO-FLUIDS AND ASSORTED |
|
UP8.00002: An electrostatic, gyrokinetic-like model for thin layers with large gradients and electrostatic potentials Harold Weitzner, Choong-Seock Chang, Frank Jenko Plasma dynamics is studied in a layer whose thickness is of the order of the geometric mean of the ion Larmor radius and the macroscopic scale length. The particle distribution functions and electric and magnetic fields all have the given gradient length. The motion of a single particle in such a state is treated. A magnetic moment adiabatic invariant exists and drift Hamiltonians are constructed. Kinetic equations are then developed, and have some properties of gyrokinetic equations. Charge neutrality, used to determine the electrostatic potential, is replaced by an equivalent condition, with explicit appearance of the potential. The linear stability of a straight, screw pinch is examined. Both stable and unstable special cases are found. Some results recover ion temperature gradient-like modes. However, large magnetic shear improves stability. [Preview Abstract] |
|
UP8.00003: The Solution of Boundary Layer Problems using Kruskal-Newton diagrams Roscoe White, Thomas Fischaleck The use of Kruskal-Newton diagrams for the solution of differential equations is illustrated with examples from the physics of pattern formation and plasma physics. The scaling of boundary layers is rapidly determined, and a systematic means of simplifying the internal layer equations is obtained. [Preview Abstract] |
|
UP8.00004: Power-law statistics in beam-driven plasma-wave system with fluctuations: a signature of self-organized criticality, or something else? Yu. Tyshetskiy, J.A. Roberts, P.A. Robinson, I.H. Cairns, B. Li Numerical simulations of reduced-parameter model [1] of stochastic growth theory (SGT) [2] describing beam-driven plasma-wave systems with parameter fluctuations, display power-law tails in wave energy distribution function P(W) for certain parameters of the model. The characteristic spatiotemporal linear correlation scales of the system quantities [3], estimated at these parameters, are large compared to both the correlation scales of fluctuations driving the system, and to the system size, possibly indicating criticality occurring in the system. We study whether these power-law tails and diverging correlation scales are indeed signatures of criticality, whether this state is robust to fluctuations and initial conditions (and hence criticality is self-organized [4]), and what model parameters define the transition into the power-law regime from the previously well studied regime of SGT with lognormal statistics [2]. We also measure exponents of the power-law tails and study their dependence on the model parameters. References: [1] P. A. Robinson, Solar Phys. \textbf{168}, 357 (1996). [2] P. A. Robinson, Phys. Plasmas \textbf{2}, 1466 (1995). [3] Yu. Tyshetskiy et al., ``Spatiotemporal correlation functions in beam-driven plasmas with fluctuations'', submitted to Phys. Plasmas. [4] P. Bak, C. Tang, and K. Wiesenfeld, Phys. Rev. Lett. \textbf{59}, 381 (1987). [Preview Abstract] |
|
UP8.00005: Hamiltonian theory of the nonlinear collisionless tearing mode F.L. Waelbroeck, P.J. Morrison, E. Tassi, D. Grasso The Hamiltonian formalism constitutes an effective framework for investigating the dynamics of fluid models. A particularly appealing feature of the Hamiltonian formalism is that it readily provides first integrals of the equations governing the configuration and propagation of nonlinear structures such as magnetic islands and solitary waves [1]. Here a newly developed noncanonical Hamiltonian formulation for a two-fluid model describing collisionless reconnection is used to investigate the effects of electron inertia on the nonlinear growth of the collisionless tearing mode. A variational principle is used to obtain a pair of equilibrium equations that take the form of coupled nonlinear elliptic equations for the magnetic flux and the ion stream-function. These equations generalize the Grad- Shafranov equation of MHD. The equilibrium solutions and conservation laws are used to calculate the saturation amplitude of the collisionless tearing mode and the result is compared with numerical simulations. \newline [1] F.~L. Waelbroeck, P.~J. Morrison and W. Horton, Plasma Phys. Control. Fusion {\bf 46}, 1331 (2004). [Preview Abstract] |
|
UP8.00006: Toroidal effects on gyrokinetic and fluid models Linda Sugiyama Fluid and gyrokinetic models of high temperature, magnetically confined plasmas both rely on expansion in a gyroradius parameter small compared to a reference size ($\rho_i/L<1$), but emphasize different aspects of physics. Fluid models retain the effects of toroidal geometry at lowest order but relegate important kinetic effects to ever higher velocity moments, by implicitly taking the gyroradius smaller than a fluid element. Gyro-orbit expansions drop inverse aspect ratio terms in zeroth order (drift kinetic equation), but retain kinetic effects, assuming that the gyro-radius is the most important small parameter. Both kinds of physics are needed to describe high temperature toroidal plasmas. Coupling through the fluid moments requires first or second order. Gyrokinetic models are rigorously derived to all orders for 2D straight field line systems. 3D configurations and toroidal effects greatly complicate the equations, particularly at higher order. Existing models are not completely consistent and these effects are discussed for toroidal plasmas. [Preview Abstract] |
|
UP8.00007: Ballooning Modes for Hall-MHD Plasmas with Shear Flow Eliezer Hameiri Our past work on ballooning modes in MHD plasmas with shear flow$^1$ has shown that such modes do exist but, rather than being stationary, they drift along the magnetic field line at a poloidal angular velocity equal to the ratio of the toroidal shear flow to the magnetic shear. In the present work we show that the ballooning modes also exist in a Hall-MHD plasma with shear flow, a fact which was in doubt up to now. The mode also drifts along a field line, as in MHD. But it should be noted that in order to go to the ballooning limit $(n\to \infty )$, one must also scale the Hall term parameter (the ion skin depth) to be comparable to the width of the mode, with both shrinking to zero. In order to investigate further the properties of the ballooning mode, the equations of which are rather complicated, we use a simplified configuration where the drift of the mode can be removed by transforming to a coordinate system moving along a field line, in a way that leaves a time-independent equation being analyzable by eigenvalues instead of Floquet behavior. So far we have worked out the MHD limit as a basis for comparison, since this case allows for very precise results. The Hall-MHD case is being pursued. $^1$E. Hameiri and S.T. Chun, \textit{Phys. Rev. A}\textbf{ 41}, 1186 (1990). [Preview Abstract] |
|
UP8.00008: Preconditioning and Scalability of Implicit Extended MHD Plasma Simulation by FETI-DP Domain Substructuring Alan H. Glasser The large range of multiple length and time scales in extended MHD plasma simulation makes it imperative to achieve efficient computation of implicit time steps on petascale parallel computers. The limiting factor is the rate of convergence of Krylov iterative solution of large, sparse matrices, which in turn depends on the condition number of the matrix, the ratio of the largest to smallest eigenvalues. As the size of the problem and the number of processors increase, it is essential that the condition number approximately approach a limit, not rise indefinitely, the property called scalability. Recent analytical work has proven this property for the application of the FETI-DP method of domain substructuring to a limited class of elliptic PDEs [1]. This method provides a coarse solution which assures scalability, and also an effective local method of preconditioning. We explore the extension of these results to more general systems, using computational rather than analytical methods to demonstrate scalability [2]. \newline \newline [1] Axel Klawonn and Olof B. Widlund, Dual-Primal FETI Methods for Linear Elasticity, Comm. Pure Appl. Math. 59, 1523-1572 (2006). \newline [2] A. H. Glasser and X. Z. Tang, The SEL macroscopic modeling code, Comp. Phys. Comm. 164, 1-3, 237-243 (2004). [Preview Abstract] |
|
UP8.00009: Ten-Moment Equations For Fluid Modelling of Plasmas Ammar Hakim High-order moment fluid equations for simulation of plasmas are presented. The ten--moment equations are a two--fluid model in which time dependent equations are used to advance the pressure tensor. With the inclusion of the full pressure tensor Finite Larmor Radius (FLR) effects are captured. In the absence of collisions, the solution the ten--moment equations can be considered as exact solutions to the Vlasov equation with special initial conditions on the particle distribution function. Collisional effects are included using two different methods. In the first method, the BGK form of the collisional operator is used and in the second, a linearized form of the Coulomb collision operator is used. The dispersion relation of the equation system, both with and without collisions, is presented. In the collisionless case it is shown that, in addition to the usual two--fluid waves, electron Bernstein waves are captured correctly. In the case in which collisions are included, collisional damping rates for the pressure tensor to isotropy are computed. Numerical solution to a few illustrative problems are presented. In the first, solutions to Riemann problems for the ten--moment equations is presented. These differ significantly from the two--fluid and ideal MHD Reimann solutions. Reconnection rate for a fast magentic reconnection problem is computed and compared to kinetic and other fluid models. The stability of a g-mode in a slab plasma is presented. [Preview Abstract] |
|
UP8.00010: Nonstationary Nonlocal Transport Theory of Fully Ionized Two Component Plasma Zhen Zheng, W. Rozmus, A. Brantov, V. Yu. Bychenkov Linearized electron transport theory that is fully equivalent to the solution of a Fokker-Planck equation (Bychenkov \textit{et.al.} Phys. Rev. Lett, 75, 4405 (1995)) has been generalized to include ion transport. Starting from the complete Lanadu collision operators expressed in terms of Rosenbluth potentials, electron and ion velocity distribution functions are expressed in terms of infinite series of angular harmonics. Hydrodynamical equations and transport closure relations are derived in response to initial perturbations. The complete set of frequency and k-number dependent transport coefficients has been discussed. Our results show reduction in ion thermal conductivity and ion viscosity for k$\lambda_{ii}>$10$^{-2}$ ($\lambda_{ii}$ - ion-ion collision mean free path, k -- wave number related to the inhomogeneity scale length) as compared to standard Chapman-Enskog theory results. Applications of this theory to the calculations of the dynamical form factor, ion-acoustic and entropy modes dispersion relations have been presented. Our results provide an exact limit for the nonlinear transport calculations. [Preview Abstract] |
|
UP8.00011: Electrostatic gyrokinetics in an axisymmetric torus Grigory Kagan, Peter J. Catto A gyrokinetic change of variables useful for describing electrostatic phenomena in axisymmetric tokamak magnetic field is introduced. In contrast to typical gyrokinetic treatments, canonical angular momentum is taken as the gyrokinetic radial variable rather than the radial guiding center location. Such an approach allows strong radial density gradients, while allowing zonal flow behavior in the presence of strong toroidal rotation. Moreover, neoclassical collision effects naturally enter when the gyrokinetic change of variables is applied to the collision operator. The new, nonlinear gyrokinetic variables are constructed to higher order than is typically the case by generalizing the linear procedure of Lee, Myra and Catto [2]. The nonlinear gyrokinetic equation obtained is expected be useful in analyzing electrostatic behavior in the tokamak pedestal and scrape-off layer. This choice of gyrokinetic variables allows the toroidally rotating Maxwellian solution of the isothermal tokamak limit to be recovered [1]. Moreover, the quasineutrality equation in the long wavelength limit is derived. To verify its consistency, the electrostatic potential obtained from it is compared to the expression found for an isothermal tokamak [1]. References: [1] P. J. Catto and R. D. Hazeltine, Phys. Plasmas \textbf{13}, 122508 (2006). [2] X. S. Lee, J. R. Myra and P. J. Catto, Phys. Fluids \underline {26}, 223 (1983). [Preview Abstract] |
|
UP8.00012: Nonlinear global gyrokinetic PIC simulations of collisionless TEM turbulence S. Jolliet, B.F. McMillan, T.M. Tran, X. Lapillonne, L. Villard, A. Bottino, P. Angelino, Y. Idomura Micro-instabilities, such as Ion Temperature Gradient modes (ITG) and Collisionless Trapped Electrons Modes (CTEM), are commonly held responsible for anomalous transport observed in tokamaks. While there have been a wide range of nonlinear studies on ITG turbulence, very little is known about the nonlinear physics of CTEM. This work presents the first linear and nonlinear simulations of ITG-CTEM turbulence performed with the global PIC code ORB5 [1]. A linear benchmark of ORB5 against other gyrokinetic codes will be shown. Numerical aspects such as numerical noise will be discussed. The simulations will focus on nonlinear phenomena including detrapping, toroidal coupling, zonal flows, profiles evolution and heat transport. \newline \newline [1] S. Jolliet \textit{et al.}, to appear in Comput. Phys. Commun. [Preview Abstract] |
|
UP8.00013: Status of edge gyrokinetic turbulence simulation in XGC1 Seung-Hoe Ku, C.S. Chang, D. Zorin, L. Greengard, M. Adams, J. Cummings, P. Worley, E. D'Azevedo, W. Lee, S. Parker, Y. Chen, Z. Lin Gyrokinetic simulation of a tokamak edge plasma is one of the highest priority research items for ITER and the magnetic fusion program. Due to the complex physical modeling required in the edge plasma (closed and open magnetic field lines with the magnetic separatrix in between, the importance of neoclassical physics, the material wall boundary, steep pressure gradients, a non-Maxwellian distribution function, and the neutral particle physics), most of the gyrokinetic simulation activities have so far been focused on the core plasmas. The status of the gyrokinetic edge turbulence simulation in the XGC1 particle code in the SciDAC Prototype FSP Center for Plasma Edge Simulation (CPES) will be reported. XGC1 includes the above mentioned edge complexities with full- f/delta-f particle technology on an unstructured mesh. Special physics/math/CS features will be discussed. Our current electrostatic turbulence/neoclassical capabilities will be presented and verified. Plans for incorporating full electromagnetic turbulence will also be discussed. [Preview Abstract] |
|
UP8.00014: Moment approach to the derivation of diffusive and general parallel closures Jeong-Young Ji, Eric D. Held In the moment expansion with the random velocity polynomials instead of the total velocity, the Coulomb collision operators are analytically calculated for any arbitrary order moment. For the electron-ion interaction, the leading terms in the small- mass ratio approximation are derived so that the momentum and energy are conserved. The general moment equations are also explicitly presented. For high collisionality, diffusive closures are derived with all nonlinear terms kept, and as a result, comparable terms to the Braginskii equations are additionally found. For arbitrary collisionality, parallel heat flux and viscous stress are also derived from the general parallel moment equations. The parallel closures can be computed by integrating the gradients of temperature and fluid velocity through kernel functions along a magnetic field line. The kernel functions are simple linear combinations of exponential functions. It is verified that the closure calculation converges with increasing number of moments and that lower collisionality requires more moments. As a practical example, the parallel heat flux is applied to simulate the temperature in SSPX. Finally, a numerical implementation that speeds up the closure calculation is also discussed. [Preview Abstract] |
|
UP8.00015: A drift ordered short mean-free path description for partially ionized magnetized plasma Andrei Simakov Effects of neutral particles are very important at the edge of a tokamak and so must be self-consistently accounted for. This has only been done so far for short mean-free path plasma under the high-flow Braginskii ordering [1]. Since plasma flow in modern tokamaks is often comparable with the diamagnetic heat flow divided by pressure it is appropriate to use drift ordering instead. Here we consider short mean-free path plasma consisting of electrons, singly-charged ions, and neutrals. We neglect neutral-neutral and elastic electron-neutral collisions and approximate the neutral-ion charge-exchange cross-section with a constant. We employ drift ordering to evaluate ion, neutral, and electron heat fluxes, viscosity tensors, and momentum and energy exchange terms and formulate a self- consistent system of electron, ion, and neutral fluid equations, thereby generalizing the drift-ordered treatment [2] of fully ionized plasma. \newline [1] P. Helander, S. I. Krasheninnikov, and P. J. Catto, {\it Phys. Plasmas} {\bf 1}, 3174 (1994) and references therein. \newline [2] P. J. Catto and A. N. Simakov, {\it Phys. Plasmas} {\bf 11}, 90 (2004). [Preview Abstract] |
|
UP8.00016: Atomic Models for High Charge State Uranium Plasmas David Fillmore, Peter Messmer We present a set of atomic process models for inclusion in electron-Uranium ion plasma simulations. These include an adaptation of the binary encounter dipole model of Kim and Rudd (1994) for electron impact ionization, the Burgess general formula for dielectronic recombination rates, and the semiclassical impact parameter approximation for ion-atom charge exchange. The orbital binding energies and oscillator strengths are estimated with a multi-configuration Dirac Fock Model. Comparisons are made to existing experimental measurements of electron impact ionization for neutral Uranium (Halle et al., 1981) and for U(10+), U(13+) and U(16+) (Gregory et al., 1990). These models have been developed for use in simulations of U(28+) through U(35+) production at the the Versatile Electron-Cyclotron-Resonance Ion Source for Nuclear Science (VENUS) at Lawrence Berkeley Laboratory. [Preview Abstract] |
|
UP8.00017: Experiments for measuring EIR and MAR in detaching plasmas in the PISCES-A divertor simulator Laizhong Cai, George Tynan, Eric Hollmann, Daisuke Nishijima Traditionally, Electron-Ion Recombination (EIR) is considered the dominant volume recombination process in detaching divertors. However, Molecular-Activated Recombination (MAR) was found both in PISCES-A and NAGDIS-II, which could be an important path to make a detaching plasma. The EIR sink rate is obtained in a pure magnetized He plasma column by absolutely calibrated high n (principle quantum number) He-I line emission associated with EIR. A small amount of H$_{2}$ gas is then injected into this plasma, resulting in the collapse of EIR emission with sufficient H$_{2}$ gas density. Since MAR produces H neutrals in a low excited or ground states in contrast to EIR associated with highly excited states, it will not bother the measurements of EIR. Using an integral form of the particle conservation equation, the MAR sink rate is derived with the measured parallel flux, ionization source, EIR sink and anomalous radial flux to the wall. Finally, the EIR and MAR are demonstrated quantitatively. The role of MAR is shown against the percentage of H$_{2}$ gas. [Preview Abstract] |
|
UP8.00018: Molecular Ion Effects in IEC Modeling Gilbert Emmert, John Santarius A 1-D model for the effect of various molecular and atomic processes (charge exchange, ion impact ionization, and dissociative processes) between deuterium ions (D$^{+}$, D$_{2}^{+}$, and D$_{3}^{+})$ and the background gas on the performance of spherical, gridded IEC devices has been developed. Ions pass through the anode grid primarily as an arbitrary mixture of D$^{+}$, D$_{2}^{+}$, and D$_{3}^{+}$ ions and, while being accelerated by the electrostatic potential, interact with the background gas to produce a source of cold ions (D$^{+}$ and D$_{2}^{+})$ through interactions with the background D$_{2}$ gas. These cold ions are accelerated by the potential and produce additional cold ions through interactions with the background gas. A formalism has been developed which includes the bouncing motion of ions in the potential well and sums over all generations of cold ions. This leads to a set of coupled Volterra integral equations. The integral equations are solved numerically to yield the energy spectrum of the ion and fast neutral flux; the resulting neutron production rate is calculated. Parametric surveys of the effect of the ion mixture in the source region and comparison with experimental data for the Wisconsin IEC devices will be presented. [Preview Abstract] |
|
UP8.00019: Deuterium cluster model for low energy nuclear reactions (LENR) George Miley, Heinrich Hora For studying the possible reactions of high density deuterons on the background of a degenerate electron gas, a summary of experimental observations resulted in the possibility of reactions in pm distance and more than ksec duration similar to the K-shell electron capture [1]. The essential reason was the screening of the deuterons by a factor of 14 based on the observations. Using the bosonic properties for a cluster formation of the deuterons and a model of compound nuclear reactions [2], the measured distribution of the resulting nuclei may be explained as known from the Maruhn-Greiner theory for fission. The local maximum of the distribution at the main minimum indicates the excited states of the compound nuclei during their intermediary state. This measured local maximum may be an independent proof for the deuteron clusters at LENR. \newline [1] H. Hora, G.H. Miley et al. Physics Letters A175, 138 (1993) \newline [2] H. Hora and G.H. Miley, APS March Meeting 2007, Program p. 116 [Preview Abstract] |
|
UP8.00020: New formula of Debye length in solid metallic hydrogen Kazunori Shibata, Ryosuke Kodama In arbitrary quasineutral states, ionic potential is somewhat screened by electrons. The degree of the screening is represented by Debye screening length. The Debye length varies proportional to square root of the temperature in plasma state. When the temperature drops to the Fermi temperature, electrons undergo Fermi degenerate and the Debye length becomes independent of the temperature. Such situation also appears in solid states. We have researched in solid metallic hydrogen because it is said to be a high temperature superconductor and is an ideal matter to treat the Debye screening by the lack of orbital electrons. By taking into account the changes in electronic quantum statistical state, the formula of the Debye length in the superconductive state was derived. The Debye length in the superconductive state again depends on the temperature by bosonization of the electrons. As a one application, we also have calculated the penetration probability by using the WKB approximation. The probability at $n=10^{31}$m$^{-3}$ and T=10K is comparative that of a particle of about 7.5 eV penetrates bare Coulomb potential. [Preview Abstract] |
|
UP8.00021: The Explanation of Quantum Teleportation and Entanglement Swapping Russell Moon, Victor Vasiliev According to the Vortex Theory, the rotation of a particle causes the surrounding three-dimensional space to rotate creating the particle's electromagnetic characteristic. Because three-dimensional space is the surface of fourth-dimensional space, this rotation extends slightly downward into the fourth-dimensional volume beneath. If two photons possessing complementary polarizations are ``entangled'', this extreme closeness forces their rotations extending into fourth-dimensional space to join together forming a vortex. When the particles are separated, the vortex between them remains. A change in the orientation of a photon at one end of the vortex travels in a wave down the length of the vortex creating a change in the orientation of the photon at the other end. Entangled separated particles of matter such as electrons are similarly connected and effected by each other. The breaking and reconnecting of these vortices also explains the phenomenon of entanglement swapping. 1 R.G. Moon, \textit{The Possible Existence of a New Particle: the Neutral Pentaquark}? Book of materials, The Scientific Seminar Ecology and Space 1, February 22, 2005, Saint-Petersburg, Russia, 2005. [Preview Abstract] |
|
UP8.00022: DIII-D II |
|
UP8.00023: Overview of DIII-D Experimental Results and Program Plans T.S. Taylor The DIII-D research program is addressing urgent ITER R{\&}D issues, improving Advanced Tokamak operation, and using an expanding set of control tools and diagnostics to better understand the physics of high performance tokamaks. Resistive wall mode experiments with counter beam injection addressed stability and feedback control in slowly rotating ITER-relevant plasmas, providing new data to compare with code predictions. ELM-control experiments using internal and external coils point to a physics basis for design of similar coils for ITER, and new disruption mitigation results show promise for suppression of runaway electrons. Steadily increasing ECH and fast wave power (2.4 MW ECH and 3.1 MW FW) provides the means to heat electrons to vary collisionality and T$_{e}$/T$_{i}$ to study turbulent transport with improved diagnostics to measure a broad spectrum of density and temperature fluctuations, and will improve current profile control in AT plasmas. In the near future, we anticipate increased long-pulse ECH power to extend noninductive high bootstrap fraction AT performance; longer-term plans include 10~s operation at full field. [Preview Abstract] |
|
UP8.00024: 3D Structure and Dynamics of ELMs T.E. Evans, J.G. Watkins, I. Joseph, J.H. Yu, M. Jakubowski, O. Schmitz Understanding the global topology and dynamics of edge localized modes (ELMs) is essential for predicting transient loading on plasma facing surfaces. Fast visible line emission images taken during ELMs consistently show filament-like helical structures that expand radially outward from the pedestal while rotating toroidally. Fast infrared camera images show a rapidly evolving splitting and broadening of the divertor heat flux footprints that appear to be correlated with non-axisymmetric divertor currents. In single-null poloidally diverted configurations, these experimental signatures appear to be topologically consistent with a splitting of the separatrix into a set of invariant manifolds resulting in an object known as a homoclinic tangle in dynamical systems theory. Here, we describe a model in which helical currents flowing along this tangle amplify its size and toroidal phase. We compare predictions from this model to experimental measurements of the properties of filament-like structures measured with various DIII-D fast diagnostic systems. [Preview Abstract] |
|
UP8.00025: Modeling of the Plasma Response to Resonant Magnetic Perturbations with the NIMROD Code S.E. Kruger, I. Joseph, E.D. Held, D.D. Schnack, T.E. Evans, R.A. Moyer Resonant magnetic perturbations (RMPs) have successfully been used to control ELMs in the DIII-D tokamak\footnote{R. A. Moyer, T. E. Evans, T. H. Osborne, et al., Phys. Plasmas {\bf 12} (2005) 056119} In these experiments, internal coils are used to tailor the mode amplitude spectrum at the separatrix with the goal of affecting transport at the edge. Intuitively, inducing islands at the separatrix would cause a stochastic edge and enhanced electron temperature transport; however, experimental evidence shows that both the ion and electron temperature gradients are relatively unchanged, while the density gradient is substantially reduced. Modeling of these experiments by initial-value extended magnetohydrodynamics codes is attractive because the essential features, magnetic reconnection, parallel and perpendicular transport, and ELM stability have all been independently studied. The challenge of modeling the experimental device lies in the extreme range of time scales with the ramp rate of I-coil currents being on the order of milliseconds and the Alfven time being sub-microseconds. Here we show results of modeling just the the field error penetration problem. We extend the numerical simulations of Fitzpatrick\footnote{Fitzpatrick, Physics of Plasmas, {\bf 10} (2003) 1782} to three-dimensional geometry, two-fluid physics, and anisotropic viscosity. [Preview Abstract] |
|
UP8.00026: Effect of RMP on Edge Density Profiles and Fluctuations in DIII-D L. Zeng, T.L. Rhodes, E.J. Doyle, G. Wang, W.A. Peebles, A.E. White, T.E. Evans, R.A. Moyer, M.E. Fenstermacher Resonant magnetic perturbation (RMP) has been used successfully to suppress Type-I edge localized modes (ELM) in DIII-D. In these ELM-suppressed operations, the detailed edge density profile and evolution of the fluctuations have been investigated in order to study the effect of RMP on edge transport. Utilizing a high-resolution profile reflectometer ($\Delta $t=25~$\mu $s, $\Delta $r $\ge $ 2 mm), it is observed that with even parity n=3 RMP, pellet injection results in a larger increase in the scrape-off layer density and a smaller increase in the pedestal density gradient, as compared with no RMP. This result is consistent with the decay time of pellet-induced core density perturbation with RMP being shorter than without RMP, indicating an enhanced particle transport during the ELM-suppressed phase. The detailed density profile and fluctuation evolution will be presented for various RMP configurations, e.g. n=1, n=3, in both low and high collisionalities. [Preview Abstract] |
|
UP8.00027: Particle Transport in RMP ELM Suppressed H-modes R.A. Moyer, J.A. Boedo, V.A. Izzo, I. Joseph, S. Mordijck, D.L. Rudakov, J.H. Yu, T.E. Evans, N.H. Brooks, T.H. Osborne, P. Gohil, J.S. deGrassie, A.W. Leonard, M.E. Fenstermacher, C.J. Lasnier, J.G. Watkins, T.C. Jernigan, M. Jakubowski, O. Schmitz, G. Wang, A.E. White, L. Zeng, G.R. McKee, C. Rost, J.R. Dorris Suppression of Type I ELMS with $n$=3 edge resonant magnetic perturbations (RMP) depends on reducing the pedestal pressure gradient below the peeling-ballooning mode stability limit. This pressure gradient reduction results from a reduction in pedestal particle density and effective particle confinement time $\tau _{p}$*. Recent experimental results suggest that this $\tau _{p}$* reduction arises from at least two mechanisms: increased ion-scale turbulence in the region 1$>$r/a$>$0.75, and improved coupling of the plasma to the pump due to strike point splitting. These mechanisms are observed to increase the density at the pump entrance, leading to improved pumping efficiency and lower $\tau _{p}$*. [Preview Abstract] |
|
UP8.00028: Modeling of Global Particle Balance in ELMing and RMP ELM-Suppressed DIII-D Discharges With SOLPS5-EIRENE S. Mordijck, I. Joseph, R.A. Moyer, G.R. Tynan, L.W. Owen, T.E. Evans, T.H. Osborne, X. Bonnin, D. Coster, D. Reiter The global particle balance in single-null DIII-D H-mode plasmas, with density control using the lower cyropump, is studied with a 2D fluid code coupled to a Monte-Carlo neutral code, SOLPS5-EIRENE. We compare DIII-D discharges with type-I ELMs to ELM-suppressed discharges using n=3 I-coil resonant magnetic perturbations (RMPs). We observe that RMP leads to a strong reduction in the density. This may result from an increase in stochastic particle advection, increased fluctuation driven transport or improved coupling of the plasma to the pump due to changes in the magnetic footprints. To identify the importance of each effect, we construct a grid and derive transport parameters to fit the radial profiles between ELMs in an ELMing discharge. We investigate the relative impact of changes in the pumping efficiency by comparing the numerical profiles to the experimental RMP profiles. [Preview Abstract] |
|
UP8.00029: Strike Point Splitting in RMP-ELM-free H-mode I. Joseph, R.A. Moyer, T.E. Evans, M.J. Schaffer, W.P. West, M. Jakubowski, A.M. Runov, R. Schneider, S.V. Kasilov, O. Schmitz, M.E. Fenstermacher, M. Groth, C.J. Lasnier, J.G. Watkins The E3D two-fluid code is used to model the effect of resonant magnetic perturbations on DIII-D thermal transport. The strike points are predicted to develop multiple striations determined by the invariant manifolds of the perturbed field, and the heat flux distribution is predicted to be well-correlated with the local connection length. Although filtered optical cameras observe striations in particle flux, the energy fluxes measured by infrared cameras and Langmuir probes do not appear to display significant splitting. This indicates that the perturbed field lines do not penetrate far inside the unperturbed separatrix. This is consistent with the fact that the predicted thermal transport is too large to match measured pedestal profiles. Both results may indicate that the rotational plasma response limits the stochastic field to a thin layer near the separatrix. [Preview Abstract] |
|
UP8.00030: Fast Imaging of ELM Structure and Dynamics in DIII-D J.H. Yu, J.A. Boedo, E.M. Hollmann, R.A. Moyer, D.L. Rudakov, P.B. Snyder Fast-framing images of CIII and D$_{\alpha }$ emission in the low-field-side (LFS) plasma boundary of DIII-D show that ELMs are helical filamentary structures that rotate toroidally. The filaments propagate radially outward at v$_{r}\sim $500 m/s during the nonlinear phase, and result in plasma-wall interactions that are poloidally localized within 15 cm of the midplane. The measured mean poloidal width of the filament is 3 cm, and the ELM toroidal mode number $n$ ranges from 10 to 35. ELM structure and dynamics vary with plasma density, possibly because ELMs are driven by a peeling type of mode in low density plasmas and are driven by a coupled peeling-ballooning mode in high density. At high collisionality ($\nu _{ped}$*=0.50), ELMs begin with an unstable filament or group of filaments at the LFS midplane region. Onset of the ELM-induced radiation in the divertor is delayed by as much as 0.8 ms compared to the midplane signals. In low collisionality ($\nu _{ped}$*=0.25) discharges, the midplane and divertor ELM signals appear simultaneously, possibly suggesting a more poloidally symmetric mode structure. [Preview Abstract] |
|
UP8.00031: ELM Triggering From Deuterium Pellets Injected into DIII-D L.R. Baylor, T.C. Jernigan, T.E. Evans, P.B. Parks, M.E. Fenstermacher, R.A. Moyer Deuterium fueling pellets have been injected into DIII-D plasmas from five different locations and under different plasma H-mode conditions. Edge localized modes (ELMs) have been triggered from pellets injected from all locations and under all the H-mode scenarios thus far explored. Pellets injected into plasmas with ELMs suppressed by a resonant magnetic perturbation are also observed to trigger one or more ELM like events. Experimental details of the pellet triggering of ELMs on DIII-D will be reviewed. In addition a pellet dropper has been installed on DIII-D for ELM pacing studies. Initial results from the slow 1mm pellets dropped into the edge plasma will be presented. [Preview Abstract] |
|
UP8.00032: Modeling of Perturbation Magnetic Field Based on Scrape-Off Layer Currents (SOLCs) During ELMs in DIII-D H. Takahashi, E.D. Fredrickson, M.J. Schaffer The poloidal perturbation magnetic field (B$_{\delta B})$ measured by the Mirnov diagnostic at the peak of ELMs in H-mode discharges on the DIII-D tokamak exhibits idiosyncratic features in its poloidal variation: the field peaks in the divertor, and is larger (``anti-ballooning'') and opposite in sign on the inboard side in comparison with the outboard side. A model, developed for calculating B$_{\delta B}$ consistent with measured scrape-off layer current (SOLC), reproduces these idiosyncratic features. A narrow magnetic structure was also observed by the Mirnov diagnostic at the outboard mid-plane in some ELMs, which appeared to be qualitatively similar to filaments reported earlier in DIII-D [1] and NSTX [2]. The SOLC-based model also reproduces such a filament structure. [1] E.J. Strait, et al., Phys. Plasmas \textbf{4}, 1783 (1997). [2] R. Maingi, et al., Phys. Plasmas \textbf{13}, 092510 (2006). [Preview Abstract] |
|
UP8.00033: Target Plate Profiles During ELM Suppression Experiments on DIII-D J.G. Watkins, T.E. Evans, C.J. Murphy, M.J. Martin, A. Nelson, M. Jakubowski, I. Joseph, R.A. Moyer, C.J. Lasnier, M.E. Fenstermacher Radial profiles of target plate plasma conditions during ELM suppressed conditions have been measured with the new DIII-D lower divertor Langmuir probe array. ELM suppression was accomplished using n=3 resonant magnetic perturbations [1]. Evidence of the n=3 mode structure of the perturbation can be seen most clearly in the V$_{f}$ profile on the target plate. The spacing of the multiple peaks in the V$_{f}$ profile is similar to predictions of the TRIP3D field line integration code. $T_{e}$ values $>$100~eV and V$_{f}$ values down to --150 V were measured. We observe resonant behavior of the target plate parameters near the q$_{95}$ value for maximum magnetic perturbation. Heat flux from the Langmuir probe measurements will be compared with infrared cameras and thermocouples. The resulting sheath power transmission factor profile will be shown. [1] T.E. Evans, \textit{et al., }Phys. Rev. Lett. \textbf{92}, 235003 (2004). [Preview Abstract] |
|
UP8.00034: Divertor Heat and Particle Fluxes During ELM Control Experiments O. Schmitz, M.W. Jakubowski, T.E. Evans, M.J. Schaffer, W.P. West, M.E. Fenstermacher, M. Groth, C.J. Lasnier, I. Joseph, R.A. Moyer, B. Unterberg, H. Frerichs In experiments exploring ELM suppression by resonant magnetic perturbation (RMP) as a technique for ITER, the manipulation of divertor heat and particle fluxes is of vital interest. To investigate these effects, a fast IR camera and CCD cameras equipped with D$_{\alpha }$, CII or CIII interference filters were used during RMP ELM control experiments at DIII-D. In general, a splitting of the inner and outer divertor strike lines was observed. This is caused by splitting of the invariant separatrix manifolds that form magnetic footprints on the wall elements. Parallel particle and heat fluxes are transported along these field lines forming a characteristic pattern on the divertor target. The measured patterns are compared to magnetic footprints modeled with the TRIP3D code to identify the topology of the heat and particle flux channels. Based on that, the occurrence of complete ELM suppression is correlated to the measured and modeled target patterns. [Preview Abstract] |
|
UP8.00035: Radiating Divertor Behavior in Single- and Double-Null Plasmas in DIII-D T.W. Petrie, N.H. Brooks, A.W. Hyatt, M.J. Schaffer, M.R. Wade, W.P. West, M.E. Fenstermacher, M. Groth, C.J. Lasnier, J.G. Watkins The ability to concentrate impurities in the divertor region to provide effective radiative divertor operation has been found on DIII-D to be sensitive to the divertor magnetic geometry and the grad-B drift direction. Argon impurities were injected into the private flux region of one divertor, while deuterium flow into the divertors was simultaneously enhanced by a combination of midplane gas puffing and divertor cryopumping. For DN plasmas it was difficult to balance the radiated power between divertors during argon injection; significant increases in radiated power and argon concentration were observed mostly in the divertor that was \textit{opposite} the grad-B drift direction. For SN plasmas, there was a higher divertor argon accumulation in the divertor when the grad-B drift direction was away from the dominant X-point, and so this setup may provide the best prospect of successfully coupling a radiating divertor approach to a high performance H-mode plasmas. [Preview Abstract] |
|
UP8.00036: The Influence of High Field Side Recycling and Impurity Sources on Divertor Detachment in Simulations of Ohmic Discharges of the ASDEX Upgrade and DIII-D Tokamaks M. Wischmeier, A. Kallenbach, A.V. Chankin, D. Coster, R. Dux, J. Harhausen, H.W. Mueller, M. Groth, X. Bonnin In the framework established under the International Tokamak Physics Activity (ITPA) Divertor and SOL working group a series of ohmic discharges with similar plasma parameters have been performed in ASDEX Upgrade and in DIII-D. The SOLPS code is tested against these experimental data at the onset of divertor detachment. The presentation focuses on identifying numerically the impact of divertor geometry, target material composition as well as high field side (HFS) recycling and impurity sources on the onset and degree of detachment along the inner target. Physics processes to explain the observed asymmetry between the inner and outer target ion fluxes with increasing line-averaged density are discussed. [Preview Abstract] |
|
UP8.00037: Measurement and Modeling of Carbon Flows in the Main SOL of the DIII-D Tokamak M. Groth, M.E. Fenstermacher, G.D. Porter, M.E. Rensink, T.D. Rognlien, J.A. Boedo, D.L. Rudakov, J.H. Yu, N.H. Brooks, R.C. Isler, J.G. Watkins Carbon ion velocities of $\sim $1/2 of the deuterium ion sound speed in the direction of the inner divertor were measured in the scrape-off layer (SOL) at the top of DIII-D low-density, low-confinement plasmas with the ion \textbf{B}x$\nabla $B drift toward the divertor. The use of upper single-null discharges and injection of methane from the bottom of the DIII-D vessel in toroidally symmetric fashion enabled simultaneous measurements of the flow of deuterium ions with a reciprocating Langmuir probe and of low charge-state carbon ions with a tangential viewing spectrometer and cameras. Modeling of the SOL with the UEDGE code predicts significantly weaker flow of deuterium ions in the SOL compared with the data, leading to an underestimate of the carbon flow in the main SOL. The dependences of the simulated flows on the assumed radial transport model and boundary conditions are investigated and will be presented. [Preview Abstract] |
|
UP8.00038: Direct Spectroscopic Measurement of Fast Ionic Flow in the Main SOL of Diverted Plasmas on the DIII-D Tokamak N.H. Brooks, W.P. West, R.C. Isler, M. Groth, J.A. Boedo Flow velocities in the SOL at the crown of the main plasma have been determined by direct measurement of Doppler shifts in the spectral line profiles of C II and C III. In single-null discharges with \textbf{B}x$\nabla $B drift into the divertor, flow toward the inner divertor is found, with C II and C III velocities of $\sim $8x10$^{5}$ and $\sim $1.5x10$^{6~}$cm/s, respectively. The magnetic field value deduced from Zeeman splitting of these spectral lines is used to spatially localize their emissions to flux surfaces outside the separatrix, with normalized $\psi $-values of 1.04 and 1.045, respectively. The spectroscopically deduced C III velocity is consistent with reciprocating probe measurements of fuel ion Mach number assuming full entrainment of the doubly ionized carbon ion on the flux surface where its emission is localized. TV images of the breakup fragments from injected methane gas provide additional confirmation of the SOL impurity flow. [Preview Abstract] |
|
UP8.00039: Indications of an Inward Pinch in the Inner SOL of DIII-D From $^{13}$C Deposition Experiments J.D. Elder, P.C. Stangeby, A.G. McLean, S.L. Allen, M.E. Fenstermacher, M. Groth, J.A. Boedo, D.L. Rudakov, B.D. Bray, N.H. Brooks, A.W. Leonard, W.P. West, W.R. Wampler, J.G. Watkins, D.G. Whyte $^{13}$C methane puffing experiments were conducted on DIII-D in both L- and edge localized mode H-mode conditions. The puffing was toroidally symmetric into the crown of a series of well-characterized LSN discharges in which the plasma conditions, carbon emissions, core carbon increment and $^{13}$C deposition pattern were measured. The hydrocarbon breakup and subsequent carbon transport were modeled using the OEDGE interpretive code. The carbon is followed in OEDGE until it deposits on surfaces. Replicating the experimental deposition requires assuming in the code both fast parallel transport as well as a pinch/drift in the inner scrape-off layer towards the separatrix of 10 to 20 m/s in both L and H mode. Radial variation of the fast parallel flow and re-erosion of carbon are investigated as alternative hypotheses to a pinch. [Preview Abstract] |
|
UP8.00040: Tritium Recovery From Carbon Co-deposits: \textit{ex situ} Studies in Preparation for an \textit{in situ} Thermal Oxidation Experiment in DIII-D C. Tsui, J.W. Davis, A.A. Haasz, B.W.N. Fitzpatrick, A.G. McLean, P.C. Stangeby, Y. Mu, S.L. Allen, W.P. West, P.L. Taylor, K.L. Holtrop, A.W. Hyatt, R.L. Boivin, K. Umstadter Tritium trapping in carbon co-deposits is potentially a major cause of T retention in ITER. A possible solution is ``thermo-oxidation,'' involving sub-atmospheric molecular O$_{2}$ at 250$^{^{o}}$-350$^{^{o}}$C. Ex situ lab tests (at U. Toronto) of graphite tiles removed from DIII-D have demonstrated the method in principle and generated a comprehensive database on removal rates. Ex situ tests were also performed for special components (e.g., diagnostic mirrors) used in DIII-D to assess potential damage. The next step is in situ demonstration in DIII-D to show that the D (as proxy for T) and C removed from the tiles actually leaves the vessel, that vessel components are not damaged, and that plasma operation can be recovered quickly. Lab results and implications for the in situ test are discussed. [Preview Abstract] |
|
UP8.00041: Results From the Upgraded Porous Plug Injection System for Studies of Hydrocarbon Dissociation and Transport in DIII-D A.G. McLean, J.W. Davis, Y. Mu, P.C. Stangeby, S.L. Allen, R. Ellis, M.E. Fenstermacher, M. Groth, C.J. Lasnier, B.D. Bray, N.H. Brooks, T.W. Petrie, W.P. West, C.P.C. Wong, D.G. Whyte, J.A. Boedo, E.M. Hollmann, D. Nishijima, D.L. Rudakov, R.J. Colchin, R.C. Isler, J.G. Watkins, S. Brezinsek, M. Jakubowski, A. Krater Calibrated spectroscopic measurements of dissociation fragments resulting from hydrocarbon influx in a tokamak divertor have been taken with the use of the Mk II Porous Plug Injector (PPI) in DIII-D. The PPI was upgraded to include a small orifice flow restrictor for more precise gas flow control, achieving flow rates corresponding to sputtering yields of 0.5{\%}-2{\%} in attached and detached divertor plasmas. Results and analysis of digital video, and medium and high resolution spectroscopic data collected are presented and compared with those of the Mk I PPI and similar experiments at JET, TEXTOR, and ASDEX. [Preview Abstract] |
|
UP8.00042: Dust Particles Observed by Laser Scattering at DIII-D B.D. Bray, W.P. West, D.L. Rudakov Studies of dust particles observed by Rayleigh/Mie scattering of ND:YAG lasers during plasma operations at DIII-D show correlations with plasma configuration. Dust particles are primarily observed outside the last closed flux surface of the plasma. The mean particle density has been observed to be near 400 m$^{-3}$ in both divertor scrape-off layer regions, corresponding to an upper or lower single-null configuration. The inferred particle size ($\sim $100 nm) indicates this represents a small carbon density relative to measured ionized carbon density in the plasma and consequently not believed to be a significant source of impurities. However, understanding the dust dynamics remains important because of its safety implications in future burning plasma reactors. In addition, in DIII-D, the dust density varies with the phase of the plasma discharge and plasma parameters. The dust density is roughly twice as large in ELMing H-mode discharges compared to QH- or L-mode. [Preview Abstract] |
|
UP8.00043: Inference of Thermal and Momentum Transport Coefficients in the DIII-D Edge Pedestal W.M. Stacey, R.J. Groebner New methodologies for inferring thermal and momentum transport coefficients from measured temperature, density and velocity profiles in the edge plasmas have been developed. The thermal interference methodology, which takes into account radiation and atomic physics cooling, recycling neutrals and convection, in the determination of conductive heat flux profiles, has been applied to the ELM-free phase of a discharge [1]. Now, time-resolved profile data sets averaged over similar inter-ELM intervals have been developed for two discharges. The methodology has been extended to take into account the measured reheating and density buildup in the pedestal between ELMs, and ion and electron thermal diffusivities have been inferred for the various sub-intervals (e.g. 10{\%}-20{\%}, 80{\%}-90{\%}) in the inter-ELM interval. An analogous methodology has been developed for inferring toroidal angular momentum and poloidal momentum transport rates from measured toroidal and poloidal velocity measurements, respectively, and applied to the same two data sets. \newline \newline [1] W.M. Stacey and R.J. Groebner, Phys. Plasma \textbf{13}, 072510 (2007). [Preview Abstract] |
|
UP8.00044: Dependence of H-mode Power Threshold on Input Torque and Toroidal Plasma Rotation in the DIII-D Tokamak P. Gohil, G.R. McKee, D.J. Schlossberg, G. Wang The power required to induce the L-H transition is dependent on the applied beam torque. For upper single-null discharges in which the ion $\nabla $B drift is away from the X-point, the L-H transition power threshold is reduced by up to a factor of 3 by changing from predominantly co (4-6 MW) to predominantly counter-injection ($<$2~MW). Lowered L-H transition power thresholds are also observed with reduced input torque in discharges with the ion $\nabla $B drift towards the X-point, but to a lesser degree. For the first time, an H-mode transition was induced by slowly reducing the input torque at constant input power by slowly varying the mix of co- and counter-beam injection. The mechanisms for such a torque dependence are being investigated from analyses of the edge plasma rotation, the edge radial electric field and the edge plasma turbulence. Preliminary results indicate large changes in the poloidal velocity shear of the edge turbulent eddies prior to the L-H transition that may be strong enough to induce the transition. [Preview Abstract] |
|
UP8.00045: H-Mode Pedestal Toroidal Velocity in DIII-D J.S. deGrassie, R.J. Groebner, K.H. Burrell, W.M. Solomon Tokamak discharges have nonzero toroidal momentum in the absence of any auxiliary injected torque [1]. It is important to understand this ``intrinsic rotation'' in order to extrapolate to burning plasmas. The interface between the plasma and the boundary can possibly provide an edge momentum flux, or nonzero velocity boundary condition. In order to see if it is possible to extract any boundary condition we have looked at the DIII-D database of measurements of toroidal and poloidal velocity in the region of the edge density pedestal in H-mode discharges, including Ohmic and ECH H-modes with no auxiliary torque, discharges with rotation driven by uni-directed neutral beam injection (NBI), and discharges with near-balanced NBI, which portend to allow measurements of intrinsic rotation at high auxiliary power levels. The consistency of an NBI momentum flux through the pedestal vs intrinsic rotation will be evaluated. Comparison will be made with applicable neoclassical and turbulence theories of intrinsic rotation. [1] J.S. deGrassie, \textit{et al., }Phys. Plasma \textbf{14}, 056115-1 (2007). [Preview Abstract] |
|
UP8.00046: Effects of Toroidal Rotation on Edge Turbulence and the L-H Power Threshold D.J. Schlossberg, G.R. McKee, M.W. Shafer, K.H. Burrell, P. Gohil, R.J. Groebner, T.C. Luce, G. Wang Edge turbulence dynamics, flows, and flow shear are found to depend strongly on the injected neutral beam torque in DIII-D plasmas. Likewise, the power threshold required to induce a transition from low- to high-confinement mode decreases by a factor of 2-3 as torque is varied from the co to counter current directions. Turbulence characteristics such as the poloidal shearing rates, correlation rates, and decorrelation times in the edge region are examined with the high-sensitivity 2D beam emission spectroscopy diagnostic on DIII-D. Poloidal flow shear in the turbulence is found to increase in all cases as the transition is approached. As the injected torque is varied from co-current to balanced, a bi-modal structure and strongly dispersive turbulence spectrum develops. At low-rotation this bi-modal structure consists of oppositely propagating flows that lead to flow shear rates increasing above the calculated decorrelation rates. [Preview Abstract] |
|
UP8.00047: Toroidal Flow in Tokamak Plasmas J.D. Callen, A.J. Cole, C.C. Hegna Many effects influence toroidal flow evolution in tokamak plasmas. Momentum sources and radial diffusion due to axisymmetric neoclassical, paleoclassical and anomalous transport are usually considered. In addition, the toroidal flow can be affected by field errors. Small, non-axisymmetric field errors arise from coil irregularities, active control coils and collective plasma magnetic distortions (e.g., NTMs, RWMs). Resonant field errors cause localized electromagnetic torques near rational surfaces in the plasma, which can lock the plasma to the wall leading to magnetic islands and reduced confinement or disruptions. Their penetration into the plasma is limited by flow-shielding effects; but they can be amplified by the plasma response at high beta. Non-resonant field errors cause magnetic pumping and radial banana drifts, and lead to toroidal flow damping over the entire plasma. Many of these processes can also produce momentum pinch and intrinsic flow effects. This poster will seek to present a coherent picture of all these effects and suggest ways they could be tested and distinguished experimentally. [Preview Abstract] |
|
UP8.00048: Neoclassical Toroidal Viscosity for Low-Density Ohmic Plasmas in DIII-D A.J. Cole, C.C. Hegna, J.D. Callen, M.J. Schaffer, R.J. La Haye A recent model [1] for field error penetration that includes resonant and non-resonant perturbed 3D magnetic fields has for the first time obtained quantitative agreement with empirical scaling studies of the error-field penetration threshold with electron density. Relevance of the new model relies on the error-field induced neoclassical toroidal viscosity (NTV) being comparable to cross-field diffusive viscosity near a resonant surface of interest. The strength and harmonic structure of NTV for low-density ohmic plasmas on DIII-D are determined from intrinsic vacuum error-field data. Preliminary analysis has shown that NTV in DIII-D is dominated by non-resonant modes. We neglect the plasma response in this initial investigation. An effective cross-field momentum transport owing to NTV is determined, for future comparison with possible cross-field momentum transport rates in ohmic discharges. \newline [1] A.J. Cole, \textit{et al., }``Effect of Neoclassical Toroidal Viscosity on Error-Field Penetration Thresholds in Tokamak Plasmas,'' to be published in Phys. Rev. Lett. (2007). [Preview Abstract] |
|
UP8.00049: Gyrokinetic Theory and Simulation of Angular Momentum Transport R.E. Waltz, G.M. Staebler, J. Candy, F.L. Hinton A gyrokinetic theory of turbulent toroidal angular momentum transport as well as modifications to neoclassical poloidal rotation from turbulence is formulated starting from the fundamental six-dimensional kinetic equation. GyroBohm-scaled transport is evaluated from toroidal gyrokinetic simulations using the GYRO code [1]. The simulations quantify the two pinch mechanisms in the radial transport of toroidal angular momentum: the slab geometry ExB shear pinch [2] and the toroidal geometry ``coriolis'' pinch due to finite parallel velocity [3]. The pinches allow the steady-state null stress (momentum transport) condition required for intrinsic toroidal rotation in heated tokamaks without an internal source of torque [4]. A predicted turbulent shift in the neoclassical poloidal rotation [5] may be significant. \newline [1] J. Candy and R.E. Waltz, J. Comp. Phys. \textbf{186}, 545 (2003). \newline [2] R.R. Dominguez and G.M. Staebler, Phys. Fluids \textbf{B5}, 387 (1993). \newline [3] A.G. Peeters, et al., Phys. Rev. Lett. \textbf{98}, 26503 (2007). \newline [4] G.M. Staebler, et al., Bull. Am. Phys. Soc. \textbf{46}, 221 (2001). \newline [5] G.M. Staebler, Phys. Plasmas \textbf{11}, 1064 (2004). [Preview Abstract] |
|
UP8.00050: A New Formulation of the Quasi-linear Transport Theory for the Trapped Gyro Landau Fluid Model G.M. Staebler, J.E. Kinsey, R.E. Waltz A quasi-linear model of turbulence evaluates the bilinear fluctuation driven fluxes using linearly unstable eigenmodes. A model for the saturated amplitude of the turbulence is used to complete the flux calculation. It is traditional to normalized the fluxes by the amplitude of the electrostatic potential (phi-norm) induced by the linear eigenmode. This is the formulation used in the GLF23 model and in the recently published Trapped Gyro-Landau Fluid (TGLF) model [1]. The normalization of the fluxes is not unique. A variety of alternate choices for normalization have been tested with a large database ($>$150 runs) of non-linear gyrokinetic simulations using the GYRO code. It was found that several alternate choices give much better fits to this database than the phi-norm for simple mixing length saturation formulas. Using the modulus of the whole linear eigenvector for the TGLF equations as the norm (v-norm) gives a nearly optimum fit to GYRO. This v-norm version of the model fits both s-alpha and shaped Miller geometries \newline [1] G.M. Staebler, \textit{et al., }Phys. Plasma \textbf{14}, 055909 (2007). [Preview Abstract] |
|
UP8.00051: Towards a Predictive Pedestal Height Model P.B. Snyder, A.W. Leonard, T.H. Osborne, H.R. Wilson The pressure at the top of the edge transport barrier (or ``pedestal height'') strongly impacts tokamak fusion performance, and first principles prediction of the pedestal height remains an important challenge. While uncertainties remain, MHD stability calculations, accounting for diamagnetic stabilization, have been largely successful in predicting the observed pedestal height, when the barrier width is taken as an input. Studies of the pedestal dependence on power input support this understanding, while providing insight into the mechanisms responsible for setting the edge barrier width. Here, we present and characterize edge MHD stability results, including an updated model of diamagnetic stabilization, and a discussion of the mechanisms which can lead to an apparent power dependence of the pedestal. In addition, we explore initial, simple models for the barrier width. The barrier width models, as well as a combination of the width model with MHD stability calculation to directly predict pedestal height, are compared with an extensive set of observations on the DIII-D tokamak. [Preview Abstract] |
|
UP8.00052: Modeling and Analysis of Phase Contrast Imaging Measurements J.C. Rost, M. Porkolab, J.R. Dorris, J. Candy, K.H. Burrell The phase contrast imaging (PCI) diagnostic on DIII-D has been operated in several configurations over its lifetime. The beam path was changed in 2003 from tangential at the midplane LCFS to a path passing through the edge at an angle near 45 degrees and reaching typically r/a=0.8, and the maximum wavenumber has been increased from 7 to 30 cm$^{-1}$. A synthetic diagnostic (SD) has been created to model all configurations of the PCI by post-processing the output of the GYRO gyrokinetic simulation. The SD includes line integration along the full path and models the detector to obtain the high- and low-$k$ cutoffs. Modeling of a plasma discharge typical of DIII-D is used to interpret the PCI spectra $S(k_{\bot }$,$f)$ in terms of turbulent ballooning modes and local $S(k_{r}$,$k_{\theta }$,$f)$. This allows us to identify parts of the PCI spectra with different plasma modes (ITG, TEM, ETG), separate effects of Doppler shift and intrinsic mode velocity in the measurement, and improve comparisons with other diagnostics. The SD will contribute to validation of the model through comparison between simulation and experiment. [Preview Abstract] |
|
UP8.00053: Development of a Synthetic BES Diagnostic for Application to Gyrokinetic Simulations C. Holland, G.R. Tynan, G.R. McKee, M.W. Shafer, J. Candy, R.E. Waltz, R.V. Bravenec The validation of microturbulence simulations requires the use of synthetic diagnostics to allow ``apples-to-apples'' comparisons of predicted and measured turbulence characteristics. We report here on progress in the development of a synthetic beam emission spectroscopy (BES) diagnostic for use with the GYRO code, and results from its application to simulations of a steady L-mode DIII-D discharge. Comparisons of simulated and experimentally measured fluctuation amplitudes, frequency spectra, correlation times and lengths will be presented, as well as a comparison of zonal flow fields obtained by time-delay estimation. The impact of increased channel spot size due to finite neutral beam atom excitation lifetimes, and sensitivity of the results to uncertainties in equilibrium profile gradients is also discussed. [Preview Abstract] |
|
UP8.00054: Localized Measurement of Short Wavelength Plasma Fluctuations Using the DIII-D Phase Contrast Imaging Diagnostic J.R. Dorris, J.C. Rost, M. Porkolab, K.H. Burrell The DIII-D phase contrast imaging (PCI) turbulence diagnostic measures density fluctuations in two operational configurations: (1)~line-integrated over the entire viewing chord, or (2) using a rotating mask system that takes advantage of the vertical variation of the radial magnetic field to make localized measurements along the PCI chord. The localized length of chord is inversely proportional to wavenumber, making this technique more favorable for short wavelength modes ($k$~$>$~15/cm). In 2006, the PCI S/N was improved by an order of magnitude by a redesigned data transmission system based on fiber optic links. This allowed measurements to be obtained showing broadband turbulent fluctuations to 20/cm. Rigorous analysis of such a localized measurement requires accurate modeling of the focused laser diffraction through the thin masking slit. Recent calibrations successfully validated our model and results will be presented. Measured evolution of turbulence will be characterized in terms of changes in global plasma parameters. [Preview Abstract] |
|
UP8.00055: Time-dependent Radial Transport of Electron Distributions Due to ECCD in DIII-D R.W. Harvey, A.P. Smirnov, R. Prater, C.C. Petty The radial transport modeling capability in the CQL3D bounce-averaged Fokker-Planck collisional-rf quasilinear code[1] has been greatly improved and the self-consistent time-dependent toroidal electric field added, making the code truly a ``Fokker-Planck-Transport'' code. The time-dependent, coupled 3D Fokker-Planck equation and the Ampere-Faraday Law equation are solved for the electron distribution, $f\left( {u,\theta _u ,\rho ,t} \right)$, and the toroidal loop voltage, $V_{loop} \left( {\rho ,t} \right)$. A fully 3D, time-implicit solution of the FP equation using sparse-matrix methods[2] is coupled to a new iterative toroidal electric field solve. The DIII-D ECH experiment is in an intermediate driven regime with $\tau _{transport} \approx \tau _{slowing} $[3] for the EC driven electrons. Results will be reported for time-evolution of radial profiles of current density, fast electrons, and toroidal loop voltage due to EC heating and current drive in DIII-D. [1] R.W. Harvey and M.G. McCoy, IAEA TCM on Advances in Simulation and Modeling of Thermonuclear Plasmas, Montreal, 1992; USDOC NTIS No. 93002962. [2] Y. Peysson \textit{et al}., Radio Frequency Power in Plasmas, 15th Topical Conference, Moran, Wyoming (2003). [3] R.W. Harvey \textit{et al.}, Phys. Rev. Lett. 88, 205001 (2002). [Preview Abstract] |
|
UP8.00056: Doppler Reflectometry Measurements of Medium Wavenumber Density Fluctuations and Zonal Flows in DIII-D L. Schmitz, G. Wang, A.E. White, J. Justiniano, T.L. Rhodes, W.A. Peebles Doppler reflectometry is a versatile diagnostic for poloidal plasma flow measurements and local density fluctuation spectra. Depending on the launch angle and frequency of the probing beam, the signal back-scattered from the plasma cut-off layer is sensitive to density fluctuations at a specific poloidal wavenumber $k_{\theta }$ (3 cm$^{-1}$~$<~k_{\_}$~$<$ 9 cm$^{-1}$, calculated using GENRAY ray tracing code). The plasma flow velocity $v_{\theta }$ is obtained with high time resolution from the Doppler shift \textit{$\omega $}$_{D}$ of the back-scattered signal ($v_{\theta }$\textit{ = $\omega $}$_{D}/k_{\theta }).$ Doppler reflectometry is well-suited for the detection of zonal flows, characterized by poloidal flow fluctuations ($v_{\theta }~=$~$k_{r}\Phi $/$B_{\phi })$. Zonal flows are thought to regulate the local turbulence level and radial correlation. We present first reflectometry measurements of geodesic acoustic modes (GAMs) and low frequency zonal flows in DIII-D L{\-}mode plasmas (0.6~$<$~$r/a$~$<$~0.9). The interaction of these time-dependent plasma flows with medium wavenumber density fluctuations is investigated in order to study turbulence self-organization. [Preview Abstract] |
|
UP8.00057: Radial Correlation Length of Density Fluctuations in DIII-D Plasmas G. Wang, W.A. Peebles, T.L. Rhodes, E.J. Doyle, L. Schmitz, A.E. White, L. Zeng, R. Nazikian, G.R. McKee The radial correlation length (L$_{r})$ of density fluctuations, which is directly related to the radial scale length of the fluctuations, is an important quantity for understanding turbulent transport in tokamak plasmas. In DIII-D, recent upgrades of the UCLA correlation reflectometer system allow high spatial resolution L$_{r}$ measurement in both low and high density gradient regions. Initial observations is presented, including: (1) In the OH plasma core, L$_{r }$decreases as major radius increases, but inferred radial fluctuation scale length $k_{r}$\textit{$\rho $}$_{s}$ (\textit{$\rho $}$_{s}$ is the ion gyroradius using T$_{e}$ and $k_{r}=2/L_{r}$ for a Gaussian radial correlation length) increases outwards; (2) In the plasma core, $k_{r}$\textit{$\rho $}$_{s}$ of the L-mode, QH-mode, and hybrid steady-state H-mode are comparable, and more than 2-3 times bigger than those in OH plasmas; (3) In both L- and QH-mode core, little change was observed in L$_{r}$ as plasma rotation is varied significantly by NB torque change. Initial results from H-mode edge pedestal will also be presented. [Preview Abstract] |
|
UP8.00058: ITG and TEM Turbulence in DIII-D L-mode Discharges J.C. DeBoo, C.C. Petty, G.M. Staebler, E.J. Doyle, T.L. Rhodes, L. Schmitz, G. Wang, A.E. White, G.R. McKee An experiment has been designed to discriminate between the effects of ion temperature gradient (ITG) and trapped electron mode (TEM) turbulence by creating discharges where one of these modes is clearly dominant. With the aid of the GKS and TGLF linear gyrokinetic stability codes, a low-density L-mode target discharge with electron cyclotron heating has been identified where TEM modes are calculated to dominate. By replacing electron cyclotron heating power with neutral beam power, ITG modes are expected to become dominant. The TEM threshold condition will also be tested by varying the local electron temperature gradient scale length to values above and below the threshold condition by employing modulated ECH near the plasma mid-radius. Results of the experiment, including comparison of turbulence measurements at low (ITG) and intermediate (TEM) wavenumbers with gyrokinetic stability code predictions of drift wave spectra and threshold predictions, will be shown. [Preview Abstract] |
|
UP8.00059: Rational-q Triggered Transport Changes With Varying Toroidal Rotation in DIII-D M.E. Austin, K.H. Burrell, R.E. Waltz, M.A. Van Zeeland, G.R. McKee, M.W. Shafer, T.L. Rhodes Comparison of rational-$q$ triggered ITBs in discharges with varying toroidal torque injection was carried out. Experiments were conducted in negative central shear discharges with different mixes of co/counter neutral beam injection (NBI) that altered the equilibrium ExB shear in conditions where transient improvements in transport occur near integer $q_{min}$ values. The transport changes were seen in high and low rotation cases; however, the latter discharges did not transition to improved core confinement. Observations support the model that sufficient background ExB shear is required for barrier formation and zonal flow effects at integer $q_{min}$ act as trigger in this case. The lack of TAE modes in the balanced injection cases indicates they are not linked to the transient confinement improvement. Fluctuation data obtained in co and balanced NBI show similar reductions in turbulence near integer $q_{min}$ as well as poloidal velocity excursions that may be further evidence of zonal flow. [Preview Abstract] |
|
UP8.00060: TURBULENCE AND TRANSPORT |
|
UP8.00061: Density profile behavior in JET and extrapolations to ITER Mikhail Maslov, Henri Weisen, Clemente Angioni A growing number of experiments on different tokamaks show that the neoclassical pinch and fuelling by edge neutrals or NBI is generally not able to explain the observed density profiles. Anomalous inward convection appears to be ubiquitous, leading to peaked density profiles at low collisionality in H-modes. Extrapolation to ITER conditions leads to high density peaking for the reference scenario which increases expected fusion gain from Q=10 to Q=30 [1,2]. A comprehensive study of JET experiments of C15-C19 campaigns (2006-2007) was done. Several hundred of density profiles in stationary H-mode phases were analyzed and the scaling law for the density peaking was derived. General agreement with the previous observations [1] was found as well as some new features arise. Effective collisionality still has the highest correlation with density peaking but on the new data one can see a dramatic change of the density profile behavior around $\nu _{eff}\sim $0.4. Density peaking increases much faster at lower collisionalities and li correlation appears also. [1] H. Weisen et al. 21st IAEA FEC, Chengdu, China [2] V. Mukhovatov et al. Nucl. Fus.43 (2003)942 [Preview Abstract] |
|
UP8.00062: Characteristics of Zonal Flows in the HL-2A and HT-7 tokamaks T. Lan, A.D. Liu, C.X. Yu, L.W. Yan, W.Y. Hong, K.J. Zhao, J.Q. Dong, J. Qian, J. Cheng, D.L. Yu, Q.W. Yang Zonal Flows, including low frequency Zonal Flow (LFZF) and geodesic acoustic mode (GAM), have been characterized in the HL-2A and HT-7 tokamaks. The three-dimension wavenumber and frequency spectrum for the GAM has been measured in the HL-2A tokamak for the first time. The poloidal and toroidal wave number spectra are peaked sharply at$k_\theta =k_\phi \simeq 0$, while the radial spectrum shows a strong peak at $k_r \rho _i \approx 0.04-0.09$ with a width$\Delta (k_r \rho _i )\approx 0.03-0.07$. The whole spectral characters of LFZF in both tokamaks are firstly presented here. Bispectral analysis, nonlinear energy transfer and experimental studies on the GAM interaction with turbulence reveal that the parametric instability is a mechanism contributing to the generating of the GAM. Wavenumber mismatch estimation, denoting the coupling of nonlinear three-wave interactions, shows that the resonance conditions are satisfied exactly at the region with highest auto-bicoherency. The turbulent transport related to GAM has been studied in HL-2A tokamak and the results are reported. [Preview Abstract] |
|
UP8.00063: Improved core confinement in JET plasmas close to q=1 and possible alternative explanations Flavio Crisanti Associazione Euratom/ENEA sulla Fusione, Frascati, C.P. 65, 00044 Frascati, Italy In some experiments [e.g. 1,2] internal transport barriers are seen in situations where the usual theoretical pictures do not provide an obvious explanation, e.g. the hybrid regime, with q0 close to 1, or even in the presence of sawteeth. In some cases these events are transient, only affecting the ion channel. However, in some JET experiments enhanced temperature gradients have been observed on both electron and ions lasting about 5s (over 20 energy confinement times and about 1 resistive time). Up to 25{\%} of the ion population were fast particles, providing up to 40{\%} of the plasma stored energy. Similar discharges with higher density did not show the same behavior. The fast ion population, which is sensitive to the density, can stabilize ITG modes [2]. The JET experimental evidence will be analyzed within this theoretical frame and the results compared with other theoretical approaches. \newline *See the Appendix of M L Watkins \textit{et al}., Fusion Energy 2006 (Proc. 21$^{st}$ Int. Conf. Chengdu, 2006) IAEA, (2006) \newline \newline [1] F Crisanti\textit{ et al}, \textit{ibid} \newline [2] G Tardini \textit{et al.}, Nucl. Fusion \textbf{47} (2007) 280 [Preview Abstract] |
|
UP8.00064: Effects of kinetic electrons on Alfvenic ITG in global electromagnetic gyrokinetic particle simulations Y. Nishimura, Z. Lin, L. Chen Employing an electromagnetic gyrokinetic simulation model\footnote{Z.~Lin and L.~Chen, Phys. Plasmas {\bf 8}, 1447 (2001).} in a global tokamak geometry,\footnote{Y.~Nishimura, Z.~Lin, and W.~X.~Wang, Phys. Plasmas {\bf 14}, 042503 (2007).} trapped electron effects on the finite beta ITG modes are investigated. With the fluid-kinetic hybrid electron model,$^1$ non-adiabatic kinetic electrons are perturbatively added on top of the adiabatic fluid electrons. Details on the formulation and the implementation of the electron drift kinetic equation is discussed. This work is supported by U.S. Department of Energy, Cooperative Agreement DE-FC02-04ER54796 and 06ER54860, and in part by SciDAC Center for Gyrokinetic Particle Simulation of Turbulent Transport in Burning Plasmas and Center for Plasma Edge Simulation. [Preview Abstract] |
|
UP8.00065: Determination of fractional transport exponents in a simple fluid drift-wave turbulence model Debasmita Samaddar, D.E. Newman, R. Sanchez, B.A. Carreras In this poster, the recently developed nonlocal (quasi-linear) renormalization scheme to derive renormalized transport equations for passive scalars in terms of fractional differential operators will be used to explore transport in a simple drift wave model. In this contribution, we use this new method to determine the existence of fractional exponents in simulations of drift-wave turbulence in slab geometry and discuss the merits and disadvantages of the method with respect to the average propagator method. Several driven and non-driven situations will be explored, in which the relative dominance of the polarization and ExB nolinearities will be tuned artificially. In this way, we also test the robustness of the fractional transport models to changes in the basic dynamics, which will help to assess the potential for application to more realistic geometries of these methods. [Preview Abstract] |
|
UP8.00066: Characterization of transport dynamics from the self-consistent interaction between fluctuations and zonal flows in ITG gyro-kinetic simulations with the UCAN code D.E. Newman, R. Sanchez, J.N. Leboeuf, V.K. Decyk , B.A. Carreras In this poster, we will describe the application of several tools imported from the theory of non-Markovian, non-local stochastic processes to the characterization of the transport dynamics that emerge from the self-consistent interaction of fluctuations and zonal flows in ion-temperature-gradient (ITG) turbulence. The simulations have been performed using the delta-f, PIC gyrokinetic UCAN code. In order to fully understand the implications of the analysis, the results from the self-consistent case will be carefully compared with two additional ITG simulations performed with UCAN using identical parameters. First, one in which the feedback action on the fluctuations carried out by the zonal flows is artificially suppressed. Secondly, a case in which in addition to suppressing the zonal flows, an externally driven flows interacts with the fluctuations. [Preview Abstract] |
|
UP8.00067: Particle characterization of transport in global gyrokinetic calculations of ion channel turbulence in tokamak plasmas Jean-Noel Leboeuf, Benjamin Carreras, Viktor Decyk, David Newman, Raul Sanchez We are in the process of characterizing transport in gyrokinetic calculations of ion channel turbulence in tokamak plasmas with the three-dimensional global toroidal nonlinear parallel particle-in-cell UCAN code. In particular, we have extended the particle manager in UCLA's own PLIB library of massively parallel particle and field managing MPI routines to automatically handle tracking/tracing of the same active simulation particles through space and time and especially multiple processors. The particle data thus tracked and stored comprise the complete set of positions and velocities for each tracked particle at each chosen instant of time (typically every 100th time step). These particle data have been analyzed with tools previously applied to passive marker particles in fluid turbulence simulations which are specifically aimed at revealing the non-diffusive aspects of particle and heat transport. The transport characteristics from UCAN calculations without and with zonal flows self-consistently generated from the fluctuations allowed to evolve will be presented. [Preview Abstract] |
|
UP8.00068: A turbulent dynamo: Generation of zonal magnetic fields by finite beta drift-ballooning modes in tokamak plasmas Robert Kleva, Parvez Guzdar The generation of zonal flows and zonal magnetic fields can play a significant role in regulating the transport in the edge region of tokamaks. In recent years the focus has been on zonal flows since they are believed to be responsible for initiating good confined modes observed in a variety of magnetic confinement devices. Here we address the generation and saturation of zonal magnetic fields by numerically solving a set of reduced Braginskii equations in a flux-tube geometry. It is shown that the dynamo action leads to the generation of these fields which yields a time-dependent saturated state in which the turbulent drive balances the classical resistivity. The magnitude of the zonal field is however quite small compared to the zonal flow. We will make comparisons from our simulations with recent measurements of zonal fields. [Preview Abstract] |
|
UP8.00069: Electron transport analysis in TCV Wendell Horton, Juhyung Kim, Elina Asp, L. Porte We have investigated the turbulent electron transport in the four current H-mode phases of the TCV discharge 29892 with high-power ECH heating. On ion inertial length scale, we break down the dynamics into collisionless wave, collisional drift wave and trapped electron mode(TEM) (Horton, Phys. Fluids 19, 711, 1976). The transition from drift wave to trapped electron mode is observed in our calculations, and the electron temperature gradient destabilizes the TEM. We find that at the mid-radius, the TEM growth rate are strongly dependent on the collisionality whereas at the outer region, no collisionality dependence is observed. we also analyze the ETG transport with well-known theory based $\chi_e$-models. Finally we show quasi-2D pseudo spectrum simulations for several $(r,t)$ points and time slices for the TEM model and the ETG model. Comparison suggests that the ETG mode is a better candidate for the electron transport in the TCV discharge. [Preview Abstract] |
|
UP8.00070: Drift-Waves and Stability in the GAMMA-10 J. Pratt, W. Horton The tandem mirror system has achieved high energy confinement times ($70-90$ ms) and radial-loss times that dominate the Pastukhov end-loss time ($>100$ ms). This high confinement regime establishes a proof of principle that the combination of electrostatic and mirror confinement can successfully insulate electrons from thermal end losses. For the first time, the stored plasma energy of the ions within the plug-barrier end cells exceeds that of the central-cell magnetically-trapped ions. Tandem mirrors exhibit a qualitatively different type of drift-wave transport than do toroidal devices, as we have shown by developing confinement time scaling predictions (J. Pratt and W. Horton, Phys. Plasmas (13), 2006). We analyze electrostatic drift-wave eigenmodes for the electrostatic potential and the magnetic perturbation in the GAMMA-10. Using teraFLOPS-speed, large-scale parallel computers, we then integrate particle orbits in these eigenmode fields. [Preview Abstract] |
|
UP8.00071: Pinch effect and chaotic motion in toroidal confinement devices G. Spizzo, R.B. White, S. Cappello Particle transport in a toroidal plasma confinement device can be non-diffusive when magnetic chaos is present but the system is not too far above the stochastic threshold. In some conditions a phenomenological fit to density and impurity profiles gives a diffusion coefficient and also a pinch effect\footnote{X. Garbet, Phys. Rev. Lett. \textbf{91}, 035001 (2003), and references therein.}. We show that the combination of diffusion and pinch is an expression of the subdiffusive and nonlocal nature of the transport, brought about by the existence of a spectrum of long distance L\'{e}vy flights. The effect is illustrated by numerical modelling of magnetic structure and particle transport in conditions relevant for the reversed-field pinch experiment at the Consorzio RFX. Simulations consist of guiding center calculations of particle motion in the spectrum of MHD modes given by the 3D code SpeCyl\footnote{S.Cappello and D.Biskamp, Nucl. Fusion \textbf{36}, 571 (1996).}, and in integration of the Montroll equation\footnote{V.M.Kenkre and E.W.Montroll, J. Stat. Physics \textbf{9}, 45 (1973).} with a kernel derived from the simulations, distinguishing between trapped and passing particles. Results are relevant for other systems with chaos induced transport, e.g. electron transport in Tokamaks. [Preview Abstract] |
|
UP8.00072: Correlation Between Accretion Theory and Spontaneous Rotation Experiments* M. Landreman, B. Coppi, C. Di Sanzo The main observations that are consistent with the accretion theory [1] of the spontaneous rotation phenomenon include: i) the reversal of the direction of rotation in the transition from the L- to the H confinement regime that is attributed, by the theory, to the inversion of the phase velocity direction of ballooning modes excited at the edge of the plasma column; ii) the propagation of angular momentum from the outer edge toward the center of the plasma column during the L-H transition; iii) the strong effects of the magnetic field topology of the outermost magnetic surfaces and of the edge plasma regimes on the magnitude and direction of the spontaneous rotation; and iv) the intrinsic connection between spontaneous rotation and the plasma transport properties. The transition in the phase velocity direction of the considered modes is related to that which led [2] to the first experimental identification of collisional drift modes by a (linear) Q-machine where the transition marked the switch-off and on of modes with different mode numbers. A quantitative analysis of the factors that enter the application of the theory to current experiments (e.g. Alcator C-Mod) is given and the developments that this involves are discussed. *Sponsored in part by the US D.O.E. and the N.S.F. \\ {[1] B. Coppi, Nucl. Fus. {\bf 42}, 1 (2002)} \\ {[2] B. Coppi, H. W. Hendel, et al, PPPL Report MATT-523 (1967)} [Preview Abstract] |
|
UP8.00073: Confinement Regime Transition: Spontaneous Rotation Reversal and Collisionality of the Plasma Edge* C. Di Sanzo, B. Coppi, M. Landreman Within the context of the accretion theory [1] of the spontaneous rotation phenomenon, the transition between the L and the H confinement regimes is associated with the reversal of the phase velocity of collisional ballooning modes that can be excited at the edge of the plasma column. These modes are driven by the combined effects of the plasma pressure gradient and the magnetic field curvature, and involve in an essential way the electron-ion and ion-neutral collision rates and the effective transverse (concerning poloidal perturbed velocities) ion viscosity. According to the accretion theory the modes eject plasma angular momentum in the same direction as that of their phase velocity. When the edge is weakly collisional and characterized by local sharp density gradients (as in the H-regime) the mode rotates in the direction of the electron diamagnetic velocity. Under the opposite conditions (L-mode) the phase velocity is in the reverse direction and the consequent recoil causes the plasma column to rotate in the electron diamagnetic velocity direction. It is argued that the quality of confinement is associated with the rate of expulsion of angular momentum. A new resistive electrostatic mode that is driven by gradients of the ion pressure and of the longitudinal flow velocity has been found. *Sponsored in part by the US D.O.E. and the N.S.F. \\ {[1] B. Coppi, Nucl. Fus. {\bf 42}, 1 (2002)} [Preview Abstract] |
|
UP8.00074: Gyrokinetic turbulence and transport with kinetic electrons in NSTX plasmas W.X. Wang, S. Ethier, T.S. Hahm, S.M. Kaye, W.W. Lee, J. Manickam, G. Rewoldt, W.M. Tang Nonlinear gyrokinetic turbulence simulations for shaped plasmas have shown that ITG driven turbulence, even without the supression due to the equilibrium shear flow, drives insignificant ion energy transport in NSTX (about the the neoclassical level). This distinct feature is in contrast to the anomalous transport level for DIII-D simulations, where ITG turbulence is shown to drive a high level of transport (10 x neoclassicallevel), even though the mean turbulence fluctuation amplitude for these two machines are actually comparable. This remarkable difference in turbulent transport properties is further investigated by taking into account the effects of kinetic electrons. Here, full electron dynamics is simulated using the split-weight scheme in our new global simulation code. The equilibrium EXB shear flow is shown not to completely suppress the fluctuations in the well-developed nonlinear turbulence regime while it can stabilize ITG instability linearly. Also reported are our ITG/TEM simulations of NSTX and DIII-D discharges with the focus on energy loss through the electron channel and the comparison of the nonlinearly saturated k-spectra with the experimental measurements. This work was supported by U.S. DOE Contract DE-AC02-76-CH03073 and the SciDAC GPS Center. [Preview Abstract] |
|
UP8.00075: A New Split-Weight Scheme for Finite-$\beta$ Gyrokinetic Plasmas W.W. Lee, E.A. Startsev, W.X. Wang The original split-weight scheme for finite-$\beta$ simulations [1], which separates the perturbed particle distribution into an adiabatic part and a non-adiabatic part, is generalized to include spatial inhomogeneities. The new scheme requires an additional separation of the fast particle response associated with quasi-static bending of the magnetic field lines. While the original scheme follows the non-adiabatic response, $\delta h$, in time, where $\delta h = F - (1 + \psi) F_0$, $F$ is the distribution, $F_0$ is the background, $ \psi \equiv \phi + \int (\partial A_\parallel / \partial t) d x_\parallel/c $ and $\phi$ and $A_\parallel$ are the perturbed potentials, the new scheme makes use of $\hat {\bf b} \cdot \nabla (F_0+\delta g)=0, $ where $\hat {\bf b}=\hat {\bf b}_0+\delta {\bf B}/B_0$, and further separates the plasma response as $ F=(1 + \psi) F_0 +\delta g %+\int dx_{||}{\kappa}_n\cdot (\nabla A_{||}\times \hat {\bf b}_0) +\delta h, $ where $ \delta g = \int dx_{||}{\kappa} \cdot (\nabla A_{||}\times \hat {\bf b}_0) $ and $\kappa$ is the zeroth order spatial inhomogeneity. The new $\delta h$ is again followed in time. The results for finite-$\beta$ stabilization of drift waves and ion temperature gradient modes in slab geometry using the new scheme with a $\beta (\equiv c_s^2/v_A^2 )$ as high as $10\%$ and a grid size of the order of the electron skin depth, are in agreement with those discussed in Refs. [2] and [3]. This work is supported by the DoE OASCR Multi-Scale Gyrokinetics (MSG) Project. [1] W. W. Lee, J. Lewandowski, Z. Lin and T. S. Hahm, Phys. Plasmas {\bf 8}, 4435 (2001). [2] J. V. W. Reynders, Ph. D. Thesis, Princeton University (1992). [3] J. C. Cummings, Ph. D. Thesis, Princeton University (1995). [Preview Abstract] |
|
UP8.00076: Nonlocal Neoclassical Calculation of Toroidal Momentum Transport G. Rewoldt, W.X. Wang, M. Bell, S. Kaye, W. Solomon, R. Nazikian Motivated by experimental observations, neoclassical equilibrium and transport have been studied using global particle simulations by the GTC-Neo code. First, the toroidal angular momentum transport due to collisional dissipation has been calculated to understand whether the spontaneous toroidal rotation observed in plasmas without external momentum sources implies any anomalous momentum transport and torque. In some cases, GTC-Neo calculates that the toroidal momentum diffusivity is 5-6 times larger than previous predictions. Second, simulations of low aspect-ratio plasmas in NSTX show that there is considerable variation of $T_i$ on a magnetic surface, with up to a 20\% difference in $T_i$ between the outer and inner sides on the mid-plane. As a consequence, plasma temperature iso-surfaces are shifted from magnetic surfaces. This finite-orbit-width toroidal effect is enhanced as the ratio of ion orbit width to temperature gradient scale length is increased, but is insensitive to the density gradient. The dependence on machine parameters, plasma rotation and collisionality have also been studied. Third, simulations have been made of the poloidal momentum transport to help understand the origin of the ``anomalous'' poloidal flow observed in DIII-D. [Preview Abstract] |
|
UP8.00077: Low-q resonances, transport barriers, and secondary electrostatic convective cells Chris McDevitt, Patrick Diamond Recent experimental observations have suggested key characteristics of ITB formation near low-q surfaces in off- axis minimum-q (OAMq) discharges. These observations identify mean profile flattening localized to the low-q surface as a transition precursor in the absence of observable magnetic field perturbations. This observation suggests an electrostatic model of ITB formation which accounts for strong transport in the immediate vicinity of the low-q surface, as well as the formation of an ITB nearby the surface. Here, a low-m electrostatic convective cell driven by modulational instability of the background drift wave turbulence is discussed in the context of ITB formation near low-q resonances in OAMq discharges. Unlike pure m=n=0 zonal flows, convective cells are capable of intense mixing near low-q resonant surfaces as well as shearing, thus relaxing mean profiles near the resonant surface. Field line bending coupled with collisional viscosity are found to strongly damp the intensity of the vortical flows except in the case of weak magnetic shear. Furthermore, collisionless convective cell saturation mechanisms such as nonlinear wave trapping are largely circumvented due to the strong mixing of the convective cell. This suggests that low-m convective cells may play a key role in the regulation of turbulent transport near low-q resonances for OAMq discharges. [1] M. E. Austin et. al., Phys Plasmas 13, 082502 (2006) [Preview Abstract] |
|
UP8.00078: Nonlinear excitation and damping of Zonal Flows using a renormalized polarization response Fred Hinton, Patrick Diamond The nonlinear interaction of drift-wave turbulence and zonal flows is considered using an analogy with dressed test-particles in a stable plasma. The incoherent mode coupling potentials from the drift waves are treated as a source of noise driving the zonal flows. The coherent mode coupling potentials are included in a renormalized nonlinear polarization response to this noise source, analogous to the shielding of test-particles. The nonlinear damping of zonal flows and the conditions for a steady turbulent state are determined from the nonlinear polarizability. This calculation attempts to systematically address the effects of fluctuations and turbulence on the otherwise 'neoclassical' zonal flow polarization response. Thus it offers the possibility of identifying new nonlinear, kinetic `channels' for the coupling of zonal flow energy to dissipation. The implications for zonal flow saturation will be discussed. This work was supported by DoE Grant No. DE-FG02-04ER54738. [Preview Abstract] |
|
UP8.00079: Propagation of transport barriers in a simple model of coupled heat and particle fluxes Mikhail Malkov, Patrick Diamond Understanding of L $\rightarrow$ H transitions is a critical problem in magnetic confinement studies. Strong nonlinearity and coupling of particle and heat fluxes result in solution multiplicity, the L-H mode coexistence. Stationary L $\rightarrow$ H transitions studied in detail earlier reveal the pedestal width to be strictly coupled to the fueling profile in the case of neglected curvature of the pressure profile. Finite pressure curvature, however, shifts emphasis to the heating rate. To better understand the mechanism of L $\rightarrow$ H transitions, particularly factors that determine the pedestal width, we study time evolution of the temperature and density profiles and the L-H interface propagation. Both the inward and outward propagation can occur. It is shown that the heat production in the core region and the fueling at the edge determine the speed and direction of interface propagation. The front propagation solutions describe the penetration of the H-mode state into L-mode state and vice versa. The impact of these findings on the problem of hysteresis of stationary solutions is discussed. [Preview Abstract] |
|
UP8.00080: Extension of electrostatic gyrokinetics to transport timescales Felix I. Parra, Peter J. Catto We outline our efforts to develop an electrostatic nonlinear gyrokinetic full $f$ model that retains transport timescales and determines the electric potential self-consistently. A set of gyrokinetic variables is defined so that the gyrophase dependent part of the distribution is absorbed into the gyrokinetic variables by extending the linear treatment of Ref. [1]. The resulting gyrokinetic equation is valid for wavelengths as small as the ion Larmor radius and allows us to evaluate the gyrophase independent part of the distribution function through order $\rho _i /L$. When evaluating the potential, the quasineutrality equation usually employed in gyrokinetics may be inadequate for long wavelengths because the terms that determine the potential and zonal flow for short wavelengths become as small as other terms neglected due to limitations of the gyrokinetic equation. Various means of investigating these limitations and suggestions for testing the radial electric field determined from the gyrokinetic and quasineutrality equations will be presented. References: [1] X.S. Lee, J.R. Myra, and Peter J. Catto, Phys. Fluids \textbf{26} (1), 1983, 223-229. [Preview Abstract] |
|
UP8.00081: Spontaneous Toroidal Rotation in Tokamaks Malcolm Haines When two-fluid MHD theory of stability is employed the resulting growth rates are complex, and the perturbing magnetic fields move with a velocity that depends both on the components of the electron drift and heat flux perpendicular to the equilibrium magnetic field and on the diamagnetic velocity. On diffusing into a resistive wall a drag force is exerted on the wall which is proportional to the square-root of the velocity of the perturbing fields. The equal and opposite force or torque will be on the plasma, centred at the singular rational surface for each mode[1]. For typical experimental conditions this leads to a spontaneous, or intrinsic toroidal rotation of 20km/s occurring in a few milliseconds for perturbing magnetic fields of 0.0025tesla. The induced poloidal rotation by this mechanism is generally much larger, but there is considerable poloidal damping due to trapped particles on the ion-ion collision time- scale[2]. Furthermore poloidal angular momentum is in general not conserved for an isolated plasma, and any up-down asymmetry can act as a source or sink[3]; for example, Pfirsch-Schluter diffusion [3 damping by trapped particles[2] and the Ware pinch[4]. [1] J.B.Taylor, Phys.Rev.Lett. 91, 115002 (2003). [2] R.C.Morris, M.G.Haines and R.J.Hastie, Phys.Plasmas 3, 4513 (1996). [3] M.G.Haines, Phys.Rev.Lett. 25, 1480 (1970). [4] M.G.Haines and P.Martin, Phys.Plasmas 3, 4536 (1996). [Preview Abstract] |
|
UP8.00082: TEQ Free Boundary Equilibrium Solver in TRANSP/PTRANSP R. Andre, D. McCune, D. Pearlstein, L. Lodestro, W.H. Meyer The TRANSP code has traditionally been used to study the results of fusion tokamak experiments. In this mode of operation, the MHD equilibrium is reconstructed inside a prescribed boundary using inverse solvers such as VMEC and ESC. Accurate magnetic field values beyond the plasma boundary are not available. In the PTRANSP project, adding predictive capability to TRANSP, such limitations are overcome by using the free boundary direct solver of TEQ. With this, the poloidal flux on the full (R,Z) grid, the separatrix, and the coil currents can be self-consistently computed. The higher fidelity representation of the field is needed for neutral beam and RF models outside closed flux surfaces and for coupling to edge models. The availability of the TEQ direct solver will also enable options for improvement of the predictive Ohm's law model. This poster will describe the implementation of the TEQ direct solver in TRANSP/PTRANSP and the status of predictive modeling enhancements based thereon. [Preview Abstract] |
|
UP8.00083: Enhancement of NUBEAM for the simulation of fast ion and RF-wave interaction based on the quasi-linear theory Jae-Min Kwon, Douglas McCune, C.S. Chang The Monte-Carlo package NUBEAM for time-dependent modeling of fast ions in a tokamak geometry has been upgraded to simulate the effects of ICRF heating on the fast ions. The RF-wave field data is provided by executing TORIC5 inside TRANSP and passed to NUBEAM. An iterative algorithm has been implemented to match the RF-power absorption value calculated by NUBEAM with the level predicted by TORIC5. The effects of RF-wave fields on the fast ions are modeled by evaluating Monte-Carlo kicks based on the quasi-linear theory. Because of the unique feature of NUBEAM, the so called ``goosing'' which enables an order of magnitude faster calculation, special care needs to be taken in the Monte-Carlo simulation. The modification of the goose algorithm in the presence of RF-wave fields will be presented. Also, the necessary features of NUBEAM for future application to self-consistent coupling with an ICRF full wave code will be discussed. [Preview Abstract] |
|
UP8.00084: TRANSP and PTRANSP at PPPL: Status and Plans. Douglas McCune, Rob Andre, Eliot Feibush, K. Indireshkumar, Jae-Min Kwon, Christiane Ludescher-Furth, Lew Randerson The PPPL TRANSP code suite is a set of tools for time dependent simulation of tokamak plasmas. The entire system consists of over a million lines of fortran-77, fortran-90, C, and C++ code. Although pieces are over 30 years old, the code has been continually upgraded and modernized, now representing over 60 man-years of labor invested. TRANSP now runs as a service on the Fusion Grid, supporting plasma physics research groups around the world. In this poster, status and plans for TRANSP and associated predictive modeling upgrades (PTRANSP) are summarized. Fusion Grid production system results will be shown. Upgrades to physics models (MHD equilibrium reconstruction, ICRF wave interaction with beam injected fast ions, predictive transport), algorithms (MPI-parallelized source models), and client software (web-browser accessible interactive visualization of run results) will be summarized. The relationship of TRANSP/PTRANSP development efforts to SciDAC and FSP will be discussed. Related posters are cross-referenced. [Preview Abstract] |
|
UP8.00085: PTRANSP Simulations of Sawtooth Oscillations in Tokamak Plasmas G. Bateman, F.D. Halpern, A.H. Kritz, A.Y. Pankin, R.V. Budny, D.C. McCune Simulations with the PTRANSP predictive integrated modeling code are used to investigate sawtooth oscillations in tokamak plasmas. Components of the Porcelli model, the PORCELLI and KDSAW modules available in the NTCC Module Library http://w3.pppl.gov/NTCC, are implemented and used in the PTRANSP code to trigger sawtooth crashes and to reset plasma profiles within the sawtooth mixing radius duringing each sawtooth crash. The \mbox{H-mode} pedestal height is computed using the NTCC PEDESTAL module. Electron thermal, ion thermal, and momentum transport are computed using the GLF23 or the Multi-Mode anomalous transport models, together with neoclassical transport computed using the NCLASS model. The sawtooth model is calibrated by adjusting the magnetic reconnection fraction as well as coefficients in the model. This is accomplished by comparing the sawtooth period and amplitude with experimental data. The effects of sawtooth crashes on fast ion heating profiles, toroidal momentum profiles, as well as electron and ion temperature profiles are investigated. The calibrated simulation protocol is used to investigate the effect of sawtooth oscillations in ITER H-mode discharges. [Preview Abstract] |
|
UP8.00086: PTRANSP Simulations of Toroidal Momentum \mbox{Transport in Neutral} Beam Heated Tokamak Plasmas F.D. Halpern, G. Bateman, A.H. Kritz, A.Y. Pankin, R.V. Budny, D.C. McCune The PTRANSP code is used to predict self-consistently the toroidal rotational frequency, electron temperature, ion temperature, and ${\bf E}\times{\bf B}$ flow shear rate. Turbulence-driven thermal transport and toroidal momentum transport are computed using several transport models. A neoclassical contribution is added to the turbulence-driven toroidal momentum transport and thermal transport. It is found that inward fluxes of momentum can be generated by the Reynolds stress in the Weiland transport model. The neutral beam injection torque input, computed using the NUBEAM code, drives rotation in the plasma core, while charge exchange can drive rotation near the plasma edge. The poloidal velocity is computed using neoclassical theory. In H-mode discharges, it is found that the largest contribution to the ${\bf E}\times {\bf B} $ flow shear is usually a consequence of toroidal rotation. The rotation frequency is investigated as a function of plasma parameters including the torque per particle. The simulated radial profiles of the toroidal rotational frequency, ion temperature, and electron temperature are compared with experimental data. [Preview Abstract] |
|
UP8.00087: PLASMA SOURCES, SHEATHS, AND THRUSTERS |
|
UP8.00088: Modeling and experiments in argon-oxygen rf ICP pulse plasma. Vladimir Demidov, Evgeny Bogdanov, Anatoly Kudryavtsev, Konstantin Serditov, Charles DeJoseph, Jr. Numerical modeling of an argon-oxygen pulsed discharge (active phase and afterglow) has been performed for a specific device [W. Guo and C. A. DeJoseph, Jr., PSST, 10, 43 (2001)] for a number of experimental conditions. Spatiotemporal behavior of densities of plasma species, as well as fluxes of charged particles including fast electrons, have been calculated. It is demonstrated that in the afterglow of the plasma, production of fast electrons from both electron detachment of oxygen negative ions and from collisions involving argon metastable atoms can be very important. Conditions for self-trapping of these fast electrons have been identified. Measured values of charged particle densities agree reasonably well with calculations. The influence of the presence of the probe on measurements of negative ion densities has been also investigated. A simple method of regulation of negative ion densities and spatial distribution will be presented. [Preview Abstract] |
|
UP8.00089: Nonlinear kinetic effects in inductively coupled plasmas via particle-in-cell simulations Aaron Froese, Andrei Smolyakov, Dmytro Sydorenko Kinetic effects in inductively coupled plasmas due to thermal motion of particles modified by self-consistent magnetic fields are studied using a particle-in-cell code. In the low pressure, low frequency regime, electron mean free paths are large relative to device size and the trajectories are strongly curved by the induced rf magnetic field. Analytic linear theories are unable to recover effects accumulated along each nonlinear path. Therefore, the simulated ICP is made progressively more complex to find the source of observed plasma behaviours. With only thermal motion modifying the wave-particle interaction, nonlocal behaviour becomes dominant at low frequencies, causing an anomalous skin effect with increased skin depth and power absorption and decreased ponderomotive force. However, when influenced by magnetic fields, the nonlocal effects are suppressed at large wave amplitudes due to nonlinear trapping. A mechanism is proposed for this low frequency restoration of local behaviour. Finally, a low rate of electron-neutral collisions is found to counteract the nonlinear behaviour, and hence reinforces nonlocal behaviour. [Preview Abstract] |
|
UP8.00090: Design and Construction of the Plasma Bubble Expansion Experiment Y. Zhang, A.G. Lynn, S.C. Hsu, M. Gilmore, Christopher Watts We will present the design and construction of a new compact coaxial magnetized plasma gun and its associated hardware systems. The plasma gun will be used for experimental studies of ``magnetic bubble'' expansion into a pre-existing lower density background plasma on the HELCAT facility at UNM. These experiments will address key nonlinear plasma physics issues pertinent to plasma models of the formation and evolution of extra-galactic radio lobes. The gun will be powered by a 120$\mu $F 10kV ignitron-switched capacitor bank. High pressure gas, controlled by a gas valve system, will be puffed into an annular gap between inner and outer coaxial electrodes. An applied high voltage ionizes the gas and creates a radial current sheet. The $\sim $ 100kA discharge current generates toroidal flux, and an external magnet will provide poloidal ``bias'' flux. This poster will describe in detail the design and construction of the various power systems for the new plasma gun source. [Preview Abstract] |
|
UP8.00091: Initial operations of the ALEXIS device with a new, helicon-type rf source A. Eadon, E. Tejero, E. Thomas, M. Cianciosa The Auburn Linear EXperiment for Instability Studies (ALEXIS) is a 170 cm long, 10 cm diameter linear magnetized plasma column. Previous investigations [E. Thomas, et al., Phys. Plasmas, 10, 1191 (2003)] on the ALEXIS device have focused on modifications of the radial electric field, which resulted in axial currents and flow shear driven electrostatic ion cyclotron instabilities. Upcoming experiments on the electromagnetic branch of these instabilities require operation under finite beta conditions and current-free plasma generation. To accomplish this, the original filament source was replaced with a helicon-type rf plasma source. This presentation will give initial measurements of the plasma parameters obtained with the rf source and will compare those parameters to those of the filament generated plasmas. Additionally, observations of the low frequency instabilities in the rf generated plasma will be presented. [Preview Abstract] |
|
UP8.00092: Gas Flow Effects on Discharge Characteristics in a Dielectric Barrier Discharge Reactor of Spray Type Woo Seok Kang, Hyun-Su Kim, Sang Hee Hong The flow characteristic of discharge gas is an essential parameter in spray-type reactors of dielectric barrier discharge (DBD) to control the inside plasma density as well as the radical effluence density outside the reactors toward the work-piece surface in etching or surface treatment processing. To understand the effects of gas flow on reactor operation and plasma property, an experimental and numerical study has been carried out for a parallel-plate narrow-gap DBD reactor of spray-type, which is operated by 10 to 20 kHz sinusoidal voltages to produce argon or helium plasma with an oxygen additive. Varying gas flow rates from 0 to 100 liter/min, the discharge characteristics, such as current-voltage, breakdown voltage, and discharge power, are measured by electrical methods. Distributions of plasma temperature and some radicals ejected along the gas flow direction are estimated by OES diagnostics in both the inside discharge region and the outside radical effluence region. For detailed understanding of radical transport in the effluence area, a simple numerical modeling is developed on the basis of computational fluid dynamics including heat and mass transfer with plasma chemistry, and the calculated results are compared with experimental ones. Finally, the effects of metal surface treatment using the spray- type DBD with different gas flow rates are predicted and compared. [Preview Abstract] |
|
UP8.00093: Nonlocal Control of Plasma Properties in a Pulsed RF ICP in Argon-Oxygen Mixtures Jon Blessington, Charles DeJoseph, Vladimir Demidov, Mark Koepke Previously [1], we showed that a simple, three-level model could explain the rapid growth of charged particles (measured by probes) following application of rf power to a noble gas. In argon + O2 mixtures, the growth rate of O-atom density is slow compared to the growth rate of the charged particle densities. This growth can be estimated from plasma emission and from numerical modeling of the discharge. As a result, the positive ion density reaches a stationary value much faster than the atomic oxygen density. Thus, by changing the duration of the rf pulse, the ratio of fast electron production, from the reaction O + O- $\to $ O2 + e (3.6 eV), compared to the ambipolar flux of positive ions to the discharge walls, can be controlled. This effect can be used for nonlocal regulation of the plasma properties [2]. [Preview Abstract] |
|
UP8.00094: Effects of Discharge Current and Gas Flow Rate on CF$_{4}$ Abatement Process by Thermal Plasma Decomposition Sooseok Choi, Hyun Seok Lee, Jun Seok Nam, Woo Seok Kang, Sang Hee Hong Perfluorocompounds (PFCs) have been widely used in semiconductor and display industry for wafer etching and chamber cleaning processes. However, it is well known that PFCs are serious global warming gases. Although thermal plasma can efficiently decompose a significant quantity of waste gas, it has demerits of large consumption of electric input power and plasma forming gas in order to commercialize its processes. In this work, effects of arc current and plasma forming gas flow rate on the thermal plasma decomposition process have been experimentally demonstrated and numerically analyzed to improve its economic feasibility. A mixture of 1 {\%} CF$_{4}$ and the rest N$_{2}$ purging gas of several hundreds slpm was decomposed by nitrogen thermal plasma generated from a plasma torch with hollow electrodes. The input powers were changeable from 40 to 60 kW depending on torch operating conditions for arc currents of 130 through 150 A and plasma gas flow rates of 70 through 90 slpm. At a high current of 150 A and a low gas flow rate of 70 slpm condition, the corresponding input power was about 40 kW and over 96 {\%} of destruction and removal efficiency was achieved for 200 slpm waste gas. [Preview Abstract] |
|
UP8.00095: Free-floating atmospheric pressure ball plasmas G.A. Wurden, C. Ticos, Z. Wang, C.J. v. Wurden A long-lived (0.3 second, 10-20 cm diameter) ball plasma floating in the air above a water surface has been formed and studied in the laboratory. A 0.4 - 1 mF capacitor is charged to 4-5 kV, and subsequently discharged (30-60 Amps, 20-50 msec duration) into central copper cathode held fixed just below the surface of a bucket of water (with a weak solution of various salts in distilled water, such as CuSO4 or CuCl2, LiCl or NaCl). An underwater ring anode completes the circuit. A bubble of hot vapor from the water surface rises up in the first few milliseconds, and changes from a mushroom cloud with stalk, to a detached quasi-spherical object, finally evolving into a vortex ring. The plasma consists of ionized water vapor, with positive salts and OH- radicals, as well as molecular species, and it completely excludes nitrogen or oxygen from the rising plasma structure. A fine boundary layer is visible in orange, in contrast to a green ball interior when using Cu/CuSO4, and filamentary structures are visible at late times. Finally, a whisp of smoke ring is observed as a residue. A variety of visible and infrared imaging (both video and still cameras) are used, along with 200-800 nm time \& space resolved spectroscopy, to identify features of this laboratory analog to ball lightning. Possible applications include a windowless ball- plasma powered pulsed copper vapor laser operating at 510 nm. [Preview Abstract] |
|
UP8.00096: Anode Sheath Transition In a Carbon Nanotube Arc Discharge Abe Fetterman, Yevgeny Raitses, Michael Keidar An atmospheric pressure helium arc discharge is used for carbon nanotube synthesis. The arc discharge operates in an anodic mode with the ablating anode made from a graphite material. For such conditions, existing models predict the electron-repelling (negative) anode sheath [1]. In the present experiments, the anode ablation rate is investigated as a function of the anode diameter. It is found that anomalously high ablation occurs for small anode diameters ($<$ 0.4 cm). This result is explained by the formation of an electron-attracting (positive) anode sheath leading to increased power losses on small anodes as compared to larger anodes. The suggested mechanism for the positive anode sheath formation is plasma convergence. The increased ablation rate due to this positive sheath could imply a greater yield of carbon nanotube production. \newline [1] M. Keidar et al, J. Nanosci. Nanotech.6 (2006) 1309 [Preview Abstract] |
|
UP8.00097: Disinfection of {\it S.~mutans} Bacteria Using a Plasma Needle at Atmospheric Pressure S. Hansen, J. Goree, Bin Liu, D. Drake The plasma needle device produces a millimeter-size low-power glow discharge at atmospheric-pressure. It is intended for dental or medical applications. Radio-frequency high voltage is applied to a single needle electrode located inside a concentric gas-flow nozzle. A low-speed helium plasma jet flows out of the nozzle and mixes with ambient air. The jet is impinges on a surface that is to be treated, which in our test was a suspension of {\it S.~mutans} bacteria that was plated onto the surface of agar nutrient in a Petri dish. {\it S.~mutans} is the most important microorganism for causing dental caries. Imaging the sample after plasma treatment and incubation reveal the conditions where bacteria are killed, and the size of the treated spot. [Preview Abstract] |
|
UP8.00098: Dynamical properties of non-equilibrium atmospheric plasma jets and their applications to plasma processing in liquids Katsuhisa Kitano, Ikawa Satoshi, Hitoshi Furusho, Yukio Nagasaki, Satoshi Hamaguchi Non-equilibrium atmospheric pressure plasma jets are discussed with the emphasis on their physics and applications. Plume-like plasmas, which may be called plasma jets, have been generated in a discharge system consisting of a dielectric/metal tube (through which He gas flows at the atmospheric pressure) and a single electrode attached to the tube, to which low-frequency, high-voltage pulses ($\sim $10kV, $\sim $10kHz) are applied. With visible light images taken by a high-speed ICCD camera, it has been confirmed that the plasma jet consists of a series of small ``plasma bullets'' that are emitted intermittently from the powered electrode in sync with the positive voltage pulses. The observed ``plasma bullet'' may be interpreted as a fast moving ionization front. The plasma jets are energetic enough to generate highly reactive charge-neutral radicals but their gas temperatures remain low. Therefore the plasma jets are ideal for processing of liquid based materials at low temperatures and some examples of process applications, such as reduction of cations, polymerization of liquid monomers, and sterilization, will be also presented. [Preview Abstract] |
|
UP8.00099: High-beta effects in a helicon plasma Rod Boswell, Cormac Corr Above an input power of 900~W and a magnetic field of 30~G in WOMBAT, a 150 cm long 80 cm diameter chamber, a narrow column of bright blue Ar II light with a diameter of $\sim $~6~cm is observed along the axis of the diffusion chamber. Although the axial plasma density is very uniform, the radial profiles are not, suggesting that a large diamagnetic current might be induced. This has been investigated by measuring the temporal evolution of the magnetic field (B$_{z})$ and the plasma kinetic pressure in a pulsed discharge mode. Although the electron pressure can exceed the magnetic field pressure by a factor of 2, a complete expulsion of the magnetic field from the plasma interior is not observed. The magnetic field displays the strongest change at the plasma centre, which corresponds to the maximum in the plasma kinetic pressure. These results can be explained by taking into account the penetration of the magnetic field into the plasma which is faster than the plasma formation time resulting in only a slight perturbation of the magnetic field in the continuous plasma. [Preview Abstract] |
|
UP8.00100: Thermal Phenomena in Gas Confinement Dielectric Tube of the VASIMR Helicon Plasma Dan Berisford, R. Bengtson, L. Raja, J. Squire, L. Cassidy, J. Chauncery, G. McCaskill A quartz dielectric tube provides gas confinement in the helicon discharge of the VASIMR (Variable Specific Impulse Magnetoplasma Rocket) experiment. Despite highly aligned magnetic field lines to confine the plasma in the discharge, significant thermal heating of the dielectric tube occurs. We perform infrared camera imaging studies of heating of the tube with varying operational parameters of the experiment. Results show decreased heating of the tube as the plasma becomes more highly magnetized and less collisional. The data follows a trend that is well represented by a Bohm transport of ions perpendicular to the magnetic field lines suggesting that ion impact on the tube rather than radiation is the primary heating mechanism. Highly localized heating is also observed directly under the antenna in regions where the coils lie closest to the tube surface. This phenomenon is attributed to capacitive coupling effects that accelerate ions under the antenna coils, increasing the local energy flux to the tube surface. [Preview Abstract] |
|
UP8.00101: On the effect of dc resistance on determination of sheath electron density profiles in collisionless plasmas David N. Walker, David D. Blackwell, Richard F. Fernsler, William E. Amatucci In recent work$^{+}$ we examined primarily the high frequency ($\omega _{pe}$/2 $<\omega \quad <\omega _{pe})$ ac impedance characteristics of a small, negatively-biased, spherical probe immersed in collisionless laboratory plasma. Theoretical solutions indicate that collisionless resistance in the sheath at a given resonant radius is a function of applied frequency and is inversely proportional to the plasma density gradient there, $i.e.$, the gradient is evaluated at the radius where the applied frequency is equal to the plasma frequency. As the calculation nears the probe radius, the gradient increases and the density decreases toward zero causing the ac resistance to vanish in the limit, $i.e.,$ d$\omega _{pe}$/dr approaches a maximum and $\omega _{pe}$, a minimum. However, experiment shows an increasing resistance as the surface is approached and no tendency to exhibit a ``cutoff'' regardless of bias. We interpret these observations as arising from the dc response of the probe. We present results of experimental studies which include the lower frequency effects.$_{ }^{ +}$Walker, D.N., R.F. Fernsler, D.D. Blackwell, W.E. Amatucci, S.J. Messer, \textbf{\textit{Phys. of Plasmas}}\textbf{, }\textbf{\textit{13}}, 032108 (2006) [Preview Abstract] |
|
UP8.00102: Convective Instabilities and Enhanced Electron Scattering Inherent to Presheaths S.D. Baalrud, C.C. Hegna, J.D. Callen The stability of a presheath in weakly-collisional unmagnetized plasma with cold ions relative to electrons is analyzed. Ions are treated with the fluid equations and electrons as a collisionless (Vlasov) plasma in our derivation of inherent instabilities that grow due to the presheath electric field and corresponding ion flow. Our model suggests ion acoustic-type instabilities that depend on the local ion fluid speed throughout the presheath for modes with wavelengths typically shorter than the local Debye length. These convective instabilities propagate along the electric field with a growth rate depending upon the non adiabatic electron response. These instabilities produce a long range collective response for discrete particles. A Lenard-Balescu type collision operator is derived that accounts for the convective instabilities and leads to enhanced electron scattering relative to conventional Coulomb scattering and, therefore, modifies the electron distribution function. The presence of convective instabilities may provide an explanation for Langmuir's paradox whereby enhanced electron scattering may lead to populating the otherwise truncated part of the electron distribution function near a plasma boundary. [Preview Abstract] |
|
UP8.00103: Exact solution for the generalized Bohm criterion in a two-ion species plasma Dongsoo Lee, Lutfi Oksuz, Noah Hershkowitz For a weakly collisional two-ion species plasma, it is shown that the minimum phase velocity of ion acoustic waves (IAWs) at the sheath/presheath boundary is equal to twice the phase velocity in the bulk plasma. This condition provides a theoretical basis for the experimental results that each ion species leaves the plasma with a drift velocity equal to the IAW phase velocity in the bulk plasma [1]. It is shown that this result is a consequence of the generalized Bohm criterion and IAW dispersion relation. It is now apparent that the results for weakly collisional two-ion species plasmas are the same as for single-ion species plasmas. In both situations, the ion drift velocity at the sheath/presheath boundary is equal to the bulk ion sound velocity. \newline \newline [1] D. Lee, G. Severn, and N. Hershkowitz, Appl. Phys. Lett. (accepted for publication). [Preview Abstract] |
|
UP8.00104: Experimental verification of the Bohm criterion for two species plasmas Noah Hershkowitz, Lutfi Oksuz, Dongsoo Lee Ion acoustic wave (IAW) phase velocities are measured near the sheath/presheath boundary in weakly collisional argon/xenon plasmas. Wave profiles vs. position are measured using a boxcar averager with a gate of 30 nsec and CW excitation. Variable gate delays allow measurement of details of the wave close to the boundary. It is shown that the phase velocity at the presheath/sheath boundary is approximately twice the phase velocity in the bulk plasma for mixture and single species plasmas. This indicates each ion species drift velocity at the boundary is equal to the IAW phase velocity in the bulk plasma. This result is independent of relative ion concentration. [Preview Abstract] |
|
UP8.00105: Overview of Stanford's Activities in the Development of a Coaxial High Energy Thruster Mark Cappelli, Flavio Poelmann, Nicolas Gascon This poster gives an overview of Stanford's current efforts in the development of a coaxial gas-fed pulsed plasma accelerator that draws steady state power on the 1 to 10 kW level, but delivers thrust through high power, high density pulses. This Coaxial High ENerGy (CHENG) Thruster operates at number densities on the order of 10$^{15}$ cm$^{-3}$ and process peak input powers of 1 MW over 10 $\mu $s pulses. The high specific impulse, high thrust density, low beam divergence and low electrode erosion originally made the Deflagration thruster very attractive for missions to the outer planets and beyond. Stanford is exploring the scalability of this device for possible applications in orbit raising or travel to neighbor planets. The poster gives and overview of experimental work and theoretical models currently being developed at Stanford. [Preview Abstract] |
|
UP8.00106: Investigation of Plasma Potential and Electron Dynamics in the Near-Field of Hall Plasma Thrusters Andrew Smith, Mark Cappelli A 3-D map of the plasma potential in the near-field of a laboratory $E$ x$ B$ Hall thruster has been experimentally obtained. The tested thruster channel spans radially from 35mm to 47mm. The measurements lie within a cubic volume 100mm on a side centered on the central axis of the thruster. The results of a 3-D discrete electron transport simulation are presented for the near-field of the thruster. For a prescribed magnetic and electric field distribution in the near-field, a staggered leapfrog time-integrating method is utilized to track electrons launched from a simulated cathode. Inter-particle collisions and field instabilities are ignored though collisions with surfaces are treated. Spatial maps of the relative electron density, mean electron energy, and estimated local Hall parameter are presented. The model results indicate that measured channel to beam current ratios may be largely governed by the field structure in the near-field, and that the local Hall parameter is anisotropic in the domain with a mean value on the order of 1. [Preview Abstract] |
|
UP8.00107: Effects of sheath instability on plasma properties in a Hall thruster discharge Dmytro Sydorenko, Andrei Smolyakov, Igor Kaganovich, Yevgeny Raitses The sheath near the electron-emitting surface may become unstable if it is characterized by the negative current-voltage characteristic, which occurs in presence of strong secondary electron emission. A 1d3v particle-in-cell code is applied to study the sheath instability effects on plasma-wall interaction in Hall thrusters. It is found that in stable stationary plasma state the final phase of cyclotron rotation of secondary electrons emitted from the thruster walls is not arbitrary but belongs to the discrete set of stability intervals [Kaganovich et al., Phys. Plasmas 14, 057104 (2007); Sydorenko et al., submitted to Phys. Plasmas (2007)]. In the limit of high discharge voltages, a new regime with relaxation oscillations is identified. In this regime, the plasma constantly switches between a state with non-space charge limited emission and a state with a space charge limited emission [Sydorenko et al., IEPC-2005-078]. [Preview Abstract] |
|
UP8.00108: Characterization of the plasma plume in the current overrun regime of cylindrical Hall thrusters Erik M. Granstedt, Y. Raitses, N.J. Fisch Cylindrical Hall thrusters (HTs)\footnote{Y.~Raitses and N.~J.~Fisch, \emph{Phys.~Plasmas} 8, 2579 (2001)} may be more promising than annular HTs for low-power scaling due to a smaller surface-to-volume ratio. High plasma plume divergence is a main drawback to cylindrical HTs as decreased efficiency and spacecraft integration issues may result. Recent measurements of the plume angle show that overrunning the discharge current above its self-sustained value can significantly decrease plume divergence. In this ``current-overrun'' regime, the half-plume angle of the cylindrical HT was reduced to $55^\circ$. Thrust measurements demonstrate that the current-overrun regime can have an anode efficiency of up to 35--40\% at 100--200~W discharge power levels: an improvement of over 60\%. Measurements of the ion energy distribution function in the plasma plume using a retarding potential analyzer reveal both increased ion current density and ion energy on-axis, indicating that these ions are ionized in a region of higher plasma potential. Also, the average energy of off-axis ions is substantially reduced, resulting in improved performance and lowered risk of damage to spacecraft components. [Preview Abstract] |
|
UP8.00109: Experiments with Non-Self-Sustained Regimes of Hall Thruster Discharge Yevgeny Raitses, Artem Smirnov, Erik Granstedt, Nathaniel J. Fisch In conventional Hall thrusters, a steady state cross-field discharge is sustained between the anode and a hollow cathode. The current density at the cathode thermionic emitter is usually large enough to provide sufficient heating for self-sustained operation. It is commonly accepted that the thruster discharge current is limited by ionization of the working gas, wall losses and electron cross-field transport, and not by the electron supply from the cathode. We report that with all thruster parameters unchanged, the discharge current can be increased over and above what is normally required for sustaining the steady state discharge by running an auxiliary discharge between the cathode and an additional electrode [1]. For the cylindrical Hall thruster geometry, such a non-self-sustained operation is characterized by improved plasma plume focusing and higher thrust [1, 2]. These results are analyzed and compared with a conventional annular geometry Hall thruster. \newline [1] Y. Raitses et al, Appl. Phys. Lett. 90 (2007) 221502 \newline [2] A. Smirnov et al, Phys. Plasmas 14 (2007) 057106 [Preview Abstract] |
|
UP8.00110: Ion flows in a two-component expanding plasma Ioana Biloiu, Earl Scime The Ar and Xe ion velocity distribution functions (ivdfs), determined by laser induced fluorescence (LIF), in the expansion region of a helicon plasma have been measured as a function of ratio of Xe to Ar gas flow rates for a constant total gas flow rate. In the magnetic field gradient region ($\sim $5cm upstream from the helicon source-expansion chamber junction) and at low pressure ($\le $ 2 mTorr) the Ar ivdf is bimodal with a fast group at a speed of $\sim $ 7 km/s and a slow group at $\sim $ 3.5 km/s. The bimodal structure persists for a wide range of Ar/Xe gas flow ratios, disappearing (due to the lack of LIF signal) at 10{\%} Xe. Increasing Xe partial pressure does not affect the Ar ion flow velocities, but does decrease the Ar LIF amplitude. Conversely, the Xe ivdf is unimodal with a bulk ion flow of 0.85 km/s. To understand the excitation mechanisms of the investigated ion states, electron energy distribution functions were obtained from planar Langmuir probe characteristics by the Druyvestein method. The eedf is sensitive to the gas composition: for 2{\%} Xe, the eedf changes from a single Maxwellian with a temperature of $\sim $ 7.5 eV to a bi-Maxwellian with a cold component at $\sim $ 2.5 eV and a hot component at $\sim $8.5 eV. The effective temperature calculated from eedf integration exhibits a sharp decrease with increasing Xe fraction, down to $\sim $4.5 eV at 40{\%} Xe and is then constant for larger Xe fractions. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700