Bulletin of the American Physical Society
2005 47th Annual Meeting of the Division of Plasma Physics
Monday–Friday, October 24–28, 2005; Denver, Colorado
Session CP1: Poster Session II |
Hide Abstracts |
Room: Adam's Mark Hotel Grand Ballroom I & II 2:00pm |
|
CP1.00001: DIII-D I AND DIVERTORS |
|
CP1.00002: New Physics Capabilities for the DIII-D National Fusion Facility R.L. Boivin New physics research opportunities will be provided by upgrades and modifications to several DIII-D systems. The DIII-D EC system is being extended to six 10 s, 1 MW gyrotrons, and will provide the needed off-current profile control for sustaining steady state advanced tokamak discharges and continued research in active stabilization of NTMs. One beamline is being changed from co- to counter-injection, allowing new research in a number of areas using co- plus counter-beam injection: QH-mode, RWM stabilization at low rotation, modulated NTM stabilization, transport studies, etc. A modification of the lower divertor will enable cryopumping and density control in lower single null and high triangularity double null divertor plasmas: density control in higher triangularity plasmas will facilitate the development of high beta steady-state scenarios, and will enable transport and boundary research in ITER relevant collisionalities. In parallel with these three major enhancements, other systems, including diagnostics, are being improved. Details of new capabilities and the research they will enable will be discussed. [Preview Abstract] |
|
CP1.00003: Momentum Model of Gas Jet Penetration in Plasma P.B. Parks The lack of penetration of supersonic gas jets used in recent disruption mitigation experiments is explained. In the model [1], a cold, thin, plasma halo forms over the surface of the jet, shielding the neutral gas interior. The magnetic field inside the jet is slightly less than the field outside as a result of the balance between sideways advection of magnetic field lines by jet motion, and inward diffusion across the resistive halo. The net magnetic force opposing the jet motion then balances the neutral gas ram pressure piling up as a shock wave behind the tip of the jet, resulting in an equation for the ``tip speed" of the jet. The tip speed to jet speed ratio $U$ is very sensitive to jet density and halo resistivity. An argon jet ($\gamma = 5/3$) cannot penetrate unless $U > U_{crit} = 1/4$, otherwise the backward propagating shock wave reaches the rear jet surface, causing neutral ``spillage" into the vacuum region. In DIII-D experiments, the jet density $\sim 6\times 10^{16}$~cm$^{-3}$ was too feeble to satisfy our penetration criterion, even when the B-field was lowered 0.5~T. Indications for improving neutral particle penetration for ITER will be made. [Preview Abstract] |
|
CP1.00004: A New ITPA Disruption Database A.W. Hyatt, J.C. Wesley, E.J. Strait, D.P. Schissel, S.M. Flanagan Following a request by the International Tokamak Physics Activity (ITPA), \hbox{DIII-D}/General Atomics is hosting a new multi-device database designed to address scientific and engineering issues specific to disruptions in tokamak plasmas. A multi-device database containing a wide range of disruption relevant information will be very useful in setting design and operational limits for ITER. This new database is now operational, and will allow a user at any participating institution access to a broad range of disruption and equilibrium data of vetted discharges drawn from each contributing institution. Contributing institutions will populate and maintain their own disruption data. The goal is a common set of data allowing scientific exploration and engineering extrapolation. Primary data storage utilizes the MDSplus format, with all scalar data mirrored in a SQL relational database format. Data from several hundred shots at \hbox{DIII-D} will be presented along with data from other devices. Data structure, accessibility and security issues will be discussed, and participation solicited. [Preview Abstract] |
|
CP1.00005: The Role of the $m/n=3/2$ Tearing Mode in the Hybrid Scenario P.A. Politzer The hybrid scenario has been proposed as a robust operating scenario for high performance operation of ITER. Understanding the physics of the hybrid regime will allow more confident implementation. In these plasmas, $J (0)$ is lower and $q(0)$ is higher than in comparable conventional plasmas. A key feature in DIII-D hybrids is an $m/n=3/2$ NTM. This island structure is associated with the reduction ($q_{95}\leq4$) or elimination ($q_{95}\geq4$) of sawteeth. Decreasing the sawtooth amplitude reduces or eliminates a trigger for the deleterious $m/n=2/1$ NTM, which limits beta in the conventional H-mode scenario. The effect of the 3/2 mode on sawteeth has been shown using localized ECCD ($\leq50\,$kA) to enhance or suppress the mode amplitude. With co-ECCD the mode is suppressed and sawteeth appear. With counter-ECCD the 3/2 amplitude increases and small pre-existing sawteeth are suppressed. A variety of physical mechanisms may be involved in the regulation of $q(0)$ and the sawteeth by the 3/2 mode. Because the stationary state always has $q(0)$ close to one, it is likely that the observed 2/2 component of the 3/2 mode is playing a role. [Preview Abstract] |
|
CP1.00006: The 3/2 Magnetic Island and Its Effect on the Central Tokamak Region M.S. Chu, V.S. Chan, M. Choi, L.L. Lao, P.A. Politzer, H.E. St. John, A.D. Turnbull, D.P. Brennan In the hybrid discharge scenario in \hbox{DIII-D}, the central plasma evolves into a quasi-steady state without sawtooth. The central safety factor (q) is pegged close to 1 and correlates with the development of a rotating 3/2 magnetic island [1]. The causal relationship between the 3/2 island and the non-sawtoothing of the discharge is investigated. Equilibria modeling the discharge with different central q are analyzed using the PEST-III stability code. The 3/2 island is found to develop a 2/2 side-band with increasing amplitude as the central q approaches 1. This near resonant Alfven wave propagates with enhanced phase speed relative to the background plasma. With sufficient phase speed, the 2/2 side-band could drive currents which impedes the further decrease in q to trigger the sawtooth. The central 2/2 side-band does not lead to appreciable enhanced trapping of the plasma; but does modify the trajectory of the trapped particles and lead to additional transport.\par \vspace{0.5em} \noindent [1]~P.A.\ Politzer, et al., 32nd EPS Conf.\ on Plasma Physics, Tarragona, Spain (2005). [Preview Abstract] |
|
CP1.00007: Comparison of Measured and Simulated MSE Signals of a Tearing Mode M.A. Makowski, C.T. Holcomb, R.J. Jayakumar, R.J. La Haye, D.P. Brennan, M. Downes Spatially resolved magnetic field fluctuations associated with tearing modes have been observed with the fast MSE diagnostic on the \hbox{DIII-D} tokamak. The interpretation of the measurements is complicated by the fact that the spatial resolution of the diagnostic varies with channel, being better on the edge channels than the core channels. This combined with the relatively narrow spatial structure of the mode may explain the result that the observed fluctuation amplitude is stronger on the edge channels that on the core channels, despite the fact that the location of the island is in the core. We have developed a model of a tearing mode and performed integrals of the mode fields over the viewing volume to arrive at a synthetic signal for comparison with measurements. The effect of the varying spatial resolution is then evaluated for the different channels. Results of these calculations will be presented [Preview Abstract] |
|
CP1.00008: Observation of Compressional Alfven Eigenmodes (CAE) in a Conventional Tokamak M.S. Kim, W.W. Heidbrink, E.D. Fredrickson, N.N. Gorelenkov, T.L. Rhodes, M.A. Van Zeeland Fast-ion instabilities with frequencies somewhat below the ion cyclotron frequency frequently occur in spherical tokamaks such as the National Spherical Torus Experiment (NSTX). NSTX and \hbox{DIII-D} are nearly ideal for fast-ion similarity experiments, having similar neutral beams, fast-ion to Alfven speed, fast-ion pressure, and shape of the plasma, but with a factor of 2 difference in the major radius. When \hbox{DIII-D} is operated at low field (0.6~T), CAE instabilities appear that closely resemble the NSTX instabilities. In particular, the mode frequencies, polarization, and beam-energy threshold are nearly identical to NSTX. CAE in high-field discharges and emission at cyclotron harmonics are also observed. As on NSTX, the basic stability properties are consistent with the idea that the instability is driven by anisotropy in the fast-ion velocity distribution and is damped predominately by Landau damping of electrons. The results suggest that these modes could be unstable in ITER. [Preview Abstract] |
|
CP1.00009: Analysis of the D$_\alpha$ Spectra Produced by Fast Ions in DIII-D Y. Luo, W.W. Heidbrink, K.H. Burrell A diagnostic that measures the fast-ion distribution function through detection of the D$_\alpha$ light from neutralized fast ions acquired data during the 2005 campaign. In addition to the fast ion signal, there are impurity lines, bremsstrahlung and emissions from other neutrals in the D$_\alpha$ range. Background from visible bremsstrahlung and non charge-exchange impurity lines are subtracted by modulating the injected beam. Halo emission and charge-exchange impurity lines are fitted by using the method of nonlinear least squares. A bar at the exit focal plane of the spectrometer blocks bright interference from edge neutrals and injected neutrals. ELMs can devastate the spectrum by elevating the signal in the D$_\alpha$ range significantly. The ELM contaminated time slices are eliminated by applying a relative and absolute criterion based on the edge D$_\alpha$ signal. Pitch angle scattering and slowing down of beam ions are studied by varying the injection energy, beam angle, plasma density and electron temperature in quiescent plasma. Results are compared to classical theory. [Preview Abstract] |
|
CP1.00010: Effects of 3D Toroidally Asymmetric Magnetic Field on Tokamak Magnetic Surfaces L.L. Lao, D.P. Brennan, M.S. Chu, A.W. Hyatt, G.L. Jackson, R.J. La Haye, M.J. Schaffer, E. Soon, E.J. Strait, T.S. Taylor, A.D. Turnbull The effects of 3D error magnetic field on magnetic surfaces are investigated using the DIII-D internal coils (I-Coils). Slowly rotating $n=1$ traveling waves at 5 Hz and various amplitudes were applied to systematically perturb the edge surfaces by programming the I-Coil currents. The vertical separatrix location difference between EFIT magnetic reconstructions that assumes toroidal symmetry and Thomson scattering $T_e$ measurements responds in phase to the applied perturbed field. The oscillation amplitudes increase with the strength of the applied field but are much smaller than those expected from the applied field alone. The results indicate that plasma response is important. Various plasma response models based on results from the MHD codes MARS and GATO are being developed and compared to the experimental observations. To more accurately evaluate the effects of magnetic measurement errors, a new form of the magnetic uncertainty matrix is also being implemented into EFIT. Details will be presented. [Preview Abstract] |
|
CP1.00011: Magnetic Field Error Correction in DIII-D M.J. Schaffer, J.T. Scoville, R.J. La Haye Two new experiments continue to show that empirical correction of magnetic field errors is not a cancellation of the pitch-resonant Fourier components of the directly measured DIII-D field errors on the $q=$ 1, 2, and/or 3 surfaces. MHD theory suggests that the plasma response is greatest to pitch-resonant perturbations. The experiments varied the correcting field geometries in two ways: (a) apply fields designed to null specific error component(s), and (b) systematically scan the correction field geometry applied by the I-coil. Only 2 out of 8 geometries tested yielded empirical corrections approximately equal and opposite to the machine error resonant field; in 6 cases there clearly was not cancellation. The corrected field error symptom in these experiments was plasma rotation braking leading to an unstable locked mode in low-density Ohmic plasmas, as in past experiments. Locked mode avoidance in the new experiments correlates better with cancellation of nonresonant error components than resonant. The empirical plasma response still needs to be better understood. [Preview Abstract] |
|
CP1.00012: DIII-D Integrated Control Design Suite in Worldwide Use for High Confidence Plasma Control A.S. Welander, R.D. Deranian, J.R. Ferron, D.A. Humphreys, R.D. Johnson, R.J. La Haye, J.A. Leuer, B.G. Penaflor, M.L. Walker, D. Gates, J. Menard, D. Mueller, R.R. Khayrutdinov Making efficient use of experimental time in tokamaks means minimizing use of operations to develop control algorithms. High power devices such as ITER will require commissioning of controllers with minimal experimental development time. A systematic process in which controllers are designed based on experimentally validated models and verified against simulations can provide the necessary high level of confidence without experimental time. We describe development of a suite of tools implementing this approach, known as ``integrated plasma control," along with examples of its application to \hbox{DIII-D} and other tokamaks worldwide. These examples include algorithms for neoclassical tearing mode suppression and efficient shape control. Use of these methods has resulted in high confidence controllers that were successful in first-time use and has motivated the use of the \hbox{DIII-D} suite of tools on NSTX, KSTAR, EAST, and ITER. [Preview Abstract] |
|
CP1.00013: Comparison of ELM models with fast diagnostics in DIII-D John Hogan, L. Owen, M. Wade, D. Coster, M. Groth, W. Meyer, M. Fenstermacher, C. Lasnier To pursue development of an extrapolable model for the ELM/wall interaction, we compare the 2-D, time-dependent evolution of predicted D$_{\alpha }$ and CIII divertor region emission calculated by solps (b2-Eirene) simulations, directly with fast time-scale spectroscopic measurements [1]. As a complementary step, we compare the calculated edge/pedestal carbon ion (C$^{6+})$ evolution with fast edge CER measurements of these profiles [2], using an ELM model based on characteristics of MHD peeling-ballooning modes. An empirical radial transport model positing the strong reduction of the radial electric field at the ELM event, with accompanying enhanced radial transport, and then a slow restoration during the intra-ELM period, is found to give a reasonable qualitative description for this phase. Results will be presented for further tests of the model for a series of ELMy H-mode discharges at higher and lower density, and with forward and reverse B$_{T}$. in which the recovery phase appears to be faster with reversed B$_{T}$. [1] M Groth, M Fenstermacher et al J Nucl Mater 2003, [2] M Wade, K Burrell et al J Nucl Mater, 2005 [Preview Abstract] |
|
CP1.00014: Study of SOL in DIII-D tokamak with SOLPS suite of codes. Alexei Pankin, Glenn Bateman, Dylan Brennan, David Coster, John Hogan, Arnold Kritz, Andrey Kukushkin, Dalton Schnack, Phil Snyder The scrape-of-layer (SOL) region in DIII-D tokamak is studied with the SOLPS integrated suite of codes. The SOLPS package includes the 3D multi-species Monte-Carlo neutral code EIRINE and 2D multi-fluid code B2. The EIRINE and B2 codes are cross-coupled through B2-EIRINE interface. The results of SOLPS simulations are used in the integrated modeling of the plasma edge in DIII-D tokamak with the ASTRA transport code. Parameterized dependences for neutral particle fluxes that are computed with the SOLPS code are implemented in a model for the H-mode pedestal and ELMs [1] in the ASTRA code. The effects of neutrals on the H-mode pedestal and ELMs are studied in this report. [1] A. Y. Pankin, I. Voitsekhovitch, G. Bateman, et al., Plasma Phys. Control. Fusion 47, 483 (2005). [Preview Abstract] |
|
CP1.00015: Multimode Analysis of the Resistive Wall Mode Instability Dmitry Maslovsky, Allen Boozer We have developed a code [D. A. Maslovsky and A. H. Boozer, \textit{Phys. Plasmas}, \textbf{12}, 42108 (2005)] that uses a full spectrum of ideal MHD plasma modes and energies calculated by Alan Glasser's DCON, to compute the plasma effective inductance matrix $\stackrel{\leftrightarrow}{\Lambda}$ and the plasma stability matrix $\stackrel{\leftrightarrow}{S}$. The $\stackrel{\leftrightarrow}{\Lambda}$ matrix provides the normal magnetic field on the plasma surface produced by a surface current in the presence of plasma, thus describing plasma response properties to an applied external magnetic perturbation. Combined with the plasma surface inductance matrix $\stackrel{\leftrightarrow}{L}_{p}$, a purely geometric quantity, the plasma stability matrix can be obtained $\stackrel{\leftrightarrow}{S} \equiv \stackrel{\leftrightarrow}{L}_{p}^{1/2} \cdot \stackrel{\leftrightarrow} {\Lambda}^{-1} \cdot \stackrel{\leftrightarrow}{L}_{p}^{1/2}$, with the stability coefficients $-s_{i}$ as its eigenvalues in a non-rotating ideal plasma. By including the complete multimode plasma representation into VALEN, which models the surrounding conducting structures, the plasma response properties for arbitrary values of plasma stability coefficients $s_{i}$, and mode coupling effects, can be accurately assessed. In particular, cases when the value of the stability parameter of the least stable mode is of the order of unity, $s_{u} \sim 1$, and when the value of the stability parameter $s_{2}$ of the second mode is closer to zero than that of the least stable mode, $|s_{2}| < |s_{u}|$ can be analyzed. We discuss ITER-relevant cases of plasma stability calculation and feedback systems optimization. [Preview Abstract] |
|
CP1.00016: RWM Feedback Simulations for ITER with Noise O. Katsuro-Hopkins, J. Bialek, G.A. Navratil Sensor noise in resistive wall mode (RWM) feedback systems is an important factor in determining feedback power requirements and system performance limits. The VALEN RWM control model has been used to simulate these effects for ITER, based on the results of DIII-D time dependent simulations. The noise in magnetic field sensor coils for ITER was modeled based on observed experimental noise amplitude and measured power spectra for the DIII-D experiment with appropriate scaling coefficient. Because the ITER feedback system includes both derivative and proportional gains, VALEN was modified to add noise to the measured values of sensor’s voltage and flux. Presence of noise in the derivative feedback requires filtering in order to achieve reasonable limits of control coil voltages. High frequency cut off, feedback system voltage limits and Kalman filters were investigated. [Preview Abstract] |
|
CP1.00017: VALEN modeling of RWM with Rotation James Bialek, Alan Boozer The VALEN RWM active control modeling code has been extended to include the effects of mode rotation. Single mode theory predicts that rotation of an unstable plasma mode near a conducting wall lowers the growth rate of the instability and rotation above a critical value will lead to passive stabilization (ref{\#}1). This expected stabilization threshold has been observed in the VALEN model. VALEN can model the 3-D passive stabilizing structure surrounding the plasma. When the passive structure is not axisymmetric the mode is found to lock at small values of the mode torque parameter. Examples of these results are presented for ITER (with and without blanket modules), DIII-D and HBT-EP. \newline \newline Ref{\#}1 A.H.Boozer, Pop 1999 [Preview Abstract] |
|
CP1.00018: Wall intersection of ion orbits induced by fast transport of pedestal plasma over an electrostatic potential hill in a tokamak plasma edge C.S. Chang, S.H. Hahn, S.H. Ku An edge localized mode (ELM) event is known to transport a significant portion of pedestal plasma across the separatrix, and increase the divertor heat load to a possibly intolerable level in a tokamak fusion reactor. In the present work, a large random-walk transport is introduced in a plasma edge in a guiding center orbit following code XGC [C.S. Chang, S.H. Ku, H. Weitzner, Phys. Plasmas {\bf 11}, 2649 (2004)] to understand the effect of the shear in the edge radial electric field $E_r$ on the wall intersection location of the large-transport orbits. It is found that without an edge $E_r$, majority of large-diffusion induced ion orbital loss is to the outer divertor near the separatrix surface. However, with a large negative $E_r$ in the plasma edge with the ion grad-B drift into the single null divertor, ion orbits with their kinetic energy less than the potential energy shift their wall- intersection locations to the inner divertor while those with their kinetic energy higher than the potential energy retain their wall-intersections at the outer divertor. If the ion grad-B is away from the single-null divertor, the ion orbital loss is always to the outer wall. [Preview Abstract] |
|
CP1.00019: Rapid change of blob structure in the outer scrape-off layer (SOL) R.H. Cohen, D.D. Ryutov Nonlinear structures (``blobs'') driven by the magnetic field curvature and highly elongated along the field lines may exist in the tokamak SOL.\footnote{S.I. Krasheninnikov. Phys. Lett. {\bf A 283}, 368 (2001)} The contact of the blob end with the divertor plate significantly affects the blob structure and velocity. However, the strong shearing of the flux-tube near the X-point makes impossible direct electrical contact of the blob in the upper SOL and the divertor, so that the sheath boundary condition (BC) has to be replaced by a BC imposed near the X point.\footnote{D. Ryutov, R.H. Cohen. Contr. Pl. Phys {\bf 44}, 168 (2004)} We show that, at larger distances from the separatrix, in the far SOL, the connection between the upper SOL and the divertor plate is re-established, and the sheath BC becomes again relevant. During the blob’s outward radial motion, this event is reflected in a sudden change of its length, from the blob extending only to the X point to the blob extending down to the plate. Likewise, a blob initially existing only in the divertor leg becomes suddenly longer, and extends to the whole SOL. [Preview Abstract] |
|
CP1.00020: Self-Consistent Calculation of Turbulence and Transport in the Tokamak Edge M.V. Umansky, L.L. LoDestro, T.D. Rognlien, R.H. Cohen, X.Q. Xu Progress is described for self-consistent calculations of turbulence and transport in the tokamak edge using the UEDGE transport code and the BOUT turbulence code. Turbulence fluxes and background profile data are exchanged between the codes, and efficient iteration schemes are explored to bridge the very different turbulence and transport time scales. The codes are run under a common Python shell, which provides a convenient environment for automatic coupling. Here a limiter geometry is studied, where a material surface is introduced into the plasma at a specific poloidal location. This produces a simple edge plasma configuration with open and closed field magnetic line regions, supporting plasma instabilities driven by the radial gradients, magnetic curvature and the sheath boundary conditions. Turbulence-driven fluxes of particles and energy, and formation of the electric field well are studied in the coupled system for different poloidal locations of the limiter plate. An important factor affecting the poloidal distribution of the turbulence is the magnetic field shear from nearby X-points, although X-points themselves are outside of the modeled region. [Preview Abstract] |
|
CP1.00021: A two-region model of turbulent transport in the edge and scrape-off layer plasma D.A. Russell, D.A. D'Ippolito, J.R. Myra 3D BOUT simulations of turbulence in diverted tokamak plasmas $\footnote{D.A. Russell, D.A. D’Ippolito, J.R. Myra, W.M. Nevins, X.Q. Xu, Phys. Rev. Lett. \textbf{93}, 265001 (2004).} $ recently demonstrated dramatically enhanced transport by blobs in the scrape-off layer (SOL), correlated with sudden electrical disconnection of the outer-midplane (OM) from the divertor sheath, suggesting that reduced modeling of turbulent transport in the OM ought to include nontrivial ``parallel physics.'' We describe a minimal, ``2-region'' version of the 3D model introduced in [2]: gradient-driven turbulence in the OM is coupled to resistive cross-field transport in the X-point region by a \textit{jump condition} on Ohm’s law applied at the boundary between the two regions. The enhancement of cross-field conductivity by field-line fanning $\footnote{ D. Farina, R. Pozzoli, and D.D. Ryutov, Nucl. Fusion \textbf{33}, 1315 (1993).}$ is modeled by an area-preserving, \textit{stretching-and-squeezing}, coordinate transformation between the two regions. The model's linear, unstable eigenmodes are distinguished by X-point resistivity.$\footnote{ J.R. Myra, D.A. D’Ippolito, X.Q. Xu and R.H. Cohen, Phys Plasmas \textbf{7}, 4622 (2000).}$ Results from numerical simulations of the fully nonlinear model equations in regimes of strong turbulence will be presented. [Preview Abstract] |
|
CP1.00022: SOL Thermal Instability due to Radial Blob Convection D.A. D'Ippolito, J.R. Myra, D.R. Russell C-Mod data\footnote{M. Greenwald, Plasma Phys. Contr. Fusion {\bf 44}, R27 (2002).} suggests a density limit when rapid perpendicular convection dominates SOL heat transport. This is supported by a recent analysis\footnote{D.A. Russell et al., Phys. Rev. Lett. {\bf 93}, 265001 (2004).} of BOUT code turbulence simulations, which shows that rapid outwards convection of plasma by turbulent blobs is enhanced when the X-point collisionality is large, resulting in a synergistic effect between blob convection and X-point cooling. This work motivates the present analysis of SOL thermal equilibrium and instability including an RX-regime model\footnote{J.R. Myra and D.A. D’Ippolito, Lodestar Report LRC-05-105 (2005).} of blob particle and heat transport. Two-point (midplane, X-point) SOL thermal equilibrium and stability models are considered including both two-field (T) and four-field (n,T) treatments. The conditions under which loss of thermal equilibrium or thermal instabilities occur are established, and relations to the C-Mod data are described. [Preview Abstract] |
|
CP1.00023: Anomalous radial convection and strong toroidal flows in tokamak scrape off layer plasma S.I. Krasheninnikov, A. Yu. Pigarov, G.Q. Yu Here we present our results of: i) 2D turbulence modeling of nonlinear evolution of step-like electron temperature profile in the SOL caused by the grad(Te) instability. We study the formation and advection of the coherent structures and investigate their role in the temperature transport; and ii) 2D modeling (with transport code UEDGE) of the macroscopic edge plasma poloidal and parallel velocities. We demonstrate parallel plasma flows with Mach number $\sim $1 in the inner SOL region and show that the B-field variation can be an important ingredient in the formation of such flows. [Preview Abstract] |
|
CP1.00024: Three dimensional Particle-in-Cell simulation of blob transport Seiji Ishiguro Three dimensional particle in cell (PIC) simulation code for investigation of dynamic plasma behavior in a scrape-off-layer in magnetic fusion devices is developed. Particle absorbing boundaries corresponding to diverter plates and first walls are introduced. Non-uniform external magnetic field is also applied. Initially a blob is placed in the system. Charge separation due to Grad B drift and blob transport across the magnetic field due to EXB drift are investigated by using this PIC code. [Preview Abstract] |
|
CP1.00025: Comparison of thermal instability theory with MARFE density limit experiment in TEXTOR Frederick Kelly, Michael Tokar The density limits of shots in TEXTOR [Tokamak Experiment for Technology Oriented Research] that ended in multifaceted asymmetric radiation from the edge (MARFE) are analyzed with respect to several thermal instability theories\footnote{ W.M. Stacey, Phys. Plasmas \textbf{3}, 2673 (1996); \textbf{3}, 3032 (1996); \textbf{4}, 134 (1997); \textbf{4}, 242 (1997); \textbf{7}, 3464 (2000).}\footnote{ F.A. Kelly, W.M. Stacey, J. Rapp and M. Brix, Phys. Plasmas \textbf{8}, 3382 (2001).}\footnote{ M.Z. Tokar, F.A. Kelly and X. Loozen, Phys. Plasmas \textbf{12}, 052510 (2005).}. [Preview Abstract] |
|
CP1.00026: Tokamak Edge Transport Simulations with an E$_{r}$ Model for Varying Collisionality Regimes T.D. Rognlien, M.V. Umansky Edge-plasma transport codes typically utilize 2D fluid models supplemented by anomalous cross-field transport from plasma turbulence. In these collisional models, substantial vertical ion and electron gradient-B drifts as well as parallel current flow are present help determine the radial electric field, E$_{r}$. In the long mean-free path regime, the gradient-B and curvature drifts lead to periodic banana orbits, and the squeezing of the ion orbits to reduce the ion particle flux. An algorithm is proposed and demonstrated that provides a smooth transition between known results in these regimes. For the collisional regime, a simple model describes the sensitivity of E$_{r}$ to radial and poloidal particle-flows and sources. The impact of the different collisionality regimes on plasma profiles is illustrated for assumed anomalous transport coefficients. This model can be further calibrated by developing kinetic edge codes. [Preview Abstract] |
|
CP1.00027: EDGE2D Simulations of Diverted Tokamak Migration Patterns Jim Strachan Campaign averaged erosion/deposition measurements from JET, DIII-D, JT-60U, and AUG, as well as $^{13}$C tracer migration results from DIII-D, JET, and AUG are modeled using EDGE2D. Experimentally, the erosion peaks at the outer strike point, while the material eroded from the main chamber wall can dominate the deposited material. The deposition can peak at the inner strike point, or can be spread away from the inner strike point displaced to the SOL side. These features challenge any single model applied to the different experiments. Seemingly, three migration paths are important: 1) migration via the private flux region essentially from the outer strike point to the inner strike point, 2) migration via the main chamber SOL resulting in deposition on the inner target displaced away from the inner strike point, and 3) migration via the core and back into the main chamber SOL, also resulting in deposition on the inner target displaced away from the inner strike point. This study has examined the primary migration patterns ignoring re-erosion effects. Effects due to ELMs and projection to ITER are being examined. [Preview Abstract] |
|
CP1.00028: Non-Maxwellian ion distribution function in edge pedestal Seunghoe Ku, Sehoon Koh, C.S. Chang Gradient scale length of plasma density and temperature in the edge pedestal can be comparable to the banana orbit width of thermal ions. In the present-day large size tokamaks and future fusion reactors including ITER, the ion orbits in the H-mode pedestal can execute a significant radial excursion before experiencing Coulomb collision. Under such circumstances, a local thermal equilibrium may not be possible for ions. When Coulomb collisions are negligible, the ion distribution function becomes a canonical distribution function. In the large collisionality limit, the ion distribution function approaches a local Maxwellian. We use the XGC (X-point included Guiding Center[1]) code to study the property of ion distribution function in a quiescent pedestal plasma with steep density and temperature gradients. Deviation of the ion distribution function from a local Maxwellian or canonical Maxwellian distribution functions are studied as function of collision frequencies. It is shown that the non-Maxwellian property of the ion distribution function is greatly influenced by the orbit squeezing and expansion from radial electric field shear. [1] C.S. Chang, S. Ku, et al, Phys. Plasmas 11, 2649 (2004) [Preview Abstract] |
|
CP1.00029: Simulation of charging and dynamics of dust particles in the tokamak plasma environment with DUSTT code. Y. Tanaka, A.Yu. Pigarov, S.I. Krasheninnikov, T.K. Soboleva Recent theoretical predictions and experimental observations indicated that dust particles in tokamak plasma can be accelerated up to large velocities; $\sim $100 m/s. The dominant role in dust dynamics (in particular, in dust acceleration) plays ion/dust drag force. The values of drag force strongly depend on plasma parameters as well as on dust size, shape, temperature, and electric charge. Theoretical models for charging and drag forces available for practical applications are limited by spherical and non-emitting particles only that introduces uncertainty in dust dynamics. In order to assess the impact of ion/dust drag force uncertainties, we perform the sensitivity study by varying Epstein's drag force coefficients and Hutchinson's fitting formula. In our studies, we use the DUST Transport (DUSTT) code to simulate the trajectories of test dust particles based on realistic plasma environment calculated by plasma transport code UEDGE. [Preview Abstract] |
|
CP1.00030: Four-Dimensional Continuum Gyrokinetic Code: Neoclassical Simulation of Fusion Edge Plasmas X.Q. Xu, A. Xiong, B.I. Cohen, R.H. Cohen, M.R. Dorr, J.A. Hittinger, G.D. Kerbel, W.M. Nevins, T.D. Rognlien We are developing a continuum gyrokinetic code, TEMPEST, to simulate edge plasmas. Our code represents velocity space via a grid in equilibrium energy and magnetic moment variables, and configuration space via poloidal magnetic flux and poloidal angle. The geometry is that of a fully diverted tokamak (single or double null) and so includes boundary conditions for both closed magnetic flux surfaces and open field lines. The 4-dimensional code includes kinetic electrons and ions, and electrostatic field-solver options, and simulates neoclassical transport. The present implementation is a Method of Lines approach where spatial finite-differences (higher order upwinding) and implicit time advancement are used. We present results of initial verification and validation studies: transition from collisional to collisionless limits of parallel end-loss in the scrape-off layer, self-consistent electric field, and the effect of the real X-point geometry and edge plasma conditions on the standard neoclassical theory, including a comparison of our 4D code with other kinetic neoclassical codes and experiments. [Preview Abstract] |
|
CP1.00031: Simulation of Plasma Transport in a Toroidal Annulus with TEMPEST Z. Xiong, X.Q. Xu, B.I. Cohen, R. Cohen, M.R. Dorr, J.A. Hittinger, G. Kerbel, W.M. Nevins, T. Rognlien TEMPEST is an edge gyro-kinetic continuum code currently under development at LLNL to study boundary plasma transport over a region extending from inside the H-mode pedestal across the separatrix to the divertor plates. Here we report simulation results from the 4D ($\theta$, $\psi$, $E$, $\mu$) TEMPEST, for benchmark purpose, in an annulus region immediately inside the separatrix of a large aspect ratio, circular cross-section tokamak. Besides the normal poloidal trapping regions, there are radial inaccessible regions at a fixed poloid angle, energy and magnetic moment due to the radial variation of the $B$ field. To handle such cases, a fifth-order WENO differencing scheme is used in the radial direction. The particle and heat transport coefficients are obtained for different collisional regimes and compared with the neo-classical transport theory. [Preview Abstract] |
|
CP1.00032: Simulation of Carbon Production from Material Surfaces in Fusion Devices J. Marian, L.A. Zepeda-Ruiz, G.H. Gilmer, C. Mundy, E.M. Bringa, T. Rognlien, J. Verboncoeur Impurity production at carbon surfaces by plasma bombardment is a key issue for fusion devices as modest amounts can lead to excessive radiative power loss and/or hydrogenic D-T fuel dilution. Here results of molecular dynamics (MD) simulations of physical and chemical sputtering of hydrocarbons are presented for models of graphite and amorphous carbon, the latter formed by continuous D-T impingement in conditions that mimic fusion devices. The results represent more extensive simulations than we reported last year, including incident energies in the 30-300 eV range for a variety of incident angles that yield a number of different hydrocarbon molecules. The calculated low-energy yields clarify the uncertainty in the complex chemical sputtering rate since chemical bonding and hard-core repulsion are both included in the interatomic potential. Also modeled is hydrocarbon break-up by electron-impact collisions and transport near the surface. Finally, edge transport simulations illustrate the sensitivity of the edge plasma properties arising from moderate changes in the carbon content. The models will provide the impurity background for the TEMPEST kinetic edge code. [Preview Abstract] |
|
CP1.00033: Microinstability Calculations for Transport Barriers Mike Kotschenreuther, Prashant Valanju, Swadesh Mahajan, James Wiley, Mikhail Pekker Comprehensive gyrokinetic stability calculations are performed using GS2 for transport barrier equilibria generated by VMEC. Edge barriers are examined for free boundary equilibria with X-points. These are the first gyrokinetic calculations in such equilibria where the magnetic geometry near the X-point could be potentially significant. Comparisons with present experiments and extrapolation to reactors will be presented. We also examine whether novel magnetic divertors can improve pedestal stability. [Preview Abstract] |
|
CP1.00034: Design of Novel Magnetic Divertors with High Heat Flux Capacity Prashant Valanju, Mike Kotschenreuther, James Wiley, Mikhail Pekker We use STELLOPT tools (VMEC and COILOPT, which were developed for designing NCSX) to design novel magnetic divertor geometries for various machines (PEGASSUS, NSTX, ITER, and Reactor designs of various aspect ratios). Two categories of divertors are explored: 1) a second X-point to create flux expansion, or 2) extracting field lines outsides TF coils. Even though non-axisymmetric coils are used to allow better access, the plasma ripple is kept low. STELLOPT allows optimization of 3-D coils with free boundary equilibrium. [Preview Abstract] |
|
CP1.00035: Helium retention and Hydrogen absorption in FLiRE Benjamin Schultz, Paul Brenner, David Ruzic The FLiRE (Flowing Lithium Retention Experiment) facility consists of a flow loop which contains a two sections to observe flow along ramps in an upper chamber. As the Li exits the upper chamber it makes a vacuum seal isolation of the upper chamber from a lower one where thermal desporption spectroscopy can take place. By applying an ion beam or a plasma pulse to the open-channel Li flow on the ramp, studies can be made of He and H retention by measuring the partial pressure of He in the lower TDS chamber. Previous studies have shown about a 1\% to 2{\%} retention of He over a time scale sufficient to exit a potential flowing Li-walled reactor. The significance of such a result is very high and needs to be verified. It is possible that He implanted in the ramp before flow was initiated was absorbed leading to the observed increase. The experiment has been altered to address this and other concerns. Research on hydrogen absorption in liquid lithium exposed to hydrogen plasma has also been conducted. Overall results and their implications towards large scale fusion reactors are given. [Preview Abstract] |
|
CP1.00036: PLASMA SIMULATION I |
|
CP1.00037: Nonlinear Finite Larmor Radius Drift Kinetic Equation H. Vernon Wong A novel and efficient method is described for deriving the nonlinear drift kinetic equation. Electric drifts can be of the order of particle thermal velocities. A maximal ordering is invoked: the smallness parameter $\epsilon <1$ is proportional to $m/e$, subject to the proviso that $E_\|$ $\sim$ $\epsilon$. The drift kinetic equation is derived up to second order in $\epsilon$. With the parallel speed $v_{\parallel}$, magnetic moment $\mu$, and position vector ${\mathbf{r}}$ as phase space variables, the equation is derived in a form such that the phase space volume following the particle phase space trajectories is manifestly preserved. The moments of the drift kinetic equation reproduce the corresponding moments of the Vlasov equation to order $\epsilon^2$. The mean perpendicular velocity, momentum flow tensor, and pressure tensor are expressed in terms of the electromagnetic fields and the velocity moments of the drift kinetic distribution function $G$. A consistent set of fluid-kinetic equations is formulated, with the fluid-like perpendicular plasma motion described by the perpendicular component of the plasma momentum equation. The drift equation is used to describe the parallel plasma dynamics, and $G$ is required to evaluate the velocity moments necessary to close the set of equations. These equations provide a basis for the development of hybrid simulation codes. [Preview Abstract] |
|
CP1.00038: Modeling Collisional Effects in PIC Simulations of Laser-Target Interactions W.M. Sharp, M. Tabak, B.C. McCandless Realistic modeling of laser-target interactions requires a self-consistent representation of collisional effects. While such effects as impact ionization, recombination, and scattering can, in principle, be modeled using particle-in-cell (PIC) methods, the small collision lengths at solid density make this approach impractical. Hybrid-particle models, in which cell-averaged collisional forces are added to a conventional particle advance, allow PIC techniques to be used at higher densities. However, existing implementations of hybrid-particle transport, such as those in LSP and ANTHEM, assume an ideal-gas equation of state (EOS), which seriously misrepresents thermodynamics at solid density. We present a method for using an arbitrary EOS in a hybrid-particle transport model and discuss its implementation in LSP using a quotidian equation of state (QEOS). Results for a 2-D laser-target simulation obtained using this new model are compared with results for an ideal-gas EOS and with experiments. [Preview Abstract] |
|
CP1.00039: Denoising PIC Codes to Approach the Performance of Vlasov Codes Bedros Afeyan, Kirk Won, Vlad Savchenko, Jean Luc Starck, Steve Rosenthal, Thomas Melhorn, Dale Welch, Viktor Decyk, Ken Struve We use multiresolution analysis and wavelet techniques in order to reduce the random noise in PIC simulations caused by subsampling phase space. We compare the performance of two different sets of wavedlet based denoising schemes one of which is specifically designed for few sample point Poisson noise (between 10 and 30 particles per cell). The techniques are applied in both active and passive modes. In the passive case, we gain insight on the nature and amount of noise that accumulates in an ordinary PIC run by post processing. While with active denoising, we reduce this noise in real time during the simulation without allowing to accumulate. The examples used to gauge the benefits of the methods are the bump on tail instability and KEEN waves. [1] \newline \newline [1] B. Afeyan et al., Proc. IFSA (Inertial Fusion Sciences and Applications 2003, Monterey, CA), 213, B. Hammel, D. Meyerhofer, J. Meyer-ter-Vehn and H. Azechi, editors, American Nuclear Society, 2004. [Preview Abstract] |
|
CP1.00040: Prospects for the Meshless Adaptive Technique for Simulations in Plasma Physics Sergei Galkin, A.A. Martynov, S.Yu. Medvedev The meshless method is gaining popularity in recent days for numerical solution of partial differential equations in computational mechanics [1], fluid dynamics and astrophysics, but still being almost unused for plasma physics simulations. Finite element, finite volume, finite difference methods have been dominated in numerical simulations for a few decades. But for complex 3D boundaries, strong anisotropy of solution etc, the progress in their use has been slowed down by a slow progress in the development of 3D grid generators. For the meshless approach node locations are really needed but not a domain partitioning, that relieves pressure on the grid generator and simplifies adaptive node insertion/relocation. Applied to nonlinear problems such as plasma equilibria with negative central current, rotating star equilibria, convection-diffusion problem with high Reynolds numbers, the method had demonstrated high accuracy, robustness and performance. Difficulties, such as unknown boundaries, boundary layers and existence of multiple solutions, were clearly overcome and accurate numerical solutions were obtained with a relatively small total number of nodes. Due to its flexibility, the approach can bring essential benefits for the problems with various singularities. The method is straightforwardly extended to 3D and to time dependant problems. [1] S. Li and W.K. Liu, Meshfree Particle Methods, 2004, \textit{Springer-Verlag}. [Preview Abstract] |
|
CP1.00041: Fast Search and Adaptive Resolution for Complex Particle Kinetics David J. Larson, Dennis W. Hewett A new plasma simulation algorithm, intended to bridge the gap between Eulerian fluid and kinetic regimes, is now being used for a variety of applications in ICF and weapon effects. The CPK method (Complex Particle Kinetic) concept [1] uses an ensemble of macro-particles with a Gaussian spatial profile and a Mawellian velocity distribution to represent particle distributions in phase space. Time evolution is modeled by a combination of Lagrangian motion and internal evolution within each individual macro-particle. Collisional particle-particle interactions [2] are facilitated by sorting particles into bins depending of the particle size. Different bin levels are connected by a linked list. Searching for neighboring particles is highly efficient because the search is limited to particles in neighboring bins with the possibility of interaction. The bin structure also allows the computation of various spatial moments at different resolutions. Combining the results of the moment calculations yields information on where and when increased resolution is necessary. We will present details of the particle binning process along with progress towards our goal of simulating the transition from continuum to fully kinetic physics. [1] D. W. Hewett, J. Comp. Phys. 189 (2003). [2] D. J. Larson, J. Comp. Phys. 188 (2003). [Preview Abstract] |
|
CP1.00042: OSIRIS 2.0: an integrated framework for parallel PIC simulations Ricardo Fonseca, Michael Marti, Samuel Martins, Luis Silva, Frank Tsung, John Tonge, Michail Tzoufras, Warren Mori, Suzhi Deng, Tom Katsouleas, Chuang Ren We describe OSIRIS 2.0 framework, an integrated framework for particle-in-cell (PIC) simulations. This framework is based on a three-dimensional, fully relativistic, massively parallel, object oriented particle-in-cell code, that has successfully been applied to a number of problems, ranging from laser-plasma interaction and inertial fusion to plasma shell collisions in astrophysical scenarios. The OSIRIS 2.0 framework is the new version of the OSIRIS code. Developed in Fortran 95, the code runs on multiple platforms and can be easily ported to new ones. Details on the capabilities of the framework are given, focusing on the new capabilities introduced, such as bessel beams, binary collisions, tunnel (ADK) and impact ionization, and new diagnostics, and also dynamic load balancing and parallel I/O. This framework also includes a visualization and data-analysis infrastructure, tightly integrated into the framework, developed to post-process the scalar and vector results from our simulations. [Preview Abstract] |
|
CP1.00043: PARSEK: A Parallel Implicit PIC Code Stefano Markidis, Giovanni Lapenta CELESTE3D [1] has been an extremely successful tool for space plasma simulation. CELESTE3D is based on fully kinetic ions and electrons and on a implicit formulation of both field solver and particle mover. The code was designed in the early 1990s and adhered to the software infrastructures of the time. Furthermore the code had general geometry and grid adaptation, a feature key to fusion applications. However, the use of the code was largely limited to space and astrophysics applications. But a new age has dawned, a new successor of CELESTE has been fully developed and tested: PARSEK. PARSEK is based on the same algorithmic approach as CELESTE but it has a number of new features: 1) The particle mover and field solver are now relativistic; 2) The software infrastructure is completely new, based on a component- based software architecture using a object-oriented language (C++ or Java); 3) All components are fully parallelized using MPI. We present the new code, its benchmark tests and some sample applications. \newline \newline [1] G. Lapenta, J.U. Brackbill, W.S. Daughton, Phys. Plasmas, 10, 1577 (2003). [Preview Abstract] |
|
CP1.00044: Fully implicit moving mesh adaptive algorithm Luis Chacon, Giovanni Lapenta In many problems of interest, the numerical modeler is faced with the challenge of dealing with multiple time and length scales. The former is best dealt with with fully implicit methods, which are able to step over fast frequencies to resolve the dynamical time scale of interest. The latter requires grid adaptivity for efficiency. Moving-mesh grid adaptive methods are attractive because they can be designed to minimize the numerical error for a given resolution. However, the required grid governing equations are typically very nonlinear and stiff, and of considerably difficult numerical treatment. Not surprisingly, fully coupled, implicit approaches where the grid and the physics equations are solved simultaneously are rare in the literature, and circumscribed to 1D geometries. In this study, we present a fully implicit algorithm for moving mesh methods that is feasible for multidimensional geometries. A crucial element is the development of an effective multilevel treatment of the grid equation.\footnote{L. Chac\'on, G. Lapenta, A fully implicit, nonlinear adaptive grid strategy, {\em J. Comput. Phys.}, accepted (2005)} We will show that such an approach is competitive vs. uniform grids both from the accuracy (due to adaptivity) and the efficiency standpoints. Results for a variety of models 1D and 2D geometries, including nonlinear diffusion, radiation-diffusion, Burgers equation, and gas dynamics will be presented. [Preview Abstract] |
|
CP1.00045: Relativistic Implicit Moment Method Cesare Tronci, Gianluca Zuccaro, Stefano Markidis, Giovanni Lapenta Many plasma physics problems in astrophysics, laboratory and space systems require the development of suitable relativistic plasma simulation methods. Collisionless particle-in-cell (PIC) methods are often used to this purpose. We investigate here the possibility of a relativistic extension of an existing implicit PIC code (CELESTE), based on a moment equation closure model (1). Four possible ways of formulating the relativistic moment expansion are analysed from a purely numerical point of view, studying truncation errors and stability properties. Based on the analysis a new method is implemented into the 1D pilot code, PARSEK (2). The classical equation of motion from CELESTE is modified in order to keep a similar implicit integration method. The field solver is also presented, taking into account the previous results of the classical version. \\ \\ (1) J.U. Brackbill, D.W. Forslund, J. Computat. Physics, 46, 271, 1982. (2) S. Markidis, G. Lapenta, W.B. VanderHeyden, Z. Budimlic, Concurrency Comput Practice Experience, 17, 821, 2005. [Preview Abstract] |
|
CP1.00046: Fully implicit particle-in-cell algorithm. Hyung Kim, Luis Chacon, Giovanni Lapenta Most current particle-in-cell (PIC) algorithms employ an explicit approach. Explicit PIC approaches are not only time-step limited for numerical stability, but also grid-intensive due to the so-called finite-grid instability.\footnote{C. Birdsall and A. Langdon, {\em Plasma physics via computer simulation}, McGraw-Hill, New York, 1985} As a result, explicit PIC methods are very hardware-intensive, and become prohibitive for system scale simulations even with modern supercomputers. To avoid such stringent time-step and grid-size requirements, the implicit moment method PIC approach (IM-PIC) was developed.\footnote{J. Brackbill and D. Forslund, {\em J. Comput. Phys.} {\bf 46}, 271 (1982).} IM-PIC advances the required moments (density, current) using Chapman-Enskop-based fluid equations, and then advances the particles with such moments. While being able to employ much larger time steps and grid spacings than explicit PIC methods, IM-PIC is limited in that the time-advanced moments and the particle moments are inconsistent, resulting in lack of energy conservation. To remedy this, we propose here a fully implicit, fully nonlinear PIC approach (FI-PIC) where the particles and the moments are converged simultaneously using Newton-Krylov techniques. This guarantees the consistency of moments and particles upon convergence. We will demonstrate the feasibility of the concept using a purely electrostatic Vlasov-Poisson model, and will show its effectiveness with several fully kinetic examples. [Preview Abstract] |
|
CP1.00047: Modeling of complex geometries with the plasma simulation code VORPAL Chet Nieter, John R. Cary, Peter Messmer, David Bruhwiler, David Smithe, Gregory R. Werner Modeling complex structures and boundaries on a Cartesian grid is a challenge for many Finite Difference Time Domain (FDTD) electromagnetic PIC codes. The simulation of a variety of devices such as accelerating cavities, plasma processing chambers, and antennas at the edge of tokamaks require conformal (curve fitting) boundaries. Since these devices are fundamentally three dimensional, the capability to run in parallel on large numbers of processors is needed. We have recently added conformal boundaries using the method of Zagorodnov to the plasma simulation code VORPAL. Our boundary approximation can be viewed with a 3D VRML viewer. Also these complex devices often include open boundaries. VORPAL includes Perfectly Matched Layer (PML) boundaries which efficiently absorb out-going waves of any frequency and angle of incidence. VORPAL's FDTD algorithms scale to thousands of processors allowing for large 3D simulations. Simulations of complex structures using VORPAL will be presented. [Preview Abstract] |
|
CP1.00048: The Discontinuous Galerkin Method for Simulation of Plasma Dynamics: The Full Two-Fluid System John Loverich, Uri Shumlak An algorithm for simulation of plasma dynamics using the discontinuous Galerkin method is presented. The algorithm works in multiple dimensions arbitrary geometries and on parallel platforms. Second, third or 4th order genuinely multi-dimensional spatial discretization with 3rd or 4th order time discretizations are used. The method is explicit, resolves shockwaves in 2 or 3 cells and is implemented for the full two-fluid system with comparisons to the GEM challenge magnetic reconnection results. Full two-fluid simulations of a z-pinch are also presented. The technique can be used as a basis for fully electromagnetic PIC, or hybrid codes. Though the technique is applied to the full two-fluid system, the same technique can easily be applied to simpler plasma fluid models including two-fluid MHD, Hall MHD and MHD. [Preview Abstract] |
|
CP1.00049: A Grad-Shafranov Refiner Using High Order Spectral Elements Alan H. Glasser Numerical computation of the outer region matching data for resistive and other singular MHD modes of axisymmetric toroidal plasmas is much more sensitive to errors in the equilibrium solution of the Grad-Shafranov Equation (GSE) than corresponding ideal MHD stability analysis. While many direct and inverse GSE solvers are capable of producing sufficiently accurate solutions for ideal analysis, most have difficulty achieving adequate resolution and iterative convergence for resistive analysis. We have developed a Grad-Shafranov refiner using high-order spectral elements to improve on the accuracy of such solutions. The inverse form of the GSE and the equation for the Jacobian of the straight-fieldline coordinate system are solved iteratively, starting from an inaccurate initial solution to obtain a highly accurate final solution in inverse form, most suitable for stability analysis. Results will be presented for numerically challenging NSTX equilibria with small aspect ratio, large vertical elongation, and high beta. [Preview Abstract] |
|
CP1.00050: Fully implicit adaptive mesh refinement MHD algorithm Bobby Philip, Michael Pernice, Luis Chacon In the macroscopic simulation of plasmas, the numerical modeler is faced with the challenge of dealing with multiple time and length scales. The former results in stiffness due to the presence of very fast waves. The latter requires one to resolve the localized features that the system develops. Traditional approaches based on explicit time integration techniques and fixed meshes are not suitable for this challenge, as such approaches prevent the modeler from using realistic plasma parameters to keep the computation feasible. We propose here a novel approach, based on implicit methods and structured adaptive mesh refinement (SAMR). Our emphasis is on both accuracy and scalability with the number of degrees of freedom. To our knowledge, a scalable, fully implicit AMR algorithm has not been accomplished before for MHD. As a proof-of-principle, we focus on the reduced resistive MHD model as a basic MHD model paradigm, which is truly multiscale. The approach taken here is to adapt mature physics-based technology\footnote{L. Chac\'on et al., \emph{J. Comput. Phys}. \textbf{178} (1), 15- 36 (2002)} to AMR grids, and employ AMR-aware multilevel techniques (such as fast adaptive composite --FAC-- algorithms) for scalability. We will demonstrate that the concept is indeed feasible, featuring optimal scalability under grid refinement. Results of fully-implicit, dynamically-adaptive AMR simulations will be presented on a variety of problems. [Preview Abstract] |
|
CP1.00051: 2D-simulation of stationary MHD flows in the ducts of rectangular cross-section Ivan Khalzov, Andrei Smolyakov, Victor Ilgisonis The numerical code for a calculation of 2D stationary MHD flows of incompressible conducting viscous fluids (liquid metals) in straight and circular ducts of rectangular cross-section is developed. The flows are driven by the electrical current perpendicular both to the duct axis and to the external magnetic field. The code generalizes the well-known iterative Gauss-Seidel method for the case of systems of elliptic equations. The algorithm developed allows us to carry out the calculations of stationary flows in a wide range of Hartmann ($Ha=1\div 10^3$) and Reynolds ($Re=1\div 10^6$) numbers. The numerical results are presented for the experimental device, which is under construction in Russia. [Preview Abstract] |
|
CP1.00052: Simulation of MHD collimation from differential rotation Christopher Carey, Carl Sovinec Recent observations indicate that astrophysical outflows from active galactic nuclei are permeated with helical magnetic fields[1]. The most promising theory for the formation of the magnetic configurations in these magnetically driven jets is the coiling of an initial seed field by the differential rotation of the accretion disk surrounding the central object. We have begun simulations that are relevant to these Poynting jets using the NIMROD code[2]. To simulate dynamics on length scales that are significantly larger than the accretion disk, the non-relativistic MHD equations are evolved on a hemispherical logarithmic mesh. The accretion disk is treated as a condition on the lower boundary by applying a Keplerian velocity to the azimuthal component of the fluid velocity and a prescribed flux of mass through the boundary. The magnetic field configuration is initialized to a dipole like field. Formation of a jet outflow is observed later in time. The initial field is coiled up and collimated, driving a large current density on the axis of symmetry. Slipping of magnetic field lines due to non-ideal effects has been investigated. \medskip \linebreak 1. Asada K. et. al., Pub. of the Astr. Soc. of Japan, 54, L39-L43, 2002 \linebreak 2. Sovinec C. et. al., J. Comp. Phys., 195, 355-386, 2004 [Preview Abstract] |
|
CP1.00053: Semi-Implicit Extended MHD Simulation C.R. Sovinec, H. Tian, D.D. Schnack, A.Y. Pankin, D.C. Barnes A semi-implicit algorithm for non-ideal MHD is extended for two-fluid effects, including Hall, $\nabla p_e $, gyroviscous, and thermal drift terms. The temporally staggered differencing of the flow velocity relative to magnetic field, number density, and temperatures is retained from the MHD algorithm [1] to reduce algebraic system sizes. Analysis of the complete advance for homogeneous equilibria with flow shows that an implicit Hall advance is necessary for stability. The advance is implemented in the NIMROD code and applied to several tests including the nonlinear GEM reconnection problem [2], linear tearing, internal kink, and edge localized modes. Performance is compared with a time-centered advance. We also report on a recent investigation of convergence properties for interchange modes and compare with an earlier numerical analysis [3] for non-ideal MHD. \newline \newline [1] C. R. Sovinec et al., J. Comput. Phys. 195, 355 (2004). \newline [2] J. Birn et al., J. Geophys. Res. 106, 3715 (2001). \newline [3] H. Lutjens and J. F. Luciani, Comput. Phys. Commun. 95, 47 (1996). [Preview Abstract] |
|
CP1.00054: Overview of The Plasma Science and Innovation Center (PSI-Center) T.R. Jarboe, C.C. Kim, G. Marklin, A.I.D. Macnab, R.D. Milroy, B.A. Nelson, U. Shumlak, S. Vadlamani, S. Woodruff, R.A. Bayliss, C.R. Sovinec, E. Held, J-Y. Ji The Plasma Science and Innovation Center (PSI-Center) has recently formed. A principal goal is the refinement of overlapping computational tools with sufficient physics, boundary conditions, and geometry to be calibrated with experiments to achieve significantly improved predictive capabilities. The Center is for ICC experiments, especially EC experiments. The PSI-Center will initially concentrate on five focus areas: 1) two fluid / Hall physics, 2) kinetic and FLR effects, 3) reconnection and relaxation physics, 4) transport, atomic physics and radiation, and 5) boundary conditions and geometry. The entire ICC community is invited to participate in this center while nine experimental programs will provide the initial database. These nine experiments are:1) Caltech reconnection experiments, 2) FRX-L, 3) HIT-SI, 4) MBX, 5) PHD, 6) SSPX, 7) SSX, 8) TCS, and 9) ZAP. (Work supported by DOE) [Preview Abstract] |
|
CP1.00055: Validation of MH4D Code Modifications Using HIT-SI Experimental Results George Marklin, Thomas Jarboe This poster will report on progress to validate recent modifications to the resistive MHD code MH4D\footnote{ S. Vadlamani, et. al., this meeting.} by doing simulations of the HIT-SI\footnote{ T. R. Jarboe, et. al., this meeting.} experiment at the University of Washington. An unstructured tetrahedral mesh was constructed to closely match the geometry of the experiment using the mesh generation code T3D\footnote{ D. Rypl, http://power2.fsv.cvut.cz/$\sim $dr/t3d.html}. Simulations will start from an initial RFP-like Taylor State in one injector and will follow the plasma dynamics as the injector flux reconnects to form closed internal flux. These initial simulations will attempt to answer two important questions which can be compared to experimental results: (1.) How high must the injector current to flux ratio be driven to initiate the reconnection process; and (2.) How high does the Lundquist number, S, have to be in order to get build up of closed internal flux. Comparisons will also be made of predicted magnetic field structure to probe and flux loop measurements. [Preview Abstract] |
|
CP1.00056: Development of MH4D for Simulation of Emerging Concept Fusion Experiments Srinath Vadlamani, George Marklin, Uri Shumlak, Tomas Jarboe, Roberto Lionello The Boundary Conditions and Geometry group at the newly formed Plasma Science and Innovation Center (PSI-Center) is developing a tetrahedral mesh MHD simulation code to accurately model Emerging Concept (EC) experiments. We are using the MH4D\footnote{ D. Schnack and R. Lionello, Eos Trans. AGU, 85(47), 2004, Fall Meet. Suppl., SH13A-1158} code since it has a tetrahedral mesh which is required to fully represent the complex 3-dimensional geometry of many EC experiments, and it implements verified parallel algorithms based on the PETSc library. This poster will report on progress to add new boundary conditions and additional physics into the code to make it suitable for low to moderate S, circuit driven fusion experiments. Initial modifications will add the capability to model insulating flux conserver boundaries with inductive coupling to external circuits. Validation of these modifications by comparison to HIT-SI experimental results will be presented in another poster\footnote{ G. J. Marklin and T. R. Jarboe, this meeting.}. Further modifications will include atomic physics and neutral interactions in the edge region to model plasma sheath boundaries with circuit driven electrodes. [Preview Abstract] |
|
CP1.00057: Verification of a Hall MHD Algorithm in Nimrod for an FRC D.C. Barnes, A.I.D. Macnab, R.D. Milroy, C.R. Sovinec The accurate and efficient computation of low-frequency two-fluid phenomena in a confined plasma remains a challenge. As an initial project of the Plasma Science and Innovation Center (PSI-Center), a time-implicit two-fluid version of NIMROD$^{1}$ is being verified and applied to the macroscopic stability of a FRC. FRC macro-stability is a problem of intrinsic interest, both from a fundamental and from a practical standpoint, and also presents a unique verification opportunity. Implicit algorithm issues are briefly discussed. Initial $n$ = 1 linear results have reproduced earlier observed$^{2,3}$ transition from the ``fundamental'' MHD internal tilt mode to modes with higher structure along $B$, with growth rates smaller but comparable to MHD growth rates. Previous long-thin analysis$^{4}$ has been extended to capture the features of these modes and results are compared with NIMROD. Modifications to the analysis suggest effects beyond HMHD, which are important for the stability of these modes. Such effects will be incorporated into future NIMROD versions and verified by comparison with the analysis. $^{1 }$C.R. Sovinec, et al., J. Comp. Phys. \textbf{195}, 355 (2004) $^{2 }$R.D. Milroy, et al., Phys. Fluids \textbf{B1}, 1225 (1989) $^{3 }$Elena V. Belova, et al., Phys. Plasmas \textbf{10}, 2361 (2003) $^{4 }$D. C. Barnes, Phys. Plasmas 10, 1636 (2003) [Preview Abstract] |
|
CP1.00058: Extension of Drift Kinetic Hot Particles to Full Orbits in NIMROD Charlson C. Kim, Carl R. Sovinec, Richard D. Milroy The primary goal of the Plasma Science and Innovation Center (PSI Center) is to refine and optimize existing MHD codes to achieve improved predictability for emerging concept (EC) experiments. Kinetic effects have been shown to play a dominant role in some EC experiments, particularly in FRC stability\footnote{\normalsize{E.~Belova, et.al. ``Numerical Study of tilt stability of prolate field-reversed configurations,'' PoP, {\bf 7}, 4996, 2000}}. The Center will extend the hybrid kinetic-MHD implementation in NIMROD\footnote{\normalsize{C.C. Kim, et.al. ``Hybrid Kinetic-MHD Simulations in General Geometry,'' CPC, {\bf 164}, 448, 2004}} from the drift kinetic model to the full kinetic model to include sufficient physics to accurately account for these effects. We outline the algorithmic issues involved, in particular as it relates to incorporation in a semi-implicit MHD code using finite element (FE) basis functions. Particular issues addressed will be particle-in-cell (PIC) in FE, timestep disparity between particles and MHD fields, and the potential for advanced timestepping algorithms for the particles in the semi-implicit code. Initial progress along these lines will be presented. [Preview Abstract] |
|
CP1.00059: Nonlocal parallel heat transport and energy confinement in SSPX J.-Y. Ji, E.D. Held, C.R. Sovinec As confinement improves and temperature increases in emerging concept experiments such as the spheromak, transport in the core of the devices becomes increasingly collisionless. Understanding transport in these experiments is one goal of the Plasma Science and Innovation Center. Recently, numerical studies of a set of pulsed, electrostatically-driven discharges in the SSPX experiment were performed using the NIMROD code \footnote{C. R. Sovinec {\it et al}., Phys. Rev. Lett. {\bf 94}, 035003 (2005).}. The resistive magnetohydrodynamic simulations show good agreement with respect to magnetohydrodynamics and fair agreement with respect to confinement. In this work, a general, parallel heat flux closure, $q_\|$, is applied in place of the collisional, Braginskii closure. Because the general form for $q_\|$ allows for arbitrary collisionality, it is applicable both in the hot core as well as the collisional edge plasma. In addition, the general, nonlocal $q_\|$ correctly assesses parallel heat flow in core regions characterized by magnetic field line chaos. We expect that these features of the general $q_\|$ lead to improved agreement between experimentally observed and simulation temperatures. [Preview Abstract] |
|
CP1.00060: The Plasma Science and Innovation Center Interfacing Group B.A. Nelson, C.C. Kim, A.P. Cassidy, S.D. Griffith, R.D. Milroy, T.R. Jarboe The Interfacing Group of the Plasma Science and Innovation Center (\urllink{PSI--Center}{http://www.psicenter.org}) facilitates simulations of collaborating Emerging Concept (EC) experiments. This includes helping set up experiments for new simulations, assisting in comparison of simulations with experimental data, acquiring and disseminating information about simulations to PSI--Center computational groups, and acting as a conduit of information between experiments. Present collaborating experiments include the Bellan Plasma Group (Cal-Tech), FRX--L (Los Alamos National Laboratory), HIT--SI (Univ of Wash), MBX (Univ of Texas--Austin), PHD (Univ of Wash), SSPX (Lawrence Livermore National Laboratory), SSX (Swarthmore College), TCS (Univ of Wash), and ZaP (Univ of Wash). The Interfacing Group also administers a local development computational cluster and network. PSI--Center 3-D resistive MHD codes include \urllink{NIMROD}{http://www.nimrodteam.org} and MH4D. Initial results from new NIMROD runs, and improved numerical diagnostics and data visualization for collaborating experiments will be presented. [Preview Abstract] |
|
CP1.00061: Extended MHD modeling on an arbitrary curvilinear adaptive grid Vyacheslav S. Lukin, Alan H. Glasser A massively parallel spectral/(hp) element code (SEL)\footnote {A. H. Glasser and X. Z. Tang, Comp. Phys. Comm. {\bf 164} (2004).} is being developed at LANL in order to advance against the present day limitations on the existing plasma fluid codes imposed by the presence of a wide range of length/time scales and high degree of anisotropy in most plasmas of interest. Having previously demonstrated a highly accurate and efficient parallel operation of SEL on fixed non-uniform meshes, we now present the results of solving a two-fluid extended MHD system of equations in a 2D magnetic reconnection geometry on a logically rectangular adaptive mesh. To make that possible, a static grid rezoning algorithm has been implemented, where the computational grid is updated from time to time, whenever deemed necessary, without interrupting the simulation. A harmonic grid generator designed to be able to both align the grid with evolving magnetic field and to concentrate it in the regions lacking the desired spatial resolution is used to update the mesh. The new code's accuracy and efficiency is tested by reproducing the results of the GEM reconnection challenge. \footnote{J. Birn, {\it et. al.}, J. Geophys. Res., {\bf 106}, pp. 3715-3720 (2001).} [Preview Abstract] |
|
CP1.00062: Stabilization of z-pinch m=0 instabilities by axial sheared flows Ioana Paraschiv, Bruno S. Bauer, Vladimir I. Sotnikov, Volodymyr Makhin The growth and saturation of the m=0 magnetohydrodynamic (MHD) instability is numerically studied in a cylindrical diffuse Bennett equilibrium in the presence of sheared plasma flows with the aid of a 2D MHD code (MHRDR). Using Fourier analysis the amplitude of different axial modes is followed from the linear to the nonlinear regime. It is found that the linear growth rates and the nonlinear saturation level of the m=0 modes decrease substantially with increasing velocity shear. Two different profiles for the axial velocity are used, one parabolic and one linear in radius. It is shown that the linear profile, which has a constant shear, is much more efficient in stabilizing the m=0 modes. [Preview Abstract] |
|
CP1.00063: Initial simulation of MHD instabilites in a high speed plasma accelerator Jin-Soo Kim, Ioan Nick Bogatu, Tom Hughes, Dale Welch, Francis Thio High density, high Mach number plasma jets are under development for a variety of critical fusion applications. These applications include fueling, rotation driving, and disruption mitigation in magnetic fusion devices. They also include a range of innovative approaches to high energy density plasmas. FAR-TECH, Inc. has begun 3D MHD simulations using the LSP code [1] to examine such high speed plasma jets. An initial study to benchmark the code is currently underway. The blow-by instability will be simulated in a coaxial plasma accelerator using the 3D LSP code and compared with the 2D MACH2 code results. \newline [1] LSP-Manual-MRC-ABQ-R-1942.pdf [Preview Abstract] |
|
CP1.00064: Accurate Evaluation of Numerical Derivatives for the Grad-Shafranov Equation Jeffrey Freidberg, Paul Nerenberg To carry out an axisymmetric toroidal MHD stability analysis one needs an accurate numerical evaluation of the flux function and its first and second derivatives. While there are standard numerical techniques for solving the Grad-Shafranov equation for the flux function, numerical evaluation of the first and second derivatives leads to increasing levels of inaccuracy. Often, very high resolution is needed for the flux function to obtain satisfactory accuracy in the second derivatives. A numerical technique is presented that allows one to calculate the first and second derivatives to the same accuracy as the flux function, using the same resolution for all calculations. The technique makes use of a novel application of Green's theorem. Specific examples are presented demonstrating the technique and the improved accuracy of the first and second derivatives. [Preview Abstract] |
|
CP1.00065: Integration of Corsica and Mathworks Simulink for Simulation of Tokamak Plasma Control System Operations W.H. Meyer, T.A. Casper, L.D. Pearlstein Using an RPC interface that allows us to do process-to-process communication, we are integrating the Corsica equilibrium and transport code into Mathworks Simulink for simulation of tokamak plasma control system operations. We have done initial testing using a simple vertical-position-control DIII-D simulation and are now in the process of defining the plasma model interface and modifying Corsica to implement that interface. The interface will be as general as possible - machine parameters will be driven from Simulink - but balanced against the need to maintain a standalone Corsica, in particular, the free-boundary equilibrium solutions. The objective is code verification by benchmarking open-loop Corsica evolution with DIII-D experimental data*. Corsica will then be used as a plasma model in Simulink for closed-loop DIII-D and ITER scenario simulations.\break \break * Leuer, J.A., et al, Proc. 20th IEEE Symp. on Fus. Eng., San Diego, USA, Oct 2003 [Preview Abstract] |
|
CP1.00066: Unconditionally Stable Explicit Quantum Lattice Representation of 1D MHD Turbulence, with Arbitrary Transport Coefficients Jeffrey Yepez, George Vahala, Linda Vahala 1D models give valuable insight into the complexities of turbulence -- like the Burgers equation for Navier-Stokes turbulence. For MHD, a magnetic field generalization of Burgers equation has been considered by Yanase and Diamond et. al. This 1D MHD model exhibits Alfvenization -- the interchange of fluid and magnetic energies. Here we present a quantum lattice representation that is an explicit unconditionally stable algorithm for any viscosity and resistivity. There are typically 4 steps in the quantum algorithm: (a) state preparation of the qubits whose excited state occupational probability are correlated to the velocity and magnetic fields, (b) local unitary collisional entanglement of the on-site qubits, (c) measurement of post-collision occupation probabilities, (d) unitary streaming to nearest neighbor nodes. The energy spectrum shows no bottleneck near the dissipative subrange, unlike the Yanase simulations that exhibit energy pile at the longest wave numbers. We discuss the interconnection between the microscopic Schrodinger, mesoscopic quantum Boltzmann, and the macroscopic 1D MHD equations. [Preview Abstract] |
|
CP1.00067: 3D Entropic Lattice Boltzmann Simulations of 3D Navier-Stokes Turbulence George Vahala, Jeffrey Yepez, Linda Vahala, Min Soe, Jonathan Carter Turbulence yields highly complex trajectories in x-space whose evolution is very difficult to accurately follow computationally. However, by appropriately projecting into a higher dimensional phase space the trajectory is simple and its evolution readily evaluated. Lattice Boltzmann (LB) representations play such a role for fluid turbulence. Moreover, in complex geometries and high Reynolds number flows, the smallest scales cannot be resolved but the usual closure schemes of eddy viscosity cannot be justified because of lack of separation of scales. At the kinetic level closure schemes can be better justified. After coarse-graining, we recover better macroscopic turbulent closure models. An Achilles' heel of LB simulations is nonlinear numerical instabilities, typically triggered as the transport coefficients tend to zero. However, by imposing a discrete H-theorem constraint on the evolution of the distribution function one achieves an entropic LB algorithm that is unconditionally stable for any level of transport coefficients. 3D turbulence simulations will be presented for the highly symmetric Kida profile as well as the Taylor-Green vortex. [Preview Abstract] |
|
CP1.00068: Scaling Laws of Confinement Parameters for the Advanced Reversed Field Pinch Jon-Erik Dahlin, Jan Scheffel A series of resistive magnetohydrodynamic numerical simulations are performed to generate scaling laws for energy confinement time \textit{$\tau $}$_{E}$ and poloidal beta \textit{$\beta $}$_{p}$ for the advanced reversed field-pinch (RFP). Strongly improved scaling with basic initial parameters is obtained as compared to the conventional RFP. Early results indicate an improved scaling of \textit{$\tau $}$_{E}$ with plasma current $I$ and line density $N$ compared to the conventional RFP on the order of \textit{$\tau $}$_{E}$ (\textit{adv.}) / \textit{$\tau $}$_{E}$ (\textit{conv.}) $\propto $ ($I$/$N)^{0.29}I^{0.29}$. The improved behavior of the advanced RFP as compared to the conventional, uncontrolled RFP stems from the introduction of current profile control (CPC). In the present numerical simulations, CPC is performed by implementation of a \textit{parameter free} automatic feedback algorithm, optimized to reduce the fluctuation caused $\left\langle {{\rm {\bf \tilde {v}}}\times {\rm {\bf \tilde {B}}}} \right\rangle $ electric field. The scheme introduces an ad-hoc electric field within the plasma volume, automatically adjusted to dynamically control the plasma into more quiescent behavior by eliminating current driven tearing mode instabilities and reducing resistive interchange modes. [Preview Abstract] |
|
CP1.00069: Wavelet analysis of gyrokinetic turbulence Ingmar Broemstrup, Marie Farge, Kai Schneider, William Dorland A method to extract coherent structures out of the data from a gyrokinetic simulation. The data is generated with GS2, which is a nonlinear kinetic code that studies low-frequency turbulence in magnetized plasma. We analyze data from simulations of ion temperature gradient-driven (ITG) turbulence, for a range of physics parameters. The parameters are chosen to produce a strong variation in the amplitude of nonlinearly generated zonal flows. To find coherent structures, the data is projected onto an orthogonal wavelet basis, a nonlinear thresholding is applied to the wavelet coefficients and the denoised data is then reconstructed in physical space. [Preview Abstract] |
|
CP1.00070: BASIC PLASMA: RECONNECTION, COMPUTATION, AND FLOWS |
|
CP1.00071: Multipactor Induced Flashover in HPM Waveguides Chris Fichtl, Keith Cartwright Modeling multipactor discharges is important for understanding window breakdown and parasitic plasma discharges in high power microwave tubes. This poster makes use of the Secondary Electron Emission (SEE) codes that have been incorporated into the ICEPIC (Improved Concurrent Electromagnetic Particle-in-Cell) code to study these phenomena. A 2-dimensional waveguide simulation looking at the formation of multipactor discharge and its role in causing flashover as a function of waveguide size and electric field will be presented. The area scaling of the minimum electric field that result in a discharge in PIC simulations will be compared to the commonly assumed scaling of $E\propto \frac{1}{\sqrt A }$. [Preview Abstract] |
|
CP1.00072: The Effect of Secondary Emission Cathode Parameters on (Near-) Brillouin Flow in Crossed-Field Diodes Christopher Fichtl, Timothy Fleming, Keith Cartwright, Christopher Lenyk The initial velocity that an electron has from the cathode can change the magnetic field needed to insulate a crossed-field diode\footnote{ A. W. Hull, \textit{Phys. Rev}. 18, 31 (1921)}. For a secondary emitting cathode the distribution of emitted electron velocities depends on the velocity distribution of electrons returning to the cathode. We have studied the evolution of the Brillouin hub in a crossed-field diode in self-consistent 1d electromagnetic Particle-in-Cell (PIC) code with thermal emission and a secondary emission model by Vaughan\footnote{ J. R. M. Vaughan, \textit{IEEE Trans. on Electron Dev}., Vol. 36, no. 9, 1989.}. The baseline simulations have thermal emission with 0.1 eV of temperature; this is compared to simulations that have both secondary emission and thermal emission. The fraction of reflected and back scattered primaries is varied to induce perturbations in the Brillouin hub. The temperature of the true secondaries is conveniently set to the thermal emission temperature of 0.1 eV. It is found that relatively few ($\sim $10{\%}) reflected and back scattered primaries allow the Brillouin hub to expand further across the diode as compared to the thermal emission cathode. [Preview Abstract] |
|
CP1.00073: Single Particle Motion in Systems with Spatially Rapidly Varying Fields Harold Weitzner Earlier work, {\it Phys. Plasmas} {\bf 12}, 012106 (2005), showed that adiabatic invariants may exist even in certain systems with large gradients in the underlying electromagnetic fields. Those results were highly implicit and hard to employ for further theoretical developments. This work shows that extensions may allow gradient lengths of order $\sqrt{\rho L}$, where $\rho$ is the larmor radius parameter and $L$ a macroscopic distance. The underlying fields may vary with this parameter in one coordinate only, say a flux variable, while turbulent electromagnetic fields of amplitude $\sqrt{\rho L}$ and scale length $\sqrt{\rho L}$ may also be present and depend on all coordinates. Such turbulent fields would have $k_{\perp} \rho \sim \sqrt{\rho /L}$, that is not quite so large as 1, but easily of order .1 or .2. The model of single particle motion is fully nonlinear and explicit in the amplitude of the turbulent electromagnetic potentials. The associated turbulence models would be far more explicit than the earlier work or the conventional gyrokinetic theories. [Preview Abstract] |
|
CP1.00074: Polarization and magnetization effects in reduced plasma dynamics Alain Brizard The concepts of polarization density and current, normally associated with the presence of a fluctuating electric field in a strong magnetic field, play important roles in the description of reduced plasma kinetic and fluid models. The addition of fluctuating magnetic fields introduce covariant generalizations of polarization effects as well as introduce the concept of magnetization current in a self-consistent way. In the present work, polarization and magnetization effects are investigated within the variational formulations of several self-consistent reduced kinetic and fluid plasma models: oscillation-center and gyrokinetic plasma kinetic models and finite-$\beta$ electromagnetic drift-fluid models. [Preview Abstract] |
|
CP1.00075: Parallel closures for plasma fluid equations E.D. Held, M.K. Addae-Kagyah, J.J. James, M. Sharma, D.R. Hatch, J.-Y. Ji Accurately incorporating the effects of rapid electron and ion motion along magnetic field lines in plasma fluid models requires a kinetic treatment that admits arbitrary collision and transit frequencies. Coupled solutions to the electron and ion drift kinetic equations (DKE's) are presented which permit the construction of accurate closure relations for plasma fluid equations. The linearized collision operator employed in this work includes both particle/field and field/particle terms which guarantee accurate transport coefficients in the collisional limit. A gyroaveraged moment expansion for the kinetic distortions is used to approximately treat speed diffusion and drag terms as well as the field/particle elements of the collision operator. In addition, flow conservation is provided for by nonlocal momentum restoring terms which couple to the closure moments. It is shown that the parallel heat flow and stress closures map continuously from the collisional to the nearly collisionless limits. It is emphasized that in the present closure scheme, thermodynamic drives are considered simultaneously thus generalizing previous work that considered the heat flow and stress drives separately. The extension of this work to include time-dependent effects as well as the generalization to toroidal geometry is discussed.\newline *Supported by the US DOE grant nos. DE-FG02-04ER54746, DE-FC02-04ER54798, and DE-FC02-05ER54812. [Preview Abstract] |
|
CP1.00076: Inhomogeneity scale lengths in a collisionless, Q-machine plasma column E.W. Reynolds, M.E. Koepke, S. Shinohara Radial inhomogeneity scale lengths for radial electric field, ion density, and magnetic-field-aligned electron drift velocity have been measured and interpreted in magnetized, low temperature, collisionless, Q-machine plasma. When the ion Larmor radius \textit{$\rho $}$_{i}$ is on the order of [much smaller than] the half-width-at-half-maximum $\ell _r \{E(r)\}$ of the radially-localized radial electric field profile $E(r)$, the radial profile of the azimuthal ion drift velocity, measured using laser-induced-fluorescence (LIF), has a peak $V_{\theta }$ that, because of finite [negligible]-Larmor-radius effects, is significantly lower than [comparable to] the peak $V_{sum}$ of the combined radial profile of the \textbf{\textit{E}}x\textbf{\textit{B}} and diamagnetic drift velocities. For example, when \textit{$\rho $}$_{i}$/$\ell _r \{E(r)\}$=1 [2] we find that $V_{\theta }$/$V_{sum}$=0.58 [0.25]. Results of an experimentally benchmarked test-particle simulation are presented and applied using experimentally relevant electric field profiles. Work supported by U.S.-NSF and the US-Japan Exchange Program. [Preview Abstract] |
|
CP1.00077: Stopping Power for Arbitrary Angle Between Test Particle Velocity and Magnetic Field Carlo Cereceda, Michel Deperetti, Claude Deutsch Using the longitudinal dielectric function derived previously for charged test particles in helical motion around magnetic field lines, the numerical convergence of the series involved is demonstrated and the double quadrature on wave vector components is performed yielding the stopping power for arbitrary angle between the test particle velocity and the magnetic field orientation. Calculations are performed for particle Larmor radius larger or shorter than electron Debye length, i.e. for protons in a cold magnetized plasma and for thermonuclear alpha particles in a dense, hot, and strongly magnetized plasma. A strong decrease is found for the energy loss as the angle varies from 0 to Pi/2. The range of thermonuclear alpha particles is also discussed in terms of the velocity angle w.r.t. the magnetic field orientation. [Preview Abstract] |
|
CP1.00078: Alpha-driven localized cyclotron wave modes in nonuniform magnetic field: challenge views on resonance K.R. Chen, Z.H. Tsai Resonance is a fundamental issue in science. It requires precise synchronization. Our recent results impact our understanding of resonance. Relativistic ion cyclotron instability, driven by MeV ions whose Lorentz factor is very close to unity, requires a small negative frequency mismatch between harmonic ion cyclotron motion and wave. Thus, it is generally believed that it can not survive the nonuniformity of magnetic field such as in realistic devices. However, our simulations have shown that localized cyclotron waves are excited by the relativistic instability when the magnetic field is with a sinusoidal nonuniformity which is much larger than the frequency mismatch required. The localized waves do not overlap and appear as spatially separated wavelets. This indicates that resonance is a consequence of the need to drive instability for dissipating free energy and increasing the entropy. When a favorable wave eigen-frequency is collectively decided in a coherent means, a special wave form in real space is created for this purpose, even without boundary. The implications for developing a wavelet kinetic theory and for resonance physics will also be discussed. [Preview Abstract] |
|
CP1.00079: Comparative simulation and theoretical studies of alpha-driven localized cyclotron wave modes in nonuniform magnetic field Z.H. Tsai, K.R. Chen, J.Q. Dong Particle-In-Cell simulation and gyro-kinetic theory have been developed to study the localized cyclotron wave modes. The localized modes are driven by MeV alpha particles via the excitation of relativistic harmonic ion cyclotron instability in nonuniform magnetic field. The nonuniformity is required to be larger than both the frequency mismatch and the difference between the Lorentz factor and unity for the excitation becoming localized. The high-resolution PIC simulation illuminates the spatial structure with the help of the quiet start numerical technique for reducing the noise level of the simulation. The gyro-kinetic theory derives an integral dispersion relation, which is solved numerically. The eigen-frequency and resultant spatial structure of the localized wave modes as well as the dependence of wave mode characteristics on plasma parameters are comparatively studied in detail. [Preview Abstract] |
|
CP1.00080: Particle-in-cell Simulation of Langmuir Probes Felipe Iza, Jae Koo Lee Ion kinetics in the sheath and pre-sheath of planar and cylindrical probes has been studied by means of 1-dimensional (1d3v) particle-in-cell Monte Carlo collision simulations. Collisionless and collisional regimes are considered and simulation results (floating potentials and the ion saturation currents) are compared with available theories. As pressure increases, the ion velocity at the sheath edge decreases below the Bohm velocity (u$_{B})$. For planar probes, this velocity is $\sim $ u$_{B}$(1+5$\lambda _{De}$/$\lambda _{i})$ where $\lambda _{De}$ is the Debye length at the sheath edge and $\lambda _{i}$ the ion mean free path. Although ionization can be neglected in the sheath region, it plays a key role in determining the voltage across the presheath. For planar probes and Maxwellian electrons, this voltage increases rapidly for electron temperatures below $\sim $2eV. For cylindrical probes, however, the voltage across the presheath can be drastically reduced by the geometrical increase of current density as ions approach the probe. At low pressure, simulation results lie between the Laframboise and the ABR theories. As pressure increases, however, collisions and ionization in the presheath becomes critical in determining the ion flux to the probe at a given bias voltage. [Preview Abstract] |
|
CP1.00081: Laboratory Study of Fast Collisionless Magnetic Reconnection in the New VTF Closed Magnetic Configuration J. Egedal, W. Fox, N. Katz, M. Porkolab A new magnetic configuration utilizing in-vessel coils has been implemented in the Versatile Toroidal Facility at MIT [1] for the study of magnetic reconnection in the collisionless regime ($n_e\sim10^{18}$ m$^{-3}$ and $T_e\sim60$eV) . In a range of the available parameter regime the reconnection process occurs in rapid bursts, with Alfvenic outflow velocities. The abrupt onset of these reconnection events provides a unique opportunity to address the ``fast trigger problem in reconnection.'' We will present our preliminary analysis of the magnetic and electrostatic structures of the reconnection region during these spontaneous reconnection events.\\[1ex] [1] J. Egedal, et al., Rev. Sci. Instrum. 71, 3351 (2000). [Preview Abstract] |
|
CP1.00082: Fine scale structure in the current sheet and electrostatic fields during driven magnetic reconnection on the VTF experiment. William Fox, Jan Egedal, Noam Katz, Miklos Porkolab We have conducted a series of experiments in the VTF reconnection experiment[1] to measure with high resolution the current channel and electric structures that form in response to driven reconnection. Preliminary measurements have revealed that the current sheet is not symmetric across the X-line, contradicting an assumption fundamental to nearly every reconnection theory. Importantly, effects related to this asymmetry can account for momentum balance for the electrons at the X-line (i.e. fulfillment of the generalized Ohm's law) via convective inertia ($m n v_\perp \cdot \nabla v_{||}$). Measurements of strong in-plane electric field structures ($E_\perp \sim $ 1~kV/m) near the X-point reveal a mechanism to efficiently heat ions, as has been recently observed by laser induced fluorescence (LIF) measurements of the ion distribution function[2].\\ \\ This work was supported by a DoE Fusion Energy Sciences Fellowship.\\ \\ $[1]$ J. Egedal, {\it et. al.} (2001), {\it Rev. Sci. Instrum.} 71, 3351 \\ $[2]$ A. Stark, W. Fox, J.Egedal, O. Grulke, T. Klinger, (2005), submitted to Phys. Rev. Lett. [Preview Abstract] |
|
CP1.00083: Fast Magnetic Reconnection Catalyzed by Trapped Electrons in Space and Laboratory Plasmas. N. Katz, J. Egedal, W. Fox, M. Porkolab The experimental studies in the open cusp configuration of the Versatile Toroidal Facility at MIT [1] have revealed a new paradigm for magnetic reconnection. Here the reconnection process is controlled and catalyzed by the dynamics of trapped electrons. The experimental data includes a rigorous study of the size and geometry of the electron diffusion region and the anomalously small current densities observed at the X-line. The low current densities are consistent with our theoretical studies solving the Vlasov equation in the measured electric and magnetic fields. A similar kinetic analysis has been applied to a unique data set obtained during reconnection in the Earth’s magnetotail by the WIND spacecraft. In this latter case the reconnection process is also controlled by electrostatically trapped electrons [2]. \\[1ex] [1] J. Egedal, et al., Rev. of Sci. Instrum. 71, 3351 (2000).\\[1ex] [2] J. Egedal, et al., Phys. Rev. Lett. 94, 025006 (2005). [Preview Abstract] |
|
CP1.00084: 3-D Vector Tomography Diagnostics for Spherical Tokamacs Alexander Balandin, Yasushi Ono Plasma emits radiation over a broad frequency range. Characteristics of the radiation are related to plasma parameters then may be applied for diagnostics. These methods are quite attractive since they do not introduce any perturbation in to plasma. In a such complex objects as plasma, the spectroscopic measurements, for example, can contain information concerning the distribution of both parameters of scalar and vector (velocity) fields. In this report, the numerical method is proposed for the inversion of the vector X-ray transform of 3-D vector fields, which is based on series expansion method. We first of all discuss the Central Slice Theorem for the vector X-ray transform and, as the corollary it is obtained that the irrotational component of the vector field gives no contribution to the path-integral, leaving only the contribution from the solenoidal part. The tomographic reconstruction of 3-D vector fields is performed by the two different methods: expansion of components of the vector field into the orthogonal system of scalar spherical harmonics and in the second one is used the decomposition over the orthogonal system of vector spherical harmonics. The first method can be used for the reconstruction of both vector field parameters and scalar ones by solving large system of linear equations with respect to expansion coefficients. The second one are applied for reconstruction only vector fields. The computer simulation, demonstrating the 3-D reconstruction of the model vector fields is presented. [Preview Abstract] |
|
CP1.00085: Laboratory Investigation of Magnetotail Reconnection Sarah Messer, W.E. Amatucci, L. Rudakov, D.D. Blackwell, D.N. Walker An experimental investigation of 3-dimensional magnetic reconnection with non-dimensional parameters comparable to those observed in Earth's magnetosphere is underway in the Space Plasma Simulation Chamber (SPSC) at the Naval Research Laboratory. Reconnection events are studied in a steady-state cylindrical plasma column with a DC magnetic field of 60 Gauss. We pulse an oppositely-directed axial magnetic field rising to about 200 Gauss in 3 $\mu $s. Magnetic reconnection arises as the field-reversed wavefront propagates inward. An array of magnetic probes allows examination of the field structure of individual events. We present both aggregate and individual-shot analyses and compare the gross structure to predictions of Hall MHD and anomalously-resistive models of reconnection. The theory of the different dynamical processes will be presented alongside experimental observations. [Preview Abstract] |
|
CP1.00086: Fast Reconnection in Electron-Positron Plasmas N. Bessho, A. Bhattacharjee We present 2D studies of magnetic reconnection in electron-positron (or pair) plasmas in which there is no equilibrium guide field, using an electromagnetic particle-in-cell code. In the generalized Ohm's law for pair plasmas, the Hall term is absent because of an exact cancellation that occurs due to the equality of electron and positron masses. Hence, there are no whistler waves. We demonstrate that fast reconnection can be realized even in such a case, which breaks the link between fast reconnection and dispersive whistler waves. The quadrupolar structure of the out-of-plane magnetic field, which is often used as a signature of collisionless reconnection, is absent in pair plasmas. We show that the off-diagonal terms in the pressure tensor for both electrons and positrons are dominant in the generalized Ohm's law, that they are localized in the diffusion region, and that such localization is key to fast reconnection. The pressure tensors and inertial terms produce essentially a large effective resistivity in the diffusion region, which facilitates the formation of an X-point and realizes fast reconnection. [Preview Abstract] |
|
CP1.00087: The Magnetic Reconnection Code: Application to the Sawtooth Instability in Tokamaks K. Germaschewski, A. Bhattacharjee The Magnetic Reconnection Code (MRC), which integrates numerically the compressible Hall MHD or two-fluid equations in flexible geometry, is a massively parallel code in an Adaptive Mesh Refinement (AMR) framework. The code implements recent developments in advanced numerical methods. AMR is an effective tool to attain the resolution necessary at small scales of the reconnection layer(s), and also provides the basis for domain decomposition needed to run simulations on massively parallel machines. Incorporating two-fluid effects comes with two numerical challenges. First, the system loses its purely hyperbolic character and becomes partially elliptic. Second, dispersive waves are introduced into the system. We apply Newton-Krylov-Schwarz methods to implicitly advance the fields in time, avoiding a stringent CFL condition and gaining additional numerical stability. We have used the cylindrical MRC to study the effect of the Hall current and electron pressure gradient on the m=1 tearing mode. While we recover the near-explosive nonlinear growth observed in A. Aydemir's four-field studies under certain conditions, we also see effects associated with a poloidally asymmetric flow that can be stabilizing linearly as well as nonlinearly. The regimes of near-explosive nonlinear growth are delineated, and comparisons will be made with large tokamak experiments. [Preview Abstract] |
|
CP1.00088: Large box gyrokinetic $\delta \! f$ simulation of field line reversing plasmas Weigang Wan, Yang Chen, Scott Parker Instabilities caused by the filed line reversing configuration are studied using a three-dimensional particle-in-cell simulation model that utilizes the $\delta \!f$-method.\footnote {Y. Chen and S.E. Parker, J. Comput. Phys. {\bf 198}, 463 (2003)} We use the model of drift-kinetic electrons and gyrokinetic ions. In the simulation with box size of 64 ion gyro radii and a wide equilibrium current, we have observed a linear instability that is not seen in the previous simulation of smaller box size. In small box simulations,\footnote {W. Wan, Y. Chen and S. E. Parker, Phys. Plasmas {\bf 12}, 012311 (2005)} the ions response can be neglected but for large box simulations an instability which is different from tearing mode is observed with gyrokinetic ions response. However, in previous studies the parallel direction is not distinguished from the external guiding field, and it is not clear whether this instability is physical or numerical. To resolve this issue, we propose to use the field line following coordinates, to clearly separate parallel and perpendicular motions, and solve the field equations strictly in the perpendicular plane. We also studied the nonlinear magnetic island evolution in a otherwise linearly stable plasma by initializing a big island. Nonlinear instability and some clues of kinetic Alfv\'{e}n waves are observed. [Preview Abstract] |
|
CP1.00089: On the rotation of magnetic islands in a tokamak X. Garbet, A. Smolyakov, E. Lazzaro, M. Ottaviani We investigate various mechanisms affecting the rotation of magnetic islands. Generally, the rotation of a magnetic island is determined by a $\sin \left[m(\theta -\zeta/q) -\omega t\right] $ component of the non-ambipolar electric current. We consider several processes leading to the transverse non- ambipolar electric currents in a tokamak such as small-scale electromagnetic turbulence, anomalous viscosity, and neoclassical transport. It is argued that in a toroidal plasma the non-ambipolar neoclassical transport is a dominant process. It results in the locking of the magnetic island into the ion poloidal motion. [Preview Abstract] |
|
CP1.00090: Perturbative analysis of tearing mode saturation Fulvio Militello, Maurizio Otaviani, Franco Porcelli, Jim Hastie New, rigorous analytic results for the tearing island saturation are presented. These results are valid for the realistic case where the magnetic island structure is non-symmetric about the reconnection surface and the electron temperature, on which the electrical resistivity depends, is evolved self-consistently with the island growth. The new equilibrium, represented by the saturated island, is constructed using a perturbation expansion which does not need an \textit{Ansatz} on the shape of the magnetic field, since it takes into account self-consistently the complete harmonic structure of the mode in the nonlinear layer. The new terms in the Rutherford equation, which are obtained with this procedure, are likely to have an impact on the overall saturation level of NTMs. [Preview Abstract] |
|
CP1.00091: Coalescence of Magnetic Islands in the High Lunquist Number, Hall MHD, Regime Dana Knoll, Luis Chacon The coalescence of magnetic islands in the high Lunquist number, Hall MHD, regime is studied. Within the resistive MHD model the coalescence rate (reconnection rate) is known to be independent of Lunquist number over a range of Lunquist numbers. Biskamp [1] has shown that the coalescence rate will stall (sloshing) in the limit of high Lunquist number (S on the order of \(10^{5}\)). This stalling has been associated with the onset of a secondary tearing mode [1]. We will demonstrate this stalling results from high magnetic pressure gradients, which result from thin current sheets. The high magnetic pressure shuts down the ion flow into the current sheet, which effectively limits the transport of fresh flux into the reconnection site. We then simulate the problem with a Hall MHD model. We demonstrate that when the current sheet scale length, \(\delta _{J}\), reaches ``Hall MHD scales'' ( i.e. \(\delta _{J} < d_{i}\), where \(d_{i}\) is the ion inertial length ) prior to the onset of sloshing, sloshing will be avoided and island coalescence will proceed at a rate independent of Lunquist number. Note that \(d_{i}\) may be small for the intial conditions of the problem. \newline \newline [1] D. Biskamp, Phys. Lett. A, vol. 87, pp. 357-360 (1982) [Preview Abstract] |
|
CP1.00092: On the Excitation of Drift-Tearing Modes in High-Temperature Plasmas V. Roytershteyn, B. Coppi, C. Yarim Contrary to relevant experimental observations in high-temperature toroidal plasmas [1], both the collisionless and weakly collisional theory of drift-tearing modes [2] in magnetically confined plasmas predict [3,4] that in the presence of a significant radial electron temperature gradient these modes should not be excited, as a result of the effects of electron Landau damping and parallel electron thermal conductivity. To reconcile the experimental observations with the theory, we propose that the presence of a background microturbulence allows for the excitation of the reconnecting mode. We have demonstrated that such effects as a local depression in the parallel electron thermal conductivity or, less likely, a local flattening of the electron temperature profile can lead to a significant reduction in the excitation threshold. In addition, a modest transverse electron thermal conductivity that can be driven by I.E.T.G. modes [5] can change significantly the topology of the drift-tearing mode and can lead to positive growth rates for realistic current density and plasma pressure distributions. [1] J. A. Snipes \textit{et al.}, \textit{Plasma Phys. Cont. Fus.}, \textbf{44} 381 (2002) [2] B. Coppi \textit{Phys. Fluids}, \textbf{8} 2273 (1965) [3] B. Coppi et al., \textit{Phys. Rev. Letters}, \textbf{42}, 1058(1979); J. Drake, \textit{et al., Phys. Fluids}, \textbf{25}, 2509 (1983) [4] V. Roytershteyn, \textit{et. al.}, Paper P2-27, 2005 Sherwood Conference [5] C. Yarim, \textit{et al.}, this conference [Preview Abstract] |
|
CP1.00093: Modulated Lower Hybrid Current Drive Suppression of MHD m=2 modes on HT-7 J.S. Mao, H. Wang, J. Zhao, Y.W. Sun, B.J. Ding, J.R. Luo, P.E. Phillips A crucial issue for the extension of advanced tokamak scenarios to long pulse operation is the avoidance of Magneto-Hydrodynamics (MHD) activity. Active Modulation of Lower Hybrid Current Drive (LHCD) was used successfully to suppress MHD activity on HT-7, a superconducting tokamak. A feedback system was used which activated modulation only when MHD activity reach a predetermined level so as to optimize the current drive through out the discharge. The LHCD modulation was varied in power and frequency, with the frequency always being less than the resistive skin time (100ms). Optimal MHD suppression was achieved when modulating LHW power $ > 200kW$, $\Delta \theta=120^\circ$ modulation, and frequency of $50Hz$. Details of the MHD suppression will be discussed in this poster. [Preview Abstract] |
|
CP1.00094: Nonlinear three-dimensional simulations of resistive wall modes with flow Masahiko Sato, Noriyoshi Nakajima Nonlinear resistive wall modes (RWMs) in low beta cylindrical tokamaks are simulated by using the reduced MHD equations. The pseudo-vacuum model is used, where the vacuum is replaced by a highly resistive plasma. First, we will show single helicity results. When the initial poloidal rotation is small, the `wall mode,' which is nearly locking to the resistive wall, is unstable. As the initial poloidal rotation become large, `plasma mode,' which is rotating against the resistive wall, becomes unstable. The electromagnetic torque damps the poloidal rotation. For `wall mode', the torque increases as the initial poloidal rotation becomes large. In this case, the final saturation level dose not depend on the initial poloidal rotation frequency due to reduction in the poloidal rotation near the plasma edge. On the other hand, for `plasma mode,' the torque decreases as the initial poloidal rotation. Thus, the poloidal rotation remains in the nonlinear stage and the final saturation level decreases. Next, we will show multi helicity results. Magnetic field line stochastization is obtained due to nonlinear coupling (m,n)=(3,1) mode and (m,n)=(5,2) mode. Effect of poloidal rotation on the extent of the stochastic region will be shown. [Preview Abstract] |
|
CP1.00095: Design of a Flow Switch for Optimized X-Ray Yield on Z Sophie Chantrenne, Michael Cuneo, Thomas Haill, Thomas Mehlhorn, Thomas Nash, Eduardo Waisman In wire array z-pinch applications it may be desirable to have a faster rising current pulse to decrease the wire array implosion time. A reduced implosion time will result in a higher implosion velocity and should therefore increase the peak radiated x-ray power at stagnation. Faster rising current pulses may also permit high x-ray power from smaller diameter wire arrays that could enhance the x-ray drive temperature from more compact hohlraums. In this work, we present results from a design study of a plasma flow switch for the Z accelerator. We would like to compress the pulse by a factor of two with a flow switch that allows the current to be stored in the vacuum for up to 50 ns before it is delivered to a wire array load on the Z Accelerator. We use 2D r-z simulations of a flow switch with the ALEGRA MHD code. Different geometries, flow switch mass, and wire array mass and other conditions were tested and compared in order to obtain an initial flow switch design. [Preview Abstract] |
|
CP1.00096: Plasma rotation in a magnetic nozzle Boris Breizman, Alexey Arefiev, Ahmet Aydemir This work generalizes the MHD description of an axisymmetric plasma flow in a magnetic nozzle [1] to the case of rotating flows. The generalization requires that the azimuthal magnetic field and radial electric field be taken into account. In the absence of a radial electric field, the MHD equations formally allow for a steady-state solution with the plasma velocity directed along the twisted magnetic field lines. However, such a solution is singular at the surface of sub- to super-Alfvénic transition. In the presence of a radial electric field, the azimuthal component of the plasma velocity must increase downstream along diverging magnetic flux tubes, because the electrostatic potential remains constant along the magnetic field lines. A regular physical solution is not possible in this case as well. In order to understand the role of plasma rotation and to understand whether perturbations can be accumulated at the surface of sub- to super-Alfvénic transition, we have examined the dynamics of small time-dependent rotating perturbations created by an external source in a steady-state plasma flow. [1] A.V. Arefiev and B.N. Breizman, Phys. Plasmas 12, 043504 (2005). [Preview Abstract] |
|
CP1.00097: Corrections to Linear Hall MHD Arising From Heat Flow and Pressure Anisotropy Jason TenBarge, Swadesh Mahajan, Richard Hazeltine Previous studies of the solutions to linear Hall MHD have neglected heat flow and pressure anisotropy. Although ignoring these quantities is safe in certain limits, in general heat flow and pressure anisotropy have the potential to play major roles in the evolution of a plasma system. Here we present a study of a two-fluid quasineutral, fully-conductive system with the inclusion of electron heat flow and pressure anisotropy. The ions receive the Chew, Golberger, Low treatment (pressure anisotropy but no heat flow). To include the effects of heat flow and pressure anisotropy, we employ appropriate fluid equations, derived previously by Hazeltine and Mahajan (2002). We find corrections to the standard Hall MHD solutions, from which the standard Hall MHD results presented in Ohsaki and Mahajan (2004) are recoverable when the fluid equations decouple from the equations of motion. \begin{thebibliography}{10} \bibitem{haz802}R. D. Hazeltine and S. M. Mahajan, Phys. Plasmas \textbf{9}, 3341 (2002). \bibitem{ohmah304}S. Ohsaki and S. M. Mahajan, Phys. Plasmas \textbf{11}, 898 (2004). \end{thebibliography} [Preview Abstract] |
|
CP1.00098: Numerical Studies of Gyroviscous Effects Using High-Order Finite Elements Nathaniel Ferraro, Steve Jardin, Joshua Breslau, Jesus Ramos We have developed a technique for incorporating a general expression of the gyroviscous force\footnote{J.~J. Ramos, PSFC/JA-05-9, MIT (2005).} into an implicit solution algorithm for the two-fluid magnetohydrodynamic (MHD) equations. We present the results of numerical simulations of six-field extended-MHD equations in two dimensions, including Braginskii's gyroviscous stress tensor, using triangular finite elements with fifth-order accuracy and continuous first derivatives ($C^1$-continuity). Our model extends that used by Jardin and Breslau\footnote{S.~C. Jardin and J.~A. Breslau, Phys. Plasmas \textbf{12}, 56101 (2005).} by including the evolution of pressure and flow compressibility, in addition to the inclusion of the gyroviscous force. The use of $C^1$-continuous finite elements allows up to four differentiations of any field variable, thus enabling the inclusion of the full gyroviscous stress tensor. The effect of this term on wave propagation and Harris-equilibrium reconnection is demonstrated. [Preview Abstract] |
|
CP1.00099: Formation of Filamentary Structure during ELMs in Spherical Tokamak Takaya Hayashi, Naoki Mizuguchi, Riaz Khan We execute nonlinear MHD simulations in full toroidal geometry to study relaxation phenomena that occur in the edge region of spherical tokamak. The simulation of a resistive MHD regime reproduces the formation of filamentary structure; the numbers of filament observed are consistent with the experimental observation. Our simulation supports the ballooning nature of filamentary structure, erupted on steep pressure gradients in the edge region as observed in recent experiments of MAST and NSTX. About 9 fingers appear in the nonlinear phase. In results eruption of filament occur from the ridges of finger. By adding a simple model of FLR term, successful stabilization of higher mode number in the linear phase and sudden excitation of those higher modes in the nonlinear phase are observed. The excitation of the higher modes is considered to be induced by nonlinear couplings among dominant lower modes. The plasma flow pattern in the MHD case follows a twin vortex pattern in the nonlinear phase, where as in the FLR case it follows a single vortex pattern. The growth rate changes with the introduction of the FLR term, both in the linear as well as the nonlinear phase. [Preview Abstract] |
|
CP1.00100: MHD simulations and theory for sheared flows and turbulence in the Helimak R. Dahlburg, W. Horton, J.C. Perez, K. Gentle The Helimak experiment was designed to study the interaction between sheared mass flows and ambient turbulence in a confined plasma. The experiment is well modeled by the classical slab used in theoretical analysis of flows and turbulence in tokamaks and channel flows. We report results of numerical simulations and theory for a magnetohydrodynamic slab model of the Helimak, using magnetic and flow profiles based on experimental data. Important features of the model include the three spatial dimensions, the presence of walls, and the inclusion of resistivity and viscosity. Linear results, computed with a Chebyshev-$\tau$ algorithm, indicate the existence of traveling unstable disturbances. We analyze the linear results by examining the stresses and perturbed dissipation. We also report progress on nonlinear simulations using our new Chebyshev-collocation--Fourier-pseudospectral code. Preliminary results show that as the linear modes attain finite amplitude, there is considerable development of multiscale plasma excitation. [Preview Abstract] |
|
CP1.00101: Nonlinear dynamics of the tearing mode for any current gradient Dominique Escande, Nicolas Arcis, Maurizio Ottaviani Within the traditional frame of reduced MHD, a new systematic perturbation expansion provides the equation ruling the nonlinear growth and saturation of the tearing mode for any current gradient. The small parameter is the magnetic island width $w$. The evolution equation depends on $w$ through a term $w \ln(w_0/ w)$ where $w_0$ is a nonlinear scale length which was absent in previous works. The technique is applicable to the case of an external forcing. The solution for a static forcing is computed explicitly and it exhibits three regimes in the dynamics. A simpler version of Thyagaraja's technique yields an independent confirmation of the unforced case. [Preview Abstract] |
|
CP1.00102: Finite Ion Orbit Effects on Magnetic Islands in Toroidal Plasmas Xinzheng Liu, Chris Hegna A kinetic theory for the interaction of an ion population with an isolated magnetic island in a tokamak plasma is presented. In this work, we examine islands whose characteristic widths are larger than the ion gyro radius but smaller than the ion banana width. In this regime, the ion response to the island has a non-local feature due to the curvature and B drifts. When solving the drift kinetic equation for the energetic ions, a change in coordinates is used to account for this behavior. A bounce averaging procedure is developed to separate out and solve the lower order distribution function. Constraint relationships found from transport equations and collision operators are used to determine the distribution function, which is treated in different velocity regions and rotation frequency values. The contribution to the perturbed current is composed of the helically flux surface-averaged bootstrap current and the perpendicular current, the neoclassical enhanced ion polarization current. The parallel current (J$_{\parallel})$ in response to the ions is calculated and compared with some recent numerical results [1]. Using this current in the ``dispersion relation'' found from asymptotic matching, the island width evolution equation is determined. A pair of self-consistent equations for the islands' width, $w$, and its propagation frequency, $\omega $, is to be derived. The results are to be compared with calculations valid for large island width. [1] E. Poli et al., Nuclear Fusion \textbf{45}, 384 (2005). [Preview Abstract] |
|
CP1.00103: Measurements of current-free double layers in two expanding helicon plasma sources Xuan Sun, Earl Scime, Samuel Cohen, Mahmood Miah This work concerns measurements of parallel ion flow in expanding plasmas generated by two different helicon plasma sources. The measurements indicate the existence of a current-free double layer in a region of diverging magnetic field. With laser-induced fluorescence (LIF), the double layer structure in both helicon plasma sources was investigated through measurements of the bulk parallel ion flow speed. Both double layers have a total potential drop of 3-4\textit{kT}$_{e}$. A stronger double layer, with a potential drop of $\sim $ 6\textit{kT}$_{e}$, was created in a uniform magnetic field region with a plasma limiting aperture plate. [Preview Abstract] |
|
CP1.00104: Momentum transport from reconnection in flowing plasmas Fatima Ebrahimi, S.C. Prager, V.V. Mirnov, C.R. Sovinec Magnetic fluctuations arising from MHD instabilities can cause momentum transport in both laboratory and astrophysical plasmas. We investigate, both analytically and numerically, momentum transport in force-free magnetic configurations using the Maxwell stress associated with tearing instabilities. We study transport from single tearing modes in the linear and nonlinear regimes, and transport arising from nonlinear mode coupling. Analytic quasilinear theory, based on matching of inner layer and outer region equations, shows that in the absence of equilibrium flow no momentum transport occurs, but that momentum transport accompanies tearing modes in plasma with equilibrium flow. The resistive, 3D, MHD code, DEBS, is used to evaluate transport in the full nonlinear state. An ad-hoc term is added to the momentum equation to represent a source of plasma flow, and the nonlinear flow dynamics is studied. Supported by NSF and DOE. [Preview Abstract] |
|
CP1.00105: Imaging Equilibrium and Perturbed Flow of Dipole-Confined Plasma B.A. Grierson, M.W. Worstell, M.E. Mauel Plasma confined in magnetospheres and laboratory terrella can be characterized by the loss current to the magnetic poles. In the Collisionless Terrella Experiment (CTX) the polar current is imaged at high speed with an array of 70 gridded particle detectors. The detectors measure electron and/or ion currents as a function of the particle's field-aligned energy and of the polar longitude and latitude. This poster will present new techniques to resolve the equilibrium and perturbed plasma structure from analysis of the detected polar currents. Fast gas injection is used to perturb the plasma with a neutral gas, and the interaction of the neutral population with the hot-electron interchange motion has been observed. Measurements of global plasma density evolution, plasma flow, and mixing time-scales will be reported. [Preview Abstract] |
|
CP1.00106: Driving Interchange Flow in a Dipole-Confined Plasma M.W. Worstell, B.A. Grierson, M.E. Mauel The Collisionless Terrella Experiment (CTX) is a device utilizing a dipole field to study interchange and flow. Instabilities excited by hot electrons\footnote{Levitt, \textit{et al.}, Phys. Plasmas, {\bf 9}, 2507 (2002).} and centrifugal forces\footnote{Levitt, \textit{et al.}, Phys. Rev. Lett., {\bf 94}, 175002 (2005).} have been observed. This poster investigates several techniques to excite interchange motion and to modify the geometry of the resulting flow patterns. One method utilizes six meshes inserted at the inner edge of the plasma around the magnetic equator. Application of a non-axisymmetric bias to these meshes can cause the plasma at different radii to mix. A second approach uses a large voltage applied to an oversized probe in order to localize the convective motion of the plasma. First results from these investigations will be presented. [Preview Abstract] |
|
CP1.00107: Magnetic Nozzle Effects for High Power Helicon Plasma Propulsion Systems Robert Winglee, Timothy Ziemba, James Prager, Nathan Stobie, Race Roberson, John Carscadden The high power helicon (HPH) is an electrode propulsion system that utilizes non-linear helicon/whistler waves to accelerate the plasma at very high densities. Because the plasma is generated in a magnetized environment at high beta, HPH has an inherent feature that the magnetic nozzles can be used to focus the plasma and increase its efficiency. Laboratory results and computer simulations are used to quantify this focusing and the changes the plasma characteristics as a function of the position and field strength on the magnetic nozzle. It is shown that the Isp of the system can be increase by 20-30{\%} by the addition of the nozzle due to the conversion of thermal energy into directed energy. As the plasma leaves the nozzle, the plasma is supersonic (mach $\sim $5) and super-Alfv\'{e}nic. This parameter region means that there are no problems associated with attachment of the plasma to the magnetic field lines and leads to a highly collimated beam well downstream from the magnetic nozzle. This collimation opens up additional applications for the high power helicon. [Preview Abstract] |
|
CP1.00108: Generation and control of rotation in the flowing magnetized plasma (FMP) experiment Zhehui Wang, Jiahe Si The Flowing Magnetized Plasma (FMP) experiment at Los Alamos examines the physics of rotating plasmas inside magnetic field when the plasma thermal energy density is comparable to or exceeds vacuum magnetic energy density. Several types of instabilities, including Magneto- rotational instability (MRI), are expected based on theory. The growth rate of each instability is a function of temperature and density gradient, magnetic field and its gradient, rotation and rotation shear. Therefore, a key experimental issue is to drive plasma rotation and tailor rotation profile. A multi-MW coaxial gun, powered by a unique sub-kV electrolytic capacitor bank, produces a meter-size rotating plasma with $n_e >$ 1.0 x 10$^{19}$ m$^{-3}$ and T$_e$ $\sim$ 1-20 eV. Rotation and profile are further controlled by floating or biasing conducting boundaries at different electrical potentials, and by inserting an electrode along the symmetry axis of cylindrical plasmas. Plasma rotation is measured directly by probes, Doppler spectroscopy, and indirectly by fast camera imaging, and phase- shift of magnetic probes. Rotation frequency at the boundary, linearly increasing with axial vacuum magnetic field, is found at a few percent of ion gyrofrequency. Results of the rotation and profile as a function of bias electrical potential of different electrodes will be presented. Relevant rotation mechanisms, ExB and diamagnetic flows in particular, and implications to MRI, will be discussed. [Preview Abstract] |
|
CP1.00109: Linear analysis of Magneto-Rotational Instability and other instabilities in a plasma Couette Flow Jiahe Si, Zhehui Wang, Hui Li The ideal magnetohydrodynamics stability of a plasma Couette flow in which the density, pressure, magnetic field are non-uniform along radial direction, is studied. For simplicity, only axisymmetric perturbations are considered. Four destabilization mechanisms are present and studied by local analysis: Kelvin-Helmholtz instability, Schwarzschild instability, magnetorotational instability and Parker instability. It is shown different instability modes prodominate depending on the equilibrium parameters. With a weak magnetic field, high temperature and moderate velocity shear, MRI is predominate over other instabilities. [Preview Abstract] |
|
CP1.00110: Pedestal Stability Changes and Boundary Plasma Response During ELM Suppression by RMP in DIII-D M.E. Fenstermacher, T.E. Evans, T.H. Osborne, P.B. Snyder, K.H. Burrell, R.J. Groebner, A.W. Leonard, J.A. Boedo, R.A. Moyer, M. Groth, C.J. Lasnier, L. Zeng ELM suppression by application of n=3 edge resonant magnetic perturbations (RMP) has been obtained for both high and low collisionality plasmas in \hbox{DIII-D} using internal coils. The effect of the RMP on pedestal stability and on the divertor plasma appears to be different in the two cases. In the high collisionality case small islands are created in the edge, pedestal plasma profiles and the divertor radiation structure do not change much, but the plasma rotation is substantially reduced. At low collisionality the edge is predicted to be stochastic, $n_e$ and $T_e$ pedestals and the divertor radiation profiles are modified significantly, and the plasma rotation initially increases. The pedestal parameters in this case show several similarities with QH-mode but no EHO is observed. The implications of these differences on the pedestal stability and the divertor plasma will be presented. [Preview Abstract] |
|
CP1.00111: Effect of Edge Neutral Source Profile on H-mode Pedestal Height and ELM Size T.H. Osborne Although the steep temperature gradient region in the H-mode pedestal is representative of the transport barrier width, the shape of the density profile responds to changes in the particle source. At higher density, neutral penetration is reduced and the steep density gradient region narrows and moves outward relative to the temperature profile; this in turn narrows the high pressure gradient region. Experiments in which the pressure width was narrowed by gas puffing at otherwise fixed parameters, or at fixed pedestal collisionality and $q$ by increasing $I_p$, resulted in reduced pedestal beta and ELM size. Peeling-ballooning (PB) mode stability calculations, using the ELITE code, indicate the eigenmode width is reduced and the toroidal mode number increases (conditions generally associated with reduced ELM size) as the high pressure gradient region is narrowed and moved outward. For ITER at low pedestal collisionality, PB mode critical pressure gradient is expected to increase at reduced width, perhaps maintaining pedestal beta with smaller ELMs. [Preview Abstract] |
|
CP1.00112: Modification of Intermittency by External Perturbations J.A. Boedo, D.L. Rudakov, R.A. Moyer, E.M. Hollmann, T.E. Evans, A.W. Leonard, M.A. Mahdavi, W.P. West, J.G. Watkins, S.L. Allen, M.E. Fenstermacher, C.J. Lasnier, A.G. McLean It is desirable to spread the particle, and most importantly, the heat load to the divertor components. Any methods that allow a degree of control of the intermittent transport may be also applicable to ELMs, which can be seen as large blobs, and therefore be quite relevant for ITER. We present results from a study on effects of applied ergodic magnetic fields using the DIII-D non-axisymmetric coil sets. The application of an ergodic field results on broadening of the SOL $T_e$ and $N_e$ profiles as the intermittent transport becomes larger. The filaments become more frequent, carry more plasma and feature increased radial velocity. [Preview Abstract] |
|
CP1.00113: Implications of Stochastic Edge Transport for DIII-D I. Joseph, R.A. Moyer, T.E. Evans, M.J. Schaffer, A. Runov, R. Schneider, S.V. Kasilov The TRIP3D field-line tracing code is combined with the E3D two-fluid Monte-Carlo code in order to calculate the effect of resonant magnetic perturbations (RMPs) on \hbox{DIII-D} plasma profiles. RMPs have been experimentally shown to suppress ELMs at high collisionality and eliminate ELMs at low collisionality during \hbox{H-mode} operation. In the latter case, experiments show that the net effect of the RWP is to reduce the edge gradients below the stability boundary. In the absence of strong plasma response, TRIP3D shows that the RMPs break the outer 5\%-10\% of flux surfaces and allow field lines to directly connect to the divertor target. The E3D code can then accurately compute the change to the profiles caused by the stochastically enhanced perpendicular and direct parallel heat conduction. Preliminary results of both field line tracing and full transport calculations suggest qualitative agreement with the experimentally measured splitting of heat flux to the divertor. [Preview Abstract] |
|
CP1.00114: The Effect of Plasma Collisionality on Pedestal Current Density in DIII-D D.M. Thomas, A.W. Leonard, T.H. Osborne, R.J. Groebner, W.P. West, K.H. Burrell The stability of the H-mode pedestal is dependent on the edge pressure gradient and the edge current density. These terms are naturally coupled though neoclassical (Pfirsch-Schluter and bootstrap) effects. On DIII-D, local measurements of the edge current density using an injected lithium beam have confirmed the close spatial and temporal correlation that exists between the measured current density and the edge pressure in H- and QH-mode pedestals, as determined using Thomson scattering and charge exchange recombination spectroscopy. In this work we investigate the behavior of the edge current for DIII-D pedestals which have a range of values for the ion and electron collisionalities ($\nu_i^\ast , \nu_e^\ast $) due to shaping and fuelling effects. Such changes in the edge collisionality are expected to significantly alter the level of the bootstrap current from the value predicted from the collisionless limit, and hence should affect the pedestal stability limits. We also discuss the extent to which the measurements agree or disagree with specific bootstrap models. [Preview Abstract] |
|
CP1.00115: Divertor Target Plate Conditions During ELM Suppression Experiments using Resonant Magnetic Perturbations on DIII-D J.G. Watkins, T.E. Evans, R.A. Moyer, D.L. Rudakov, C.J. Lasnier This paper describes the target plate plasma conditions measured by a fixed array of Langmuir probes during ELM suppression experiments on DIII-D using resonant magnetic perturbations. At high collisionality, the target plate particle flux drops by 30\%, the temperature drops by 50\%, and the probe floating potential (V$_f$) goes to zero across the floor when the perturbation coil is energized. In low collisionality, after the ELMs are suppressed, the target plate particle flux profile broadens and increases by about 50\%. The electron temperature at the plate increases and the V$_f$ goes negative at the outer strike point and positive at the inner strike point when the ELMS go away. The background (between ELMs) heat flux calculated from the probe data increases $\sim$100\% after the ELMs are suppressed but the (5X) larger peaks associated with ELMS are no longer present. A steady-state, ELM-free, heat flux is desirable for actively cooled fusion reactor target plates. [Preview Abstract] |
|
CP1.00116: Effects of ELMs and ELM Suppressed Operation on Density Profiles in DIII-D L. Zeng, E.J. Doyle, G. Wang, T.L. Rhodes, W.A. Peebles, T.E. Evans, A.W. Leonard The dynamics of density perturbations associated with edge localized modes (ELMs) and ELM suppressed operation have been investigated on \hbox{DIII-D} using an upgraded profile reflectometer system with high temporal ($\geq 10~\mu$s) and spatial ($\sim$4~mm) resolution, and with a density coverage of $0-6\times 10^{19}\, m^{-3}$. During Type~I ELMs, a large radial expansion velocity ($\sim$600~ms) has been observed, and the radial particle transport time is comparable to the parallel transport time. The scaling of pedestal density loss, pedestal width, density scale length and the radial particle flux vs.\ $n_e$ and $q_{95}$ will be presented for both Type-I and III ELMs. ELM suppressed operation has been achieved by the use of resonant magnetic perturbations, where it has been observed that the radial extent of the SOL profile modification does not reach the vessel wall. A detailed study of density profile and fluctuation behavior during ELM–suppressed operation will be presented. [Preview Abstract] |
|
CP1.00117: Interpretation of Hydrocarbon (CD) Imaging and Chemical Sputtering in the Lower Divertor of DIII-D M. Groth, M.E. Fenstermacher, C.J. Lasnier, W.H. Meyer, G.D. Porter, N.H. Brooks, A.G. McLean The measured hydrocarbon emission in the lower DIII-D divertor is toroidally localized near the gaps of adjacent tiles, and shifted, with respect to the tile gaps in the inner divertor leg, in the direction of the toroidal magnetic field. The emission of the CD hydrocarbon radical was imaged at the (0,0) band head at 430.7 nm with tangentially viewing cameras, and spectroscopically resolved using a multi-chord spectrometer. These emission profiles indicate that the production of hydrocarbon by chemical sputtering occurs predominately near the gaps of these facetted tiles, correlating with elevated surface temperatures of, and larger heat and particle fluxes onto the exposed tile edges. Enhanced chemical sputtering in the tile gap region can also be due to formation of amorphous hydrocarbon layers (a:C-D) at the side walls of the tiles, and re-erosion by impinging neutrals. [Preview Abstract] |
|
CP1.00118: OEDGE Code Interpretation of Carbon Co-deposition Experiments in DIII-D L- and H-Mode Plasmas J.D. Elder, A.G. McLean, P.C. Stangeby, S.L. Allen, M.E. Fenstermacher, M. Groth, B.D. Bray, N.H. Brooks, W.P. West, D.L. Rudakov, J.A. Boedo, W.R. Wampler, J.G. Watkins, D.G. Whyte, G. Wang, W.M. Solomon A carbon-13 tracer experiment in DIII-D involved injection of C$_{13}$-methane into ITER-like partially detached ELMy H-mode plasmas, as a follow-up to an earlier L-mode experiment. Experimental details are given by McLean et al, this meeting. For the L-mode experiments, the OEDGE interpretive code is able to successfully replicate the C$_{13}$-deposition pattern, edge spectroscopic emissions and the increment to the C-ion density in the main plasma. A fast ($M\sim 0.5$) edge flow toward the inner divertor and no erosion of the C$_{13}$-deposits was assumed in L-mode. OEDGE will be applied to the data set for the H-mode experiments; ELMs and $\sim$10X greater input power make redistribution of the C$_{13}$-deposits by plasma erosion more likely. OEDGE modeling will address the time-varying plasma conditions and carbon re-erosion. [Preview Abstract] |
|
CP1.00119: Explicit Representation of Main-Chamber Recycling in the OEDGE Boundary Code S. Lisgo, P.C. Stangeby, A.G. McLean, J.D. Elder, M.E. Fenstermacher, M. Groth, B.D. Bray, N.H. Brooks, W.P. West, D.L. Rudakov, J.G. Watkins, D.G. Whyte Recent tokamak experiments indicate that significant plasma contact with the main-chamber walls can occur in some operating regimes, in addition to recycling at the divertor targets. In order to investigate the role of main-chamber recycling on core fuelling and impurity behavior, the OSM-EIRENE-DIVIMP (OEDGE) code has been modified to explicitly include magnetic field lines that terminate on $toroidally-symmetric$ main-chamber surfaces, which allows the model to include simultaneous plasma contact for both the divertor targets and the wall. This generalization of the computational mesh is made feasible by the 1D nature of the numerical methods employed in the OSM plasma solver. The analysis is restricted to discharges where the majority of the main-chamber plasma contact occurs on a toroidally-symmetric limiter surface (``window-frame" geometry). Results are presented for a low-density, \hbox {L-mode} DIII-D plasma. [Preview Abstract] |
|
CP1.00120: Migration of Micron Size Carbon Dust in the DIII-D Divertor C.P.C. Wong, N.H. Brooks, T.E. Evans, A.W. Hyatt, W.P. West, D.L. Rudakov, G. Antar, J.A. Boedo, E.M. Hollmann, S.I. Krasheninnikov, R.A. Moyer, A. Pigarov, M.E. Fenstermacher, M. Groth, C.J. Lasnier, R. Maingi, A.G. McLean, W.M. Solomon We studied migration of pre-characterized carbon dust in a tokamak environment by introducing about 30~mg of micron-size dust flakes (5-10~$\mu$m in diameter) in the lower divertor of \hbox{DIII-D} using the DiMES sample holder. In two separate experiments dust was exposed to strikepoint sweeps of high power lower-single-null discharges in an ELMing \hbox{H-mode} regime. When the outer strike point passed over the dust holder, 1-2\% of the total dust carbon content (equivalent to a few million of dust particles) penetrated the plasma core, raising the core carbon density by a factor of 2-3. Individual dust particles were observed moving at velocities of 10-100~m/s, predominantly in the toroidal direction consistent with the ion drag force. The observed behavior of the dust is in qualitative agreement with modeling by the DustT code. [Preview Abstract] |
|
CP1.00121: Porous Plug Injection System for Studies of Hydrocarbon Dissociation and Transport in DIII-D J.W. Davis, A.G. McLean, P.C. Stangeby, A.A. Haasz, S.L. Allen, R. Ellis, M.E. Fenstermacher, M. Groth, B.D. Bray, N.H. Brooks, W.P. West, C.P.C. Wong, D.G. Whyte, D.L. Rudakov, J.G. Watkins, S. Brezinsek Calibrated spectroscopic measurements of hydrocarbon dissociation fragments in a tokamak divertor were obtained by admitting methane through a porous graphite surface, such that fragments of the injected molecules returned to a carbon surface in a similar way to fragments due to natural chemical sputtering. The porous surface was made from a graphite plate with 1004 holes 0.25 mm in diameter spread over a 3 cm region, and was viewed by calibrated spectrometers. The gas flow rate was 7-40$\times$10$^{17}$ CH$_4$/s, simulating expected chemical erosion yields. Intensities of CD band and CI and CII line emissions were recorded. It was thereby established that chemical sputtering contributed a minority of the carbon atoms naturally sputtered from the outer target. [Preview Abstract] |
|
CP1.00122: Comparative Study of C I Line Profiles Resulting from Methane Puffing in the DIII-D Divertor, with those from Surface Sputtering N.H. Brooks, W.P. West, C.P.C. Wong, A.G. McLean, D.L. Rudakov Outstanding discrepancies between the expected temperature of carbon atoms generated by plasma breakup of hydrocarbons and that actually measured in DIII-D under conditions in which chemical sputtering dominates [1] are investigated by the controlled injection of methane through a porous plug [2]. Surprisingly, the methane injection yields an effective temperature $<$1 eV, rather than the several eV anticipated from modeling the breakup dynamics of the injected CH$_4$ molecule. At a spot at the same major radius as the porous plug but toroidally displaced, an asymmetric C I profile, characteristic of physical sputtering, is observed; this profile is significantly broader than that in the puff. The buildup of a soft C:D film on the face of the porous plug is explored as a possible explanation for the anomalously narrow C I profile on the spectrometer channel viewing the porous plug. [1] R.C. Isler et al., Phys. Plasmas $\bf {8}$, 4470 (2001). [2] A.G. McLean et al., this conference. [Preview Abstract] |
|
CP1.00123: Compatibility of the Radiating Divertor with Hybrid Plasmas in DIII-D T.W. Petrie, N.H. Brooks, A.W. Hyatt, A.W. Leonard, M.J. Schaffer, M.R. Wade, W.P. West, M.E. Fenstermacher, M. Groth, C.J. Lasnier, J.G. Watkins Seeding the divertor with impurities to enhance the power radiated upstream of the divertor target was successfully carried out in the DIII-D ``hybrid" H-mode regime. Argon was puffed directly into the divertor, while deuterium gas $D_2$ was injected into the scrape-off layer upstream of the divertor. When the argon gas injection rate $\Gamma_{AR}$ was raised to a level which caused the ratio of total radiated power to power input to increase from 0.46 to 0.62, we found equal increases in the radiated power from the divertor and the core, reductions in heat flux at the divertor target of almost a factor of two, and energy confinement unchanged. For a fixed $D_2$ injection rate, both the radiated power and argon density in the core plasma were proportional to $\Gamma_{AR}$, suggesting that feedback control over the radiated power using $\Gamma_{AR}$ may be possible. Divertor enrichment $>$30 were determined for these plasmas. [Preview Abstract] |
|
CP1.00124: Thermal Ion Loss From Confined QH-mode Plasma in the Presence of Alfven Eigenmodes C.J. Lasnier, T.L. Rhodes, G. Wang, L. Zeng, K.H. Burrell, J.S. deGrassie, M.A. VanZeeland, G.R. McKee, J.C. Rost, J.G. Watkins During QH-mode discharges in \hbox{DIII-D}, in the upper single-null configuration, we use infrared cameras to observe heating of the upper outer baffle far outside the strike point. We attribute this heating to impact by moderate energy ($\sim$5~keV) ions lost from the core plasma, with supporting data from charge exchange recombination (CER) measurements and fixed Langmuir probes. Examining millimeter-wave scattering data, we find correlation of the baffle heating with the presence of core Alfven eigenmodes. In this presentation we show this correlation and explore the characteristics of Alfven modes that may be contributing to warm ion losses. We examine beam emission spectroscopy (BES), multi-chord fast interferometry, fast magnetic pickups, reflectometry, and phase contrast imaging data for additional information. [Preview Abstract] |
|
CP1.00125: The Effect of Plasma Shaping on Edge Profiles and Stability During QH-Mode Operation on DIII-D W.P. West, A.W. Leonard, B.D. Bray, K.H. Burrell, P. Gohil, T.H. Osborne, P.B. Snyder, D.M. Thomas, T.A. Casper, E.J. Doyle, G. Wang, L. Zeng Edge profiles of electron density and temperature, ion temperature, C$^{+6}$ density, and magnetic field pitch angle were measured in ELM-free QH mode plasmas as the triangularity and squareness of the separatrix was varied. Preliminary analysis indicates that the pressure at the top of the pedestal increases with stronger shaping. The profiles are used along with magnetic and motional Stark effect data in a reconstruction of the plasma equilibrium that includes the edge bootstrap current. Generally good agreement between the reconstructed equilibria and the Li beam polarimetry measurements of the pitch angle in the region of the pedestal is found. Stability of these equilibria against coupled peeling/ballooning modes will be presented. [Preview Abstract] |
|
CP1.00126: RWM Stabilization in DIII-D Using I-Coils With High Speed Actuators G.L. Jackson, A.G. Kellman, R.J. La Haye, J.t. Scoville, E.J. Strait, J.M. Bialek, A.M. Garofalo, O. Katsuro-Hopkins, G.A. Navratil, H. Reimerdes, Y. In, A. Nagy, M. Okabayashi, H. Takahashi A new prototype actuator system driving 12 internal coils (I-coils) was used to help stabilize resistive wall modes (RWMs) up to $\beta_N\sim 4$. This approach is an alternative to rotational stabilization, which may not be adequate for fusion devices. VALEN modeling shows that as $\beta_N$ approaches the ideal wall limit, higher bandwidth and lower system delay time are required to stabilize the larger RWM growth rates. This actuator system consists of 6 transistor amplifiers (dc-40 kHz), configured in 3 pairs, each driving 4 I-coils in an n=1 configuration. Initial experiments include the combination of I-coils for fast RWM stabilization and external C-coils with higher current capability for slower response dynamic error field correction. Effects of noise, maximum actuator current, and feedback system delay time on maximum achievable $\beta_N$ will also be presented. [Preview Abstract] |
|
CP1.00127: Resistive Wall Mode Internal Structure Identification by Soft X-Ray Contrast Enhancing Technique I.N. Bogatu, Y. In The understanding of the internal structure of the slowly growing resistive wall mode (RWM) is important for improved identification and control. The real-time RWM stabilization is a critical topic for advanced tokamak regimes, for example, in ITER. During an early RWM evolving stage it is difficult to distinguish the RWM-associated perturbations from equilibrium changes. But a newly-developed soft x-ray contrast enhancing technique (SXR CET) was shown to be sensitive to the evolving RWM. Based on time or spatial derivative of signals, the SXR CET is a useful diagnostic to reveal RWM internal structures in conjunction with magnetic sensors. Defining the requirements for an ideal SXR CET, we present the analyzed results of SXR CET using \hbox{DIII-D} experimental data. The SXR CET correlations with magnetic field perturbation, electron temperature, and current density profile are presented. We also evaluate the possibilities of SXR CET application to discriminate rotating and non-rotating plasmas for understanding of the slowly growing RWM. [Preview Abstract] |
|
CP1.00128: Model-based RWM Identification, ELM-Filtering, and RWM Control on the DIII-D Tokamak Y. In, J.S. Kim, M.S. Chu, J.R. Ferron, D.A. Humphreys, G.L. Jackson, R.J. La Haye, R.D. Johnson, J.T. Scoville, E.J. Strait, A.D. Turnbull, M.L. Walker, A.M. Garofalo, H. Reimerdes, M.S. Chance, M. Okabayashi Resistive-wall-mode (RWM) modeling helps not only to identify the RWM mode, but also to optimize model-based control. Based on modeled sensor signals, a static matched filter is constructed to identify the RWM mode. Using a ``picture frame" wall model, a dynamic Kalman filter has been developed to discriminate edge localized modes (ELMs) from RWMs. From this ELM-filtered RWM mode identification, an enhanced RWM feedback controller has been implemented for \hbox{DIII-D} plasmas. Recent experiments showed the effectiveness of the Kalman filter scheme; the feedback coils were rarely excited during ELMs, while responding to RWMs. To investigate proper RWM responses, an optimized Kalman filter parameter set has been found and evaluated. Meanwhile, using a thin wall treatment, an eigenmode approach has been adopted to represent the vessel more accurately. [Preview Abstract] |
|
CP1.00129: Interaction of Edge-Localized Modes and Resistive Wall Modes in DIII-D E.J. Strait, M.S. Chu, G.L. \hbox{Jackson}, R.J. La Haye, J.T. Scoville, P.B. Snyder, A.M. Garofalo, H. Reimerdes, Y. In, M. Okabayashi Edge-Localized Modes (ELMs) in the DIII-D H-mode edge pedestal include a magnetic perturbation with a broad spectrum of toroidal mode numbers, which is observed to couple to the Resistive Wall Mode (RWM) branch of the n=1 ideal kink mode at high beta. The damping rate of the driven RWM provides a passive probe of the RWM's stability, decreasing as beta increases. In some cases the ELM appears to trigger the growth of an unstable RWM, perhaps by altering the plasma rotation when conditions are near the instability threshold. The amplitude of the ELM-excited RWMs may thus be a factor in determining the minimum current requirements for RWM feedback stabilization. Experimental results will be presented. [Preview Abstract] |
|
CP1.00130: ELM Trigger Model: Sheath Breakdown Caused by Thermoelectrically Driven SOL Current in ELM Precursor Phase H. Takahashi, E.D. Fredrickson, M.J. Schaffer, N.H. Brooks, T.E. Evans, G.L. Jackson, L.L. Lao, M.E. Austin, J.G. Watkins The breakdown of the sheath at the interface of the SOL plasma and divertor tiles, through which thermoelectrically driven scrape-off-layer current (SOLC) flowing along open field lines enters the tokamak structure, is investigated [1] in \hbox{DIII-D} as a possibly causal element in the ELM triggering process. Sheath breakdown may occur when the ion saturation current density ($j_{sat}$) at the sheath falls to the SOLC current density, increasing the potential drop across the sheath. The limiting $j_{sat}$ varies cyclically through the ELM cycle, reaching a minimum, as the SOL density is pumped out, just before the ELM crash. The increased SOL current following breakdown may contribute toward destabilizing MHD modes. Evidence for sheath breakdown is sought from measurement by the SOLC and Langmuir probe diagnostics. [Preview Abstract] |
|
CP1.00131: Application of a 2D Ideal MHD Linear Stability Code to Finite-Amplitude Nonaxisymmetric Perturbations A.D. Turnbull, M.S. Chu, L.L. Lao Nonaxisymmetric perturbations from unintended error fields, fields imposed from external coils, or from saturated instabilities are becoming increasingly important in tokamaks. Depending on the plasma response, the fields can saturate in a slightly perturbed state or resonate with internal rational surfaces, opening large islands. It is important to study the plasma response of 2D configurations against finite perturbations to optimize and maintain their robustness. An ideal MHD stability code can study this response in several ways. The formulation [1] provides a complete description of the response from external fields in principle. An intuitive approach to analyze final saturated states is to impose ideal modes as finite small-amplitude boundary displacements for a new 3D equilibrium. Both approaches will be discussed, as well as the relationship between them and the normal mode approach used to study resonant field amplification of resistive wall modes [2].\par \vspace{0.5em} \noindent [1]~C.~Nuhrenberg and A.H.\ Boozer, Phys.\ Plasmas ${\bf 10}$, 2840 (2003).\par \noindent [2]~M.S.\ Chu, et al., Nucl.\ Fusion ${\bf 43}$, 441 (2003). [Preview Abstract] |
|
CP1.00132: Bifurcated Equilibria and Magnetic Islands in Tokamaks Paul R. Garabedian The NSTAB code captures islands successfully despite a nested surface hypothesis made in the coordinate system that is employed. The resolution of the code can be checked by applying it to the vacuum field of stellarators where islands are known to exist in equilibria found by other methods. The same numerical construction produces slender islands in tokamaks like the DIII-D and ITER. When those three-dimensional solutions of the tokamak problem are used in Monte Carlo computations of the energy confinement time, no anomalous transport occurs in the results. The physical significance of finding many three-dimensional MHD equilibria in axially symmetric tokamaks needs further investigation. More specifically, one can ask how much their effect might contribute to the prompt loss of $\alpha $ particles or to disruptions. [Preview Abstract] |
|
CP1.00133: Massive Gas Injection System for Disruption Mitigation on the DIII-D Tokamak T.C. Jernigan, L.A. Baylor, S.K. Combs, D.A. Humphreys, P.B. Parks, J.C. Wesley Massive injection of deuterium or noble gases ($>$10$^{22}$ molecules) has been very effective at mitigating disruptions in \hbox{DIII-D} [1]. Both the divertor heat loads and the first wall forces were reduced by more than a factor of four. Total electron densities (free and bound) of $\sim$10$^{21}$ m$^{-3}$ have been achieved, close to that required to prevent avalanche multiplication of runaways. Two tested configurations are described. Both use a fast solenoid valve with an orifice diameter of 4~mm with a flow rate in helium of 5$\times$10$^4$~Pa~m$^3$/s at 7~MPa. A new valve with a 20~mm orifice will be tested on \hbox{DIII-D} in 2006. This valve is actually close to that required for ITER. Calculations show that a set of four such valves can reach the no-avalanche density in ITER in $\sim$0.25~$t_{co}$ where $t_{co}$ is the plasma current quench time. How the gas jet interacts and mixes with the plasma is under investigation.\par \vspace{0.5em} \noindent [1]~D.G.\ Whyte, et al., Phys.\ Rev.\ Lett.\ ${\bf 89}$, 55001 (2001). [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