Theory Department

2012 Theory Department Weekly Highlights

Nov. 30~

Nov. 23~

Nov. 16~

Nov. 9~

THEORY
Many researchers from the theory department attended the Annual Meeting of the APS Division of Plasma Physics in Providence, Rhode Island October 29-November 2. Three theory department members gave invited talks; Ilya Dodin " Quantum Physics of Classical Waves in
Plasma", Guoyong Fu "M3D-K Simulations of Beam-Driven Alfven Modes in DIIID" and Igor Kaganovich " Plasma-Wall Interaction in Presence of Intense Electron Emission from Walls. A recent Princeton University graduate, theory department student Dr. Paul Schmit presented an
invited talk titled "New Wave Effects in NonStationary Plasma".

COMPUTATIONAL PLASMA PHYSICS GROUP
Professor Mo and his group visited PPPL and gave a seminar on JASMIN (J Adaptive Structured Meshes applications INfrastructure) on November 8. JASMIN is a parallel software platform using adaptive multi-level structured grids. It has been developing in the Laboratory of Computational Physics (Beijing, China) led by Professor Zeyao Mo. The main objective to accelerate the large scale simulations of complex on parallel computers. Written in C++ and Fortran mixed-language model and built on public domain standards of MPI, OpenMP, C++,
HDF5, it is layer and component designed to facilitate and ease the learning curve of user application code development. Tens of parallel programs have been reconstructed or developed to achieve high performance on thousands of processors with major applications in ICF and
CFD. JASMIN promotes a new paradigm of parallel programming in the Chinese scientific community.

Nov. 2~

COMPUTATIONAL PLASMA PHYSICS GROUP
A TRANSP Users Group meeting was held in Providence on October 31 in conjunction with the annual APS-DPP meeting. Material was prepared by each of the TRANSP developers to inform users of new and planned capabilities. Highlights of these presentations were: Robert Andre: The free boundary equilibrium package ISOLVER has now been released and can be used with TRANSP in either interpretive or predictive mode. In the latter, the equilibrium is evolved self consistently coupled with the flux diffusion equation. It works both in a "least squares mode" where boundary shape points are specified, and in a "coil currents mode" where the plasma shape is determined self-consistently from plasma, conductor, and coil currents. Marina Gorelenkova: Several new capabilities have been added to NUBEAM. It is now coupled
to TORIC with the addition of a RF kick operator. A new 3D Halo model has been added to more accurately track the halo neutrals generated by charge exchange. A new energy and pitch angle dependent anomalous diffusion of the beam ions as computed by the TGLF/DEP codes has been implemented.

Xingqiu Yuan: The new solver package PT-SOLVER is now operational. It was demonstrated that it is capable of solving temperature evolution equations when TGLF is used as the transport model. The capability is being released first to beta-test users, but is expected to be generally released in the next six months. Users wishing to learn more about any of these capabilities should contact the appropriate developer.

Oct. 26~

THEORY GROUP
H. Mynick attended the IAEA Fusion Energy Conference in San Diego, CA and presented a paper, "Turbulent Optimization in Stellarators & Tokamaks via Shaping". A. Hakim contributed to several sections of the report "Multiphysics Simulations: Challenges and Opportunities", which is the outcome of a multidisplinary workshop organized by David Keyes, Lois McInnes, and Carol Woodward. The report, to appear in the International Journal of High Performance Computing, is available at http://www.ipd.anl.gov/anlpubs/2012/01/72183.pdf. The space physics group at PPPL hosted a NASA LWS team meeting on "Integrating Kinetic Effects in Global Models" at PPPL on October 22-23. The group discussed collaborative projects considering comparative studies of reconnection in global systems using MHD, Hall-MHD, and fully kinetic models. The team also considered how to model wave-particle interactions that control temperature anisotropies in the solar wind and planetary magnetospheres.

COMPUTATIONAL PLASMA PHYSICS GROUP
At the request of TRANSP user David Pace and Ron Waltz of General Atomics, X. Yuan and M. Gorelenkov have implemented a new model for fast ion diffusion due to background turbulence into TRANSP. The model makes use of the micro instabilities calculated by TLGF on each flux surface. An auxillary program DEP then computes the expected diffusion coefficient for fast particles on this surface as a function of particle velocity and pitch angle. This information in transmitted to the neutral beam package NUBEAM where a diffusion operator is applied to each particle depending on its location and velocity coordinates. This capability will be used in David Pace's upcoming APS invited talk to assess the role of background turbulence on the beamdriven current profiles in DIII-D.

Oct. 19~

Theory Group
Members of the Theory Department presented multiple posters at the IAEA Fusion Energy Conference in San Diego, California. These are: E. Belova, "Numerical Simulations of NBIdriven GAE modes in L-mode and H-mode Discharges in NSTX", A. Hakim, "Integrated Fusion Simulations of Core-Edge-Wall Thermal and Particle Transport Using the FACETS Code", S. Jardin, "Simulation of Sawteeth and other Global Macroscopic Dynamics of Tokamak Plasmas on the Transport Timescale". Jardin was also co-author of five other papers presented at the
meeting. S.K. Ku, "Flux-driven Full-fGyrokinetic Study of the Nonlocal Edge-Core Ti interaction", W. W. Lee, "Steady-State Particle-In-Cell Simulations of Microturbulence in Tokamaks, "Turbulent Optimization in Stellarators & Tokamaks via Shaping", W. Wang, "Turbulence Generated Non-inductive Current and Shear Flow Driven Turbulent Transport in Tokamaks", L. Zakharov "Current Sharing Between Plasma and Walls in Tokamak Disruptions".

In addition, members of the Theory Department attended ITPA Topical Group on MHD, Disruptions and Control meeting, which was also in San Diego, California. S. Jardin gave a talk titled "Working Group 10 Report: Halo Current Modeling with TSC and DINA", and G. Fu gave a talk on "M3D-K Simulations of Beam-Driven Alfven Modes in DIIID".

COMPUTATIONAL PLASMA PHYSICS GROUP
S. Ethier presented a CPPG Seminar on "What is GPU Computing". The talk addressed "GPU Computing", which aims to run scientific codes on Graphics Processing Units designed for fast rendering of complex 3D graphics on high resolution displays. While there are claims of 100X speedups in the time-to-solution over their CPU counterparts, Dr. Ethier explained under what conditions one might expect substantial speedup and when one might not. He also addressed what is involved in converting a modern HPC code that is optimized for CPU computing to GPUs.

Oct. 12~

Theory Group
Peter Damiano presented a seminar to the Space Physics Research Group at the Space Sciences Laboratory (UC Berkeley)  on Oct. 9 entitled "Electron energization and wave dispersion in a mirror kinetic Alfven wave". While there, he was involved in research discussions with Chris Chaston on Alfven wave induced particle energization in magnetospheric plasmas including the analysis of field and particle observations from the FAST (Fast Auroral SnapshoT) NASA satellite mission. 

Oct. 5~

THEORY
The Theory Department organized a Retreat, September 24-25 and 27 on Princeton University's main campus. The purpose of the Retreat was to develop a strategic plan for the Department spanning four areas: (1) Transport and Confinement, (2) Global MHD, including Energetic Particles, (3) Basic Plasma Science, and (4) High-Performance Computing Support. The Retreat was organized around plenary talks and breakout sessions. The plenary talks were mostly given by PPPL experimental leaders and Dr. Ronald Waltz from General Atomics, who gave a talk on"Core and Edge Transport and Confinement: Past, Present, and Future". The breakout sessions were moderated by experimentalists, and led by members of the Theory Department. This forum enabled lively and stimulating discussions, and has led to the development of key elements of a strategic plan.

COMPUTATIONAL PLASMA PHYSICS GROU
S. Jardin and S. Lazerson (Advanced Projects) participated in a teleconference with a ITPA working group that is tasked with making an assessment of the likely edge displacement in ITER due to (a) long-lived core MHD instabilities and (b) application of RMPs. Jardin presented 3D nonlinear results from M3D-C1 on edge displacements from sawteeth and helical deformations in both NSTX and ITER. Lazerson presented results using VMEC to calculate the effect of n=3 RMPs. Action items include performing more detailed cross-code comparisons including the comparison of M3D-C1 results with those obtained by Tony Cooper with ANIMEC, and the comparison of NSTX results with data.

Sep. 28~

COMPUTATIONAL PLASMA PHYSICS GROUP
Jin Chen visited Institute of Computational Mathematics and Scientific/Engineering Computing, CAS from September 8-24, and gave a series of lectures to graduate students on numerical methods with applications in solving MHD related numerical problems. Jin met two groups of researchers and software developers and had very fruitful discussions with them. One group is working on Adaptive Structured Meshes applications INfrastructure(JASMIN) with applications in ICF; the other group is working on Parallel Hierarchical Grid (PHG), a parallel computing environments and platforms for design and implementation of 3D self-adaptive finite elements on distributive hierarchical unstructured grids. The useful inputs may directly benefit our present work in MHD simulations.

Sep. 21~

COMPUTATIONAL PLASMA PHYSICS GROUP

S. Jardin and S. Kaye led a delegation to DOE on Friday to give a briefing on the development plan for the TRANSP/PTRANSP code. The others presenting were P. Bonolli (MIT) and A. Kritz (Lehigh). They met with John Mandrekas, Steve Eckstrand, Jim VanDam, and Ed Synakowski. The two hours of presentations covered planned physics development in the areas of free-boundary equilibrium, non-linear solver development, anomalous diffusion of fast ions, and RF/fast-ion interactions. Productive discussions were held regarding funding this activity in
the future.

Sep. 14~

THEORY

COMPUTATIONAL PLASMA PHYSICS GROUP
The TGLF transport model [Kinsey, et al, Nuclear Fusion 51 083001 (2011)] has been implemented in the new PTRANSP solver, PT-solver in a way that it is parallel over both wave number (interior to TGLF) and exterior (over flux surfaces). This has now been fully integrated into the TRANSP code and runs have been completed for both DIII-D (0.5 - 6.5 sec) and ITER discharges in which this parallel capability was combined with RF calculations using parallel TORIC and neutral beam calculations using parallel NUBEAM. These and other results will be presented at the upcoming IAEA meeting in paper [ ITR/P1-29 by R. Budny, X. Yuan, et. al. "PTRANSP tests of TGLF and predictions for ITER"].

Sep. 7 ~

THEORY
Version 4.5 of the DEGAS2 Monte Carlo neutral transport code (htttp://w3.pppl.gov/degas2) was released on August 24 by Daren Stotler. DEGAS2 can now be run in a time-dependent mode and arbitrary volumetric sources of neutral particles can be specified. Time dependence was also incorporated into the DEGAS2 interface to the XGC0 guiding center neoclassical particle transport code being developed within the Partnership for Edge Physics Simulation (EPSI; PI: C. S. Chang, PPPL). The corresponding modifications to XGC0 were released to EPSI team members on August 15. The principal new capability is a fully time dependent coupling between XGC0 and DEGAS2, with the spatial distribution of the neutral atoms simulated by DEGAS2 being determined self-consistently between the two codes.

Aug. 24~

THEORY
Dr. Ihor Holod of University of California at Irvine is currently visiting PPPL for the
implementation of the double split weight (DSW) scheme into Gyrokinetic Toroidal Code
(GTC). GTC is a global nonlinear Particle-In-Cell (PIC) code for simulating turbulence in fusion plasmas, which has recently been extended to include the electromagnetic capabilities based on the fluid-kinetic hybrid(FKH) electron model by using the expansion of the electron response into adiabatic and nonadiabatic part, and by treating electrons in the lowest order as a massless fluid. This allows the relaxation of the numerical restrictions associated with the electron Courant condition and removes unnecessary high-frequency modes. However, the FKH model excludes the microtearing physics, for which the electron response expansion breaks down near the mode rational surface. To improve the electromagnetic capability of GTC, the UCI team headed by Prof. Zhihong Lin is now working with E. Startsev and W.W. Lee of PPPL to implement the DSW scheme into GTC. The DSW scheme retains electron inertia near the mode rational surface by treating the electron response non-pertubatively. The most important step would be the implementation of the Ampere’s law to determine the vector potential, which is currently calculated from the Faraday’s law in GTC. In the next step, the additional Poisson-like equation will be introduced to calculate the effective potential. Simple test case will be designed to test the new capabilities. If successful, the scheme can also be incorporated into GTS and XGC codes at PPPL in the future.

Aug 17~

THEORY
Mr. Feng Wang, a graduate student from Dalian University of Technology, Dalian, China, returned to China after an extended visit to PPPL from January 2011 to July 2012. During his visit, he worked with his host, Dr. Guoyong Fu, and Josh Breslau on nonlinear simulation of non-resonant kink modes in NSTX using the M3D code. He found that plasma rotation can greatly reduce the m=2/n=1 island width. The fast beam ion distribution is flattened by the saturated m=1/n=1 mode. Feng Wang will continue this collaborative research in China as a part of his Ph.D. thesis work.

Thomas Dickerson of St. Michael's College, a NUF student, worked with Wei-li Lee and Ed Startsev of the Theory Department on some numerical problems associated with shear-Alfvén waves in tokamaks. Specifically he investigated: 1) the time step in the simulation as a function of plasma beta, and 2) the formation of shear-Alfvén eigenmodes in the presence of plasma inhomogeneities, also as a function of plasma beta. Results will be presented at the upcoming APS meeting.

COMPUTATIONAL PLASMA PHYSICS GROUP
This summer, the CPPG mentored two National Undergraduate Fellowship (NUF) program interns, two Science Undergraduate Laboratory Interns (SULI), and two high school science interns. NUF student David Perkins (BYU) worked with S. Ethier and W. Wang on improving the performance and scalability of the GTC-NEO particle code, while Andrew Ritchie worked with S. Jardin on benchmarking M3D-C1 linear resistive results with an asymptotic matching code. SULI student Jeffrey Lestz (Washington University) and high school student Sadik Shahidain (Princeton High) also worked with Stephane Ethier and Weixing Wang on the development of a MATLAB-based analysis tool to explore the microturbulence data generated by the GTS simulation code. SULI student Matthew McMillan (Wheaton College) worked with Sam Lazerson and Eliot Feibush on modeling, benchmarking, and visualizing neutral beams in 3-D MHD equilibria. Finally, high school student Hadar Lazar (Bergen County Academy) was supervised by Eliot Feibush worked with the MRX team investigating the scaling and merging of magnetic islands as guide fields are applied.

Aug 10~

THEORY:
Five researchers from PPPL will attend the International Symposium on Heavy Ion Inertial Fusion in San Francisco. They will present the following invited talks: R. Davidson, National Research Council Assessment of the Prospects for Inertial Fusion, L. Grisham, Studies of Electrical Breakdown Processes across Vacuum Gaps,I. Kaganovich, Drift compression and Final Focus (review), E. Startsev, Effects of Beam-Plasma Instabilities on Neutralized Propagation of Intense Ion Beams in Background Plasma, H, Qin, on Design considerations for a wobbler in a HIF driver, and E. Gilson on Plasma Sources for Drivers and NDCX-II.

COMPUTATIONAL PLASMA PHYSICS GROUP (S. JARDIN):
Francesca Poli has implemented the CDBM transport model [ Fukuyama et al ,Plasma Phys. Controlled Fusion, (1995) 37, 611] in the Tokamak Simulation Code (TSC). In this model, the transport coefficients are based on the theory of self-sustained turbulence due to current diffusivity driven modes. The CDBM model includes a correction to the thermal diffusivity due to weak or negative magnetic shear and large Shafranov shift, as well as the ExB shear turbulence stabilization mechanism. Having the trigger for an ITB, this model can predict ITB formation in reversed shear plasmas.

The paper "Numerical Calculation of Neoclassical Distribution Functions and Current Profiles in Low Collisionality, Axisymmtric Plasmas" (PPPL-4775) by B. C. Lyons (PPPL student), S. C. Jardin, and J. J. Ramos (MIT) has been accepted for publication in Physics of Plasmas. The paper describes a numerical method for computing the non-Maxwellian part of the distribution function and thereby closing the MHD equations. It is demonstrated that the method obtains the correct result for the neoclassical resistivity and bootstrap current when applied to an axisymmetric (tokamak) plasma. This method forms the basis for extending the 3D 2F MHD code M3D-C1 to include neoclassical kinetic effects.

Aug 3~

In the Earth’s magnetosphere, low frequency (mHz) global scale Alfven wave modes (also known as Field Line Resonances) are observed to stand on the Earth’s closed magnetic field lines and have been linked to the formation of some auroral arcs. The arcs result when electrons, accelerated in the waves, precipitate into the atmosphere. However, the exact means by which electrons are accelerated in these waves is not completely understood. Alfven waves are known to carry significant parallel electric fields when the perpendicular scale length is on the order of the electron inertial length (and b << me/mi – inertial Alfven wave) or the ion acoustic length (and b >> me/mi – kinetic Alfven wave), but neither of these limits can produce the kilovolt parallel potential drops necessary to accelerate electrons to the energies needed to explain auroral observations for these global modes. Using a self-consistent global 2D hybrid MHD-kinetic electron model in dipolar coordinates (Damiano et al., Phys. Plasmas, 14, 062904, 2007) we have studied the effects of the magnetic mirror force on the evolution of the wave (a limit that can be termed “the mirror kinetic Alfven wave” - Nakamura, J. Geophys. Res., 105, 10729, 2000). In inertial and kinetic Alfven waves, electron mass and electron pressure effects respectively provide the impedance for electron motion along the field line necessitating a parallel electric field to accelerate electrons in order to carry the parallel current and maintain quasi-neutrality. In the mirror kinetic Alfven wave, that impedance is due to the mirror force that results from the converging topology of the Earth’s closed magnetic field lines. Using the code, we illustrated, for the first time in simulation, that kilovolt order parallel potential drops could self-consistently develop in these waves for realistic magnetospheric parameters (Damiano and Johnson, Geophys. Res. Lett., 39, L02102, 2012). The resulting electron acceleration is a significant sink of wave energy and we noted a dispersion of wave energy perpendicular to the magnetic field relative to ideal MHD simulations for consistent parameters. This dispersion is due to the perpendicular Poynting flux associated with the parallel electric field generated by mirror force effects and dominates the dispersion due to electron inertial effects [Damiano and Johnson, PRL submitted, 2012]. This work is the first time that it has been noted that mirror force effects can affect the dispersive characteristics of the wave and alters the traditional picture of energy confinement in these global modes. The study also suggested that the dispersive and dissipation characteristics associated with the mirror force effects on the electron dynamics could be modeled effectively in the fluid limit with the introduction of a transport coefficient in the
MHD description.

Professor Alain Brizard from Saint Michael's College, Vermont, visited PPPL for two days and gave a theory seminar on " Beyond Linear Polarization in Gyrokinetic Theory", discussing a guiding-center contribution to polarization that is often dropped, and nonlinear (quadratic in the potential) gyrocenter polarization. He also discussed the relation between gyrokinetic polarization and high-frequency oscillation-center polarization (used for laser-plasma interactions or other high-frequency phenomena). During his visit he had extensive discussions with several theorists and graduate students about subtle aspects of higher order gyrokinetics. An abstract for his talk and the slides are available at: https://w3.pppl.gov/theory/Seminars2011.html

A paper on "Kink modes and surface currents associated with vertical displacement events", by Janardhan Manickam, Allen Boozer, and Stefan Gerhardt, Princeton Plasma Physics Laboratory, Princeton University, Princeton, New Jersey, USA was published in the August issue of Physics of Plasmas. The abstract is as follows. The fast termination phase of a vertical displacement event (VDE) in a tokamak is modeled as a sequence of shrinking equilibria, where the core current profile remains constant so that the safety-factor at the axis, qaxis, remains fixed and the qedge systematically decreases. At some point, the n=1 kink mode is destabilized. Kink modes distort the magnetic field lines outside the plasma, and surface currents are required to nullify the normal component of the B-field at the plasma boundary and maintain equilibrium at finite pressure. If the plasma touches a conductor, the current can be transferred to the conductor, and may be measurable by the halo current monitors. This report describes a practical method to model the plasma as it evolves during a VDE, and determine the surface currents, needed to maintain equilibrium. The main results are that the onset conditions for the disruption are that the growth-rate of the n¼1 kink exceeds half the Alfven time and the associated surface current needed to maintain equilibrium exceeds one half of the core plasma current. This occurs when qedge drops below a low integer, usually 2. Application to NSTX provides favorable comparison with non-axisymmetric halo-current measurements. The model is also applied to ITER and shows that the 2/1 mode is projected to be the most likely cause of the final disruption.

COMPUTATIONAL PLASMA PHYSICS GROUP (S. JARDIN):
Xingqiu Yuan gave a presentation, which was broadcast via ReadyTalk, and led a discussion on "Status and Plans for Density Prediction in PT-SOLVER. Density prediction is a new capability for PTRANSP/TRANSP made possible by the introduction of the new package for solving the surface-averaged transport equations, PT-SOLVER. Presently, the solver works with the anomolous transport model TGLF and the neoclassical model NEO, but other tranport models are now being added. This will be released into the production version of TRANSP within the next few weeks.

Eliot Feibush and Sam Lazerson represented PPPL at the annual DOE Computational Science Graduate Fellowship conference in Arlington, Virginia on July 27. They presented practicum research opportunities to the students who have been awarded fellowships.

July 27~

Theory Group

Professor Hameiri from NYU concluded his visit to PPPL on July 20. During his six-week visit Professor Hameiri worked closely with Dr. Guazzotto from University of Rochester on the problem of ideal-MHD equilibrium with flow in the transonic regime, with particular reference to the long-standing issue of the hyperbolic region appearing in the otherwise elliptic equilibrium problem for poloidal velocities close to the poloidal sound speed (the so-called "first hyperbolic region"). This problem is of interest because observed poloidal flow in of poloidal/total magnetic field (the poloidal sound speed), in particular near the plasma edge. At this range, the governing Grad-Shafranov (GS) equation changes its type twice. From elliptic at the low velocity region to hyperbolic in the region where the velocity is near the poloidal sound speed, then again to elliptic when the velocity is higher still near the plasma edge. The existing code FLOW, used to calculate MHD equilibrium with mass flow [L. Guazzotto, R. Betti, J. Manickam, and S. Kaye, Phys. Plasmas 11, 604 (2004)] has the problem that the computation solves the GS equation as if it was elliptic in the whole computation domain, namely, as a boundary value problem. This is generally inappropriate for a transonic case with a hyperbolic region. Moreover, a hyperbolic region never shows up in the calculation, presumably because it is extremely narrow. The question that arises is whether or not the existing code suffers from a fatal error that makes its results invalid for transonic flows. Convincing physical arguments have been used in the past to justify the numerical approach and the computational results, but a mathematical proof of the validity of the transonic results in the transonic hyperbolic region was until now missing. The result of the present work by Guazzotto and Hameiri conducted for a month and a half this summer at PPPL (and for a month last summer) is that the existing code is actually proper. This conclusion was reached by constructing and investigating two models, a one-dimensional nonlinear and a two-dimensional (axisymmetric) linear. The nonlinear model has an equation which is practically identical to the GS equation, with a minor simplification based on a large-aspect-ratio expansion. The remarkable result that comes out is that the two points of sonic (type) transitions can actually be required to be points of regularity of the solution. That is, such a requirement produces a continuous equilibrium state throughout the plasma. This result is entirely nontrivial, and would not have been obtained without constructing yet another nonlinear equation, based on minimum energy subject to dynamical constraints [E. Hameiri, Phys. Plasmas 5, 3270 (1998)] that can be solved exactly. The linear two-dimensional model is an equation that is not very close in form to the GS equation, but does contain two transition points as needed. It also has the advantage that after Fourier transforming the poloidal angle variable, the equation has an exact solution in terms of special functions. It can then be seen directly that requiring the transition points not to be singular, i.e., connecting the solution across such points under the requirement that it remains continuous and nonsingular, yields an acceptable unique solution. Remarkably, a numerical solution based on a scheme analogous to the one used in FLOW converges to the correct solution, even if the size of the hyperbolic region is finite. The conclusion from both this linear and the previous nonlinear models is that the transition of the plasma flow from sub to super “sonic” (with the Mach number becoming greater than unity) does not suggest any additional modification of the code as is, for the case where the hyperbolic region is so narrow as to remain unseen. For a larger and visible hyperbolic region, one would have needed to impose a continuity requirement (which is not necessary in the linear model problem). Fortunately, such a situation will not happen for any plasma beta appreciably less than unity at the location of the transonic transition.

A paper by Alexander Khrabrov and Igor Kaganovich titled "Electron Scattering in Helium for Monte Carlo Simulations” has been accepted for publication in Physics of Plasmas. The work developed an approximation for differential cross-section of electron scattering on helium atoms. It is intended for use in Monte Carlo simulations, where reproducing essential physics needs to be combined with numerical efficiency. The model is based on the screened-Coulomb differential cross-section with energy-dependent screening. For helium, a two-pole approximation of the screening factor is found to be highly accurate over a wide range of energies.

Matt Landreman recently visited PPPL from MIT for a few days to discuss possible areas of collaboration in gyrokinetic studies of the edge region of tokamaks and stellarators. He also gave a theory seminar on "Omnigenity as Generalized Quasisymmetry in Stellarators" (the slides are available at https://w3.pppl.gov/theory/Seminars2011.html). Here is a summary of his talk:“A stellarator reactor will need to be nearly omnigenous, meaning the radial guiding-center drift velocity averages to zero over time for all particles. Omnigenity is easier to achieve than quasisymmetry, and omnigenous plasmas may be highly non-quasisymmetric, yet several properties of the flows and current in quasisymmetric plasmas apply directly or with only minor modification to the larger class of omnigenous plasmas. All three types of quasisymmetry (helical, poloidal, and axisymmetry) have an omnigenous generalization, and eneralized poloidal symmetry is not prohibited near the magnetic axis as true poloidal symmetry is. The radial electric field in quasisymmetric plasmas is determined by turbulence, while the electric field in omnigenous plasmas is determined by neoclassical physics.

July 20~

Theory Group
Leonid E. Zakharov and Luca Guazzotto have implemented the psi-form of the MHD energy principle (based on vector potential perturbations, rather than on plasma displacement) into a new perturbed tokamak equilibrium code. At present, the code uses the reduced MHD model, but its algorithm, based of both equilibrium and Hamada coordinates, allows a straightforward extension to the full MHD as well as to 3-D configurations. The plasma surface model is made suitable for disruption and Wall Touching Kink Mode studies. The code is a section in a more general Cbstb-code, which is under development. It will utilize all advantages of the psiform of MHD, which is unique in providing opportunities to investigate the plasma edge, where the resonant surfaces are densely packed and the conventional stability models are inapplicable. In particular, this form makes obvious the plasma edge stability in the LiWF regimes, what was used in 2005 in predictions of ELMs stabilization by Li conditioning.

An article titled "Hybrid Simulation of Energetic Particle Effects on Tearing Modes in Tokamak Plasmas", by Dr. Huishan Cai of University of Science and Technology of China and Dr. Guoyong Fu of PPPL, has been published in the 2012 July issue of Physics of Plasmas. The abstract reads: "The Effects of Energetic Ions on Stability of Tearing Mode are Investigated by Global Kinetic/MHD hybrid Simulations in a Low Beta Tokamak Plasma. The kinetic effects of counter circulating energetic ions from the non-adiabatic response are found to be strongly destabilizing while the effects from the adiabatic response are stabilizing. The net effect with both adiabatic and non-adiabatic contributions is destabilizing. On the other hand, the kinetic effects of co-circulating energetic ions from the non-adiabatic response are calculated to be weakly stabilizing while the corresponding adiabatic contribution is destabilizing for small energetic ion beta. The net effect is weakly stabilizing. The dependence of kinetic effects on energetic ion beta, gyroradius, and speed is studied systematically and the results agree in large part with the previous analytic results for the kinetic effects of circulating particles. For trapped energetic ions, their effects on tearing mode stability are dominated by the adiabatic response due to large banana orbit width and strong poloidal variation of particle pressure. The net effect of trapped energetic particles on tearing modes is much more destabilizing as compared to that of counter circulating particles at the same beta value."

Leonid E Zakharov gave a talk on "Toward High Performance Plasma Regimes on EAST and Beyond" to the US-PRC Magnetic Fusion Collaboration Workshop (July 10-12, San Diego, California). After invention of the flowing liquid lithium system, the innovative Lithium Wall Fusion regimes became realistic for both EAST and the next step device in China. These regimes resolve well-known issues of the tokamak fusion.

COMPUTATIONAL PLASMA PHYSICS GROUP
A new visualization has been developed of the 3D vector field of plasma flow computed by global gyrokinetic simulations using the GTS code. The visualization shows the direction, magnitude, and structure of turbulence generated intrinsic rotation in a tokamak. Vectors indicate the clockwise and counter-clockwise flows around the torus. Color-coded vectors are drawn at each grid point on the poloidal planes. A color scale was developed to maximize contrast within the most heavily populated range of data while preserving visibility of the global minimum and maximum values. Technical highlights include transferring large amounts of simulation data from NERSC to PPPL using multiple streams, parallel rendering by the VisIt software, and multiple nx client sessions connecting to a persistent server session. Each of the 1,000 time steps is rendered to a high definition image. The images are assembled into an animated movie that is compressed for efficient, high quality playback. A workflow is in place for producing visualizations of new simulations. This will be presented at the upcoming APS meeting by E. Feibush

July 13~

Theory Group

A few Theory Department members attended the 39th European Physical Society Conference on Plasma Physics & 16th International Congress on Plasma Physics held in Stockholm, Sweden, July 2-6, 2012. Jay Johnson presented an invited talk in the basic, space, and astrophysical plasmas sessions entitled " The effect of heavy ions on magnetopause transport", Seung-Hoe Ku gave an invited talk, entitled "Intrinsic rotation in flux-driven global ITG
turbulence simulations" Nat Fisch presented a poster titled "Very High Laser Intensities via Resonant Raman Compression" and the poster "Possibility of Type-I ELMy H-mode operation in ITER hydrogen phase" was prepared by A. Polevoi, N. N. Gorelenkov, S.Yu.Medvedev, S. Konovalov et.al. on a possibility to study important physics questions on TAE driven quasilinear diffusion, NTM and peeling-ballooning limits. The work is a result of the continuing collaboration of PPPL, ITER organization, Keldysh and Kurchatov institutes in Russia. In it a special case of a hydrogen plasma was considered with low B field which is stable or close to being stable with the respect of the modes addressed.

Jay Johnson also visited the International Space Science Institute in Bern, Germany, July 9- 12, 2012; Jay led a team meeting on plasma entry and transport.

On Tuesday, July 10th, Dr. Bruce Scott gave a special theory seminar entitled "Symmetry of Momentum Conservation in Gyrokinetics." A brief overview of the gyrokinetic field theory was given and the specific role of the time-dependent polarization current was discussed. It was shown that the contributions in the zonal toroidal momentum transport equation due to higher-order field components in the Hamiltonian can be put into the same symmetric structure as the lowest-order piece. Hence arguments about symmetry and cancellation should be applied equally to all orders, not just the lowest. Measurements of the PDF of these terms in computations were given.

On Thursday, July 12th, Dr. Kimin Kim gave a theory seminar entitled "delta-f Monte-Carlo calculation of NTV in perturbed tokamaks and demonstration of bounce harmonic resonance." A new delta-f particle code (POCA) was discussed capable of addressing the nonambipolar transport associated with non-axisymmetric magnetic perturbations. The code is using a modified pitch-angle collision operator, preserving momentum conservation. Basic features of POCA were introduced with benchmarking tests in axisymmetry. Neoclassical toroidal viscosity (NTV) torques were calculated and compared with a combined and 1/n theory, indicating essential physics of NTV transport such as the dependence on the perturbed magnetic field and strong resonance. The first numerical demonstration of bounce harmonic resonance was given. Also discussed was the observation of a new type of bounce orbit, modified by the bounce harmonic resonance, which prevents phase mixing of bounce orbit thus enhancing the NTV. Finally, a preparation for experimental application of POCA was addressed.

COMPUTATIONAL PLASMA PHYSICS GROUP

The M3D-C1 group met with Sherry Li and others in the NERSC/LBNL Linear Solvers group to discuss development of an improved linear solver for the preconditioning phase of M3D-C1 at high processor count (10,000p - 40,000p) as is planned for production runs in 2013. The SuperLU_Dist and MUMPS based block Jacobi preconditioners start to deviate from ideal weak scaling in this range of processors. The M3D-C1 formulation approximately decomposes the velocity field into the 3 MHD wave components by way of the Helmholtz decomposition of the velocity vector and the application of annihilation operators on the momentum equation. Our plan is to take advantage of this partial separation in PDSLin, which is an iterative solver based on SuperLU_Dist, to quickly form concurrent approximate factorizations corresponding to each wave type on each toroidal plane and then iteratively refine the solution to take into account the wave-wave coupling. We agreed to start working evaluating this technique as applied to test matrices of this size in the coming weeks.

July 6~

Theory Group

M3D-C1 nonlinear 2F 3D MHD code is now operational in a production mode and is being used to study the different types of sawtooth, snake, and "long-lived mode" behavior that can occur in a tokamak. Separate equations are solved for the 3D time evolution of the magnetic field, the velocities, the electron and ion temperatures, and the density. Sources of heat, particles, current drive, and momentum must be supplied, as well as transport models, much like the TSC code in 2D. This 3D code is unique in that it can take very long time steps, even in the highly nonlinear regime. An individual time step can be 10-100 Alfven wave transit times, allowing it to span the MHD stability, reconnection, and transport timescales in a single simulation which can cover the time range of over 10^5 Alfven wave transit times. This allows us to study the effect of heating and current drive systems, transport processes, and plasma shaping on the MHD activity in the core of the tokamak. Calculations are run long enough so that the system reaches either a stationary state or exhibits time-periodic behavior. Initial studies are focused on validation studies in which we compare results with a series of experiments on the DIII-D tokamak where dramatic differences in the periodic sawtooth behavior were observed in two otherwise identical discharges that had different cross-sectional shapes. The code is also being applied to ITER plasmas to map out under what conditions a stationary internal mode can develop and what effect this has on the 3D distortion of the plasma boundary. This latter activity is part of a new MHD ITPA Working Group to assess 3d plasma distortions that could occur in ITER.

COMPUTATIONAL PLASMA PHYSICS GROUP
NUBEAM is the ion Monte-Carlo orbit integrator used in TRANSP and also available via the NTCC modules library. The standard ion orbit integration is done using a magnetic-flux coordinate system. However, in order to implement a RF operator which is very localized in space, a second orbit integrator was added to NUBEAM in 2007. This, RZ orbitintegrator, uses a cylindrical coordinate system which simplifies coupling with the advanced RF codes such as TORIC, that produce RF diffusion operators in a ( R,Z, v_Per, v_Par) space. The RF "kick" operator as well as the RZ orbit integrator are now again operational in coupled NUBEAM and TORIC routines within TRANSP, that are now both run in a parallel processing mode using MPI. Initial verification runs have been performed with the parallel coupled system and the results presented in the 2007 APS conference have been reproduced.

Jun 29

THEORY
Paper titled "Absence of Debye Sheaths Due to Secondary Electron Emission" by M. D. Campanell, A. V. Khrabrov, and I. D. Kaganovich have just been published in Phys. Rev. Lett. 108, 255001. The abstract reads "A bounded plasma where the hot electrons impacting the walls produce more than one secondary on average is studied via particle-in-cell simulation. It is found that no classical Debye sheath or space-charge-limited sheath exists. Ions are not drawn to the walls and electrons are not repelled. Hence the unconfined plasma electrons travel unobstructed to the walls, causing extreme particle and energy fluxes. Each wall has a positive charge, forming a small potential barrier or “inverse sheath” that pulls some secondaries back to the wall to maintain the zero current condition."

Two models on the quasilinear (QL) theory of energetic particles in the presence of Alfven modes are being developed by N.N.Gorelenkov, H. Berk and K. Ghantous. The first is prepared for publication (IAEA abstract is accepted) in which a truncated phase space is reduced to 1.5 dimensions. It has produced predictions that correlate well for the expected neutron signal drop with the experimental data on DIII-D, although the predicted relaxed EP distribution function should not have all the expected modifications. In a second, QL model, more complete full QL phase space code that is able to bridge the gap between isolated mode interactions to the typical QL theory is under development. In the 1.5D model the effect of TAEs on fast ion beta profiles is evaluated by making expressions for the local growth and damping rates equal and evaluating them either analytically or numerically using the NOVA-K code, which accurately captures
the absolute values of growth rates and thresholds for TAE excitation. In attempts to validate the 1.5D quasi-linear diffusion model we have compared its predictions with experimental DIII-D data. A DIII-D plasma, formed with reversed magnetic shear, created favorable conditions for EP induced TAE excitations. The neutron production rate was calculated from the EP distribution predicted by the QL model. The results were compared with experimental measurements. It was found that the QL prediction for the change in the neutron fusion production due to EP relaxation from TAE instability accurately replicated as a function of time the observed experimental neutron count rate. The quasi-linear investigation is being expanded by increasing the phase space description to 2.5 dimensions (angular momentum, energy change at constant magnetic moment). The diffusion coefficients are calculated for each of the wave-particle resonances. In addition our quasi-linear transport equation treats the cases where there is no mode overlap, partial overlap and full mode overlap. The essential aspect is to allow the resonance width of the wave to be related to the trapping width of the mode amplitude so that the saturation level produced by the code for a single resonance is the same as in analytic theory and/or computer simulation (see the reference). Then the mode width that is achieved determines whether resonances remain isolated, or whether there is mode overlap, which would cause global diffusion of the alpha particles. Thus the more complete, 2.5 D, quasi-linear diffusion covers all the cases. Depending on system parameters, one obtains steady state, nonlinear oscillatory, bursting and even avalanche behavior. H. L. Berk, B. N. Breizman, J. Fitzpatrick, et. al. , Phys. Plasmas 3, 1827 (1996).

COMPUTATIONAL PLASMA PHYSICS GROUP
A web-conference call was held this week between the PPPL NUBEAM group and D. Greene (ORNL) and P. Bonoli and J. Wright (MIT) to discuss collaborating in the coupling of TORIC and NUBEAM in TRANSP. The immediate focus of the PPPL group is to reproduce a TRANSP JET run made in 2007 by J.M. Kwon in which a RF-kick operator is applied to the fast-ions in NUBEAM using the electric fields calculated in TORIC. This is being performed now with parallel versions of NUBEAM and TORIC, whereas in 2007 these codes were serial. D. Greene described his work in coupling a simplified Monte Carlo code, sMC, with AORSA both by the kick-operator and by a more complete collision operator, and comparing the results. The MIT group expressed interest in substituting the faster TORIC code for AORSA in these comparisons. It was agreed that another call should be made later in the summer to report on progress the different institutions.

Jun 22

THEORY
Bill Tang attended an invitation-only International High Performance Computing workshop (HPC 2012) in Cetraro, Italy on June 24-29. He presented an invited talk on "Extreme Scale Computational Science Challenges in Fusion Energy Research" and served on a special panel that led the workshop participants in discussions on the topic: "Five years into Exascale Exploration: What Have We Learned?"

Velocity shear is well-known to drive Kelvin-Helmholtz instability in continuous fluid (or across interface between two fluids with sufficient velocity difference). Flow shear in plasmas, on the one hand, can suppress various turbulence through the associated ExB shear, while on the other hand the free energy associated with it may drive a negative compressibility mode. The plasma flow shear driven modes have been observed in linear machines. However, little attention has been paid to it in tokamak devices because it was generally assumed to be barely driven unstable due to the magnetic shear effect. Recent results from our global nonlinear gyrokinetic simulations with the GTS code show that strong flow shear can drive the negative compressibility mode unstable in tokamak geometry in some experimentally relevant parameter regimes and that the associated turbulence may produce significant momentum and energy transport. In addition to momentum diffusion, the flow shear turbulence can also produce an intrinsic torque in the co-current direction. Since this is a poorly understood new mechanism for driving toroidal plasma transport, the generic nature of flow shear turbulence including linear properties, mechanisms for nonlinear saturation and the influence of the q-profile structure, has been investigated systematically using plasma parameters relevant to tokamak experiments. Linearly, the flow shear driven modes reside in a low-k range, similar to that of ITG mode, with smaller but almost constant growth rate over a wider ktheta range, while the mode frequency increases strongly with ktheta. More interestingly, the flow shear modes show significantly finite k//, unlike ITG and TEM. It is shown that the nonlinear energy transfer to longer wavelength via toroidal mode coupling and corresponding strong zonal flow and geodestic acoustic mode (GAM) generation play a critical role in the nonlinear saturation of the instability. The influence of the q-profile structure in a tokamak is quite different than what was expected from the linear theory in simple geometry. Remarkably, strong “resonance” in the shear flow turbulence and associated transport peaks at the lowest order rational surfaces with integer q-number (rather than fractional), consistent with theoretical calculation. As a consequence, local “corrugations” are generated in all plasma profiles (temperatures, density and toroidal rotation), potentially impacting transport barrier formation near the rational surface. The detailed results of this work,including the study with respect to flow optimization for minimizing plasma transport, will be reported in the upcoming EU-US TTF meeting in September and IAEA FEC 2012 Conference in October. The work has been carried out by W. X. Wang, S. Ethier and their collaborators.

COMPUTATIONAL PLASMA PHYSICS GROUP
Several weeks ago, it was observed when performing a numerical convergence study that a numerical instability developed in the M3D-C1 code at the highest toroidal resolutions (64 Hermite cubic polynomial elements and above). This was reported at the June 12-13 CEMM meeting in Madison. After exhaustive analysis, it was found that this numerical instability arose because the variable called "small f" which appears in the (R,Z) component of the vector potential was not being treated fully implicitly, only it's Laplacian was. The time advance algorithm in M3D-C1 has now been rewritten so that all variables entering into the magnetic field and current evaluation are treated fully implicitly for both resistive MHD and two-fluid MHD. This has cured this numerical instability with only a minimal increase in running time, and the numerical convergence tests show that sawtooth simulations using 32 and 64 Hermite cubic polynomial elements now give very similar results, thus confirming the results from the lower resolution study which are being presented at the upcoming IAEA meeting. This work was done in close consultation with Nate Ferraro of General Atomics.

Jun 15

THEORY
The article, "General Cause of Sheath Instability Identified for Low Collisionality Plasmas in Devices with Secondary Electron Emission" by M. D. Campanell, A. V. Khrabrov, and I. D. Kaganovich has been published online in the June 8 issue of Physical Review Letters (Vol.108, No.23): http://link.aps.org/doi/10.1103/PhysRevLett.108.235001. Kinetic studies by the PPPL researchers using full Particle-in-Cell (PIC) code revealed for a weakly-collisional plasma of Hall thrusters, the EVDF is non-Maxwellian, strongly anisotropic, depleted at high energie, and even non-monotonic. The average kinetic energy of a plasma bulk electrons, which are confined by the sheath potential barrier is several times larger than that in the direction normal to the walls. The plasma bulk electrons reach the walls mostly after scattering off neutral atoms. These collisions are so rare that secondary electrons emitted by the walls propagate through the plasma almost freely, without energy exchange with the plasma electrons. Thus, a wall is bombarded by both the scattered plasma bulk electrons and the electrons emitted from the opposite wall. It appears that while the plasma is heated and the secondary electron emission (SEE) intensity increases, the balance of electron and ion fluxes to the wall is maintained not through the development of the commonly assumed space-charge limited (SCL) sheath regime, but through the modification of the EVDF of the plasma-beam system. This new balance mechanism creates an unusual situation in which the plasma sheath is not in a steady SCL state even though the secondary electron emission produced by the plasma bulk electrons is so intense that the corresponding partial emission coefficient exceeds unity. Instead, the plasma-sheath system performs relaxation oscillations by switching quasi-periodically between the space charge limited and non-space-charge limited states. A condition for sheath instability is derived. When the SEE coefficient of the electrons bordering the depleted loss cone in energy space exceeds unity, the sheath potential is unstable to a negative perturbation. This result explains three different instability phenomena observed in Hall thruster simulations including a newly found state with spontaneous 􀀁20 MHz oscillations. When instabilities occur, the SEE propagating between the walls becomes the dominant contribution to the particle flux, energy loss and axial transport.

On June 12, Dr. Wilfred Cooper gave a special theory seminar entitled "3D Magnetohydrodynamic Tokamak Equilibria; Helical Core, RMP and Ripple." Discussed were the results obtained using the ANIMEC code with fixed boundary, modeling JET and DIII-D, and with free boundary, modeling TCV and MAST, which show that equilibrium states with large internal 'snake-like' helical structure can be obtained in both cases. It was shown that bifurcated equilibria (helical and axisymmetric) exist for weakly core-shear-reversed plasmas with a range of minimum q-values in the neighborhood of unity. Also discussed were free boundary calculations for the MAST tokamak that combine the effects of a helical core structure with Resonant Magnetic Perturbation (RMP) and finite ripple deformations, showing that RMP coils induce the largest plasma edge modulation followed by finite TF coil ripple effects, the 3D helical structure causing only a weak deformation.

On June 13, Dr. Renaud Gueroult gave a special theory seminar entitled "New Plasma Techniques for Mass Separation." Discussed were the causes for the growing need of highthroughput mass separation devices for nuclear waste remediation, and the advantages of plasma filters over other types of mass separation methods. Next a new type of plasma mass filter was presented which operates in the collisional regime and in which the separation of light and heavy elements is accomplished axially, along the axis of rotation, rather than in the radial direction. The mass separation results from the competition between centrifugal and magnetic mirror forces. The axial separation makes the collection of separated elements easier and reduces the overall contamination of the device compared to other plasma centrifuges.

This is the last week of an eight week visit to PPPL by Dr. Tobias Goerler from the Max Planck Institute for Plasma Physics (Garching, Germany), as he departs on June 22. He recently gave a tutorial on how to use the GENE gyrokinetic turbulence code (http://gene.rzg.mpg.de), with 11 theorists and experimentalists attending. During his visit, Dr. Goerler has been interacting with W. Guttenfelder and others on studies of NSTX plasmas and cross-code comparisons, with H. Mynick and B. Faber on stellarators, and with G. Hammett and others on tokamak turbulence simulations in various regimes.

COMPUTATIONAL PLASMA PHYSICS GROUP
A two-day meeting of the SciDAC Center for Extended Magnetohydrodynamics (CEMM) was held June 12-13 at the University of Wisconsin in Madison. Presentations and discussions focused on code development activities, verification and validation, and application to experiments. Significant progress was reported in the abilities of the NIMROD and M3D-C1 codes to include kinetic effects beyond the standard 2F model, as required to model NTMs and NTV. All 22 presentations are posted on: https://w3.pppl.gov/cemm/Madison2012/index.htm.

Jun 8

THEORY
The article, "General Cause of Sheath Instability Identified for Low Collisionality Plasmas in Devices with Secondary Electron Emission" by M. D. Campanell, A. V. Khrabrov, and I. D. Kaganovich has been published online in the June 8 issue of Physical Review Letters (Vol.108, No.23).

On June 5, Dr. Tobias Goerler gave a special theory seminar entitled "Gyrokinetic finite-size effect investigations with GENE." The extension of the Eulerian gyrokinetic turbulence code GENE from flux tube (local) to full torus (global) domain was discussed. Local limit was reproduced in both linear and nonlinear simulations. A modified rho* value which reflects the width of the profiles, not of the device, was identified as a useful parameter. Two extreme cases, an ASDEX-Upgrade L-mode discharge and a TCV electron internal transport barrier, were addressed using simulations with as much physics as available in the present code version (realistic geometry/profiles, electromagnetic effects, collisions).

On June 7, Dr. Peter Damiano gave a theory seminar entitled "Electron Energization and Wave Dispersion in a Mirror Kinetic Alfven Wave." Discussed were hybrid MHD-kinetic electron simulations in a dipole geometry which suggest that potential drops sufficient to accelerate electrons to keV energies can result from the effect of equilibrium mirror force on kinetic Alfven wave dynamics. The wave energy is dissipated by the electrons within only a few Alfven cycles. The perpendicular Poynting flux associated with the parallel electric field disperses wave energy perpendicular the the magnetic field leading to a broadening of the parallel current region as electrons are accelerated along adjacent field lines. This mirror force induced wave dispersion dominates over that associated with electron inertial effects.

COMPUTATIONAL PLASMA PHYSICS GROUP
Professor Ravi Samtaney from KAUST, Saudi Arabia (and former PPPL CPPG staff) presented a CPPG seminar on "Plasmoid Dynamics in Resistive MHD Simulations of Magnetic Reconnection". He described 2D numerical simulations of reconnection at very high Lundquist number (S) that have confirmed the validity of the linear theory of plasmoid formation, and have shown that plasmoids quickly grow to become wider than the thickness of the original SP current sheet, thus effectively changing the underlying reconnection geometry. He also presented some preliminary results from three-dimensional simulations of resistive high-Lundquist number reconnection in the presence of a guide field.

Jun 1

THEORY
On May 24, Imène Rym Goumiri gave a seminar entitled "Reduced-order-model Based Feedback Control of the Hasegawa-Wakatani Equations." Development of model-based feedback control for stabilization of an unstable equilibrium in the Hasegawa-Wakatani (H-W) model was discussed. The (H-W) model was first reduced using snapshot-based approximate balanced truncation, and then a model-based feedback controller was designed for this reduced order model using linear quadratic regulators. The controller was then applied to the original linearized and nonlinear H-W equations to stabilize the unstable equilibrium and suppress transition to drift-wave turbulence.

Gianluca Spizzo from CNR Instituto GAS Ionizzati in Italy has been visiting the Theory department for the past few weeks. During his stay his host has been Roscoe White and they have been working on the construction of a model describing the potential controlling ambipolar flux near the edge of RFX in the presence of the stochastic fields produced by the known tearing mode spectrum, and simulating the ion and electron transport in this region.

COMPUTATIONAL PLASMA PHYSICS GROUP
Significant progress has been made in the incorporation of a new solver algorithm in TRANSP that is able to advance the temperature profiles using the TGLF routine to compute the heat fluxes. The parallel, implicit algorithm makes use of techniques described in [G. V. Pereverzev, et al. Comput. Phys. Comm, 179, p 579 (2008)]. Good agreement was obtained with XPTOR for several DIII-D time slices in which flow shear was included. Successful tests were also performed reproducing experimental data on JET and TFTR. A complete simulation of a 2- second TFTR discharge using 16 processors finished in 18 hours. Additional tests are being performed to demonstrate the applicability of the algorithm over a wide range of discharge parameters. It is expected that this capability will be made available to TRANSP users sometime over the summer.

May 25

THEORY
ITER plans to rely upon the so-called “H-mode” (high confinement) edge plasma operation in order to produce ten times higher fusion energy than input heating energy. It has been long known empirically that one of the essential conditions in achieving the H-mode operation at a lower plasma heating power is a good wall conditioning. A better wall conditioning will reduce neutral particle influx into edge plasma, as well as the impurity particle influx. The confinement enhancement from reduction of incoming neutrals has been further suggested by coating of material wall with lithium in NSTX and EAST. For the first time, we have succeeded in a multiscale gyrokinetic simulation of tokamak edge turbulence together with the neutral particle recycling effect, using the XGC1 multiscale edge gyrokinetic code in realistic diverted edge geometry including a magnetic separatrix. XGC1 indeed shows that the reduction of neutral particles reduces the edge plasma turbulence and the consequential heat transport. The new capabilities and findings we report here may provide another basic ingredient in the H-mode transition physics, which remains to be one of the highest priority items for ITER research.

COMPUTATIONAL PLASMA PHYSICS GROUP
M. Gorelenkova has generated a module for NUBEAM in TRANSP to calculate excited-states beam deposition rates using ADAS atomic cross sections. These self-consistent calculations improve those of theapproximate model (Janev-Boley-Post) previously used in TRANSP. R. Budny used this to compare the total beam deposition rate for a DIII-D shot (147626) with offaxis beams. The peak deposition rate is reduced by 5-20% compared with the value from the approximate model. The shine-through rate is increased by 6%. He reported on similar comparisons using PTRANSP for ITER baseline and pre-activation predictions at the ITPA IOS meeting in Madrid. Those showed less differences.

May 18

Mr. Wei Shen, a graduate student from Institute for Fusion Theory and Simulation (IFTS), Zhejiang University, China, has finished a six month visit to PPPL from November to May, and returned to China. He was hosted by Dr. Guoyong Fu. During his visit, he worked with Guoyong Fu and Josh Breslau on nonlinear simulation of sawteeth and energetic particle transport in tokamak plasmas using M3D code and presented the initial results in a recent theory seminar at PPPL. He will continue this collaborative research in China as a part of his Ph.D. thesis work.

May 11

S. Ethier, E. Feibush, and S. Jardin attended the two-day PICSciE Conference "Research Computing at Princeton" held at the Princeton Center for Theoretical Science. Ethier gave a presentation "ITER-Relevant Simulations of High-Temperature Fusion Plasmas in Tokamaks" and Jardin presented a poster "Multiple Timescale Simulations of the Global Dynamics of
Highly Magnetized Plasmas".

Allen Sanderson from Utah State University visited for two days to discuss enhancements he has made to the VISIT visualization software that was requested by the M3D, M3D-C1, and GTS projects. He also engaged in discussions about future work, in particular a diagnostic for automatically calculating the widths of magnetic islands.

A TRANSP Developers WEB-Conference Call was held on Friday to discuss the TRANSP code Development Plan with external developers. Besides PPPL, participants were from Lehigh, ORNL, MIT, and JET. This call concentrated on the plans for TRANSP regarding how to handle multiple parallel components such as NUBEAM, TORIC, and PT-SOLVER.

May 4

Theory Group

On April 30, Dr. Xianzhu Tang gave a special theory seminar entitled "Plasma Physics Issue in ICF Target Performance." Recent theoretical and computational work was discussed, emphasizing the importance of plasma physics effects on predictions of target performance in inertial confinement fusion (ICF) which is commonly investigated using neutral fluid radiationhydrodynamics models. The talk focused on how self-generated magnetic field in target implosion can affect the hydrodynamic mix model and the target energy transport; and how electric field and other thermodynamic cross terms in a multi-component collisional plasma can drive fuel ion separation which can significantly affect the fusion yield. On April 25, Joshua Burby gave a special theory seminar entitled "Gyrosymmetry: Global Considerations." The general condition for when the unit vectors perpendicular to the magnetic field, and therefore the gyrophase, cannot be defined globally, was presented. Specific examples which do not satisfy this condition were given and it was demonstrated that global perpendicular unit vector field always exists in toroidal geometries. Finally it was shown that a global convention for measuring gyrophase is unnecessary for the validity of the gyrocenter expansion of the single particle equations of motion.

COMPUTATIONAL PLASMA PHYSICS GROUP
A new version of the ElVis scientific graphics software was deployed for improved interoperability with TRANSP rplot. The new configuration increases performance andreliability. ElVis runs on the user's personal computer and starts rplot on the PPPL cluster. The legacy rplot command line interface is displayed in an xterm and automatically connected to the ElVis display. The user enters commands and scripts in rplot directly. All plots from rplot sessions are accumulated in ElVis graph windows for interactive exploration and publication quality output. An elvis module has been added to the PPPL cluster to organize access to the software.

S. Jardin participated in a review of the German Institute of Plasma Physics (IPP) at the request of the Max-Plank Society. The 2-day review covered the ASDEX-U, Wendlestein 7X and theory and computation programs.

Apr 20

Theory Group

Members from the Theory Department attended the 2012 International Sherwood Fusion Theory Conference, in Atlanta, Georgia on March 31-April 3. Allen Boozer gave a plenary talk titled,"Rotation of Tokamak Halo Currents", and C.S. Chang, gave a plenary talk titled " Unveiling the Kinetic Mechanism of RMP Penetration". Ilya Dodin presented a talk titled "Axiomatic Approach to Wave-particle Interactions and its Applications to Waves with Trapped Particles". Guoyong Fu presented three posters and the following department members, H. Mynick, W. W. Lee and S. Ethier, Nikolai Gorelenkov, Greg Hammett, Roscoe White, Luca Guazzotto and Theory Graduate Student Jessica Baumgaertel presented posters.

Theory department members also attended the 2012 Transport Taskforce Workshop in Annapolis, Maryland on April 10-13. C. S. Chang, gave a plenary talk titled "Bootstrap Current for Edge Pedestal Plasma in Diverted Tokamak Geometry" Seung-Hoe Ku presented a talk titled"Intrinsic Momentum Generation and Inward Transport in Realistic Diverted Tokamak Geometry with Neutral Recycling" and Wenjun Deng presented a talk titled, “Gyrokinetic Particle Simulations of Reversed Shear Alfv en eigenmodes in DIII-D tokamak". Guoyong Fu gave an oral presentation titled "M3D-K Simulation of Beam-Driven Alfven Modes in DIIID" and Theory Graduate student Katy Ghantous gave an oral presentation titled "Revisiting Line Broadened QL Burst Model: Applications on TAE-EP Interaction". Weixing Wang, presented a poster titled "Flow Shear Driven Turbulence and Transport in Tokamak Plasmas".

On April 16, Dr. Natalia Tronko gave a special theory seminar entitled "A Close Guiding-center Look at Gyrokinetic Theory." Two representations of higher order Hamiltonian guiding-center theory were identified, symplectic and Hamiltonian. The higher order effects were derived in the symplectic representation using guiding-center Banos hierarchy to help guide the derivation. Magnetic moment invariance was shown to guarantee the equivalence of the two representations. Guiding-center polarization and magnetization were shown to arise naturally from higher-order
guiding-center theory within the two-step perturbative approach to nonlinear gyrokinetic theory, i.e. particle to guiding-center, and guiding-center to gyrocenter.

On April 19, Professor Anatoly Spitkovsky, gave a theory seminar on "Plasma Physics of Collisionless Shocks: from Astrophysics to Laboratory." Outstanding issues in plasma physics that govern the structure of collisionless shocks and their ability to accelerate particles were discussed. Shock properties as function of flow composition and velocity were addressed in context of large scale ab-initio kinetic simulations. Also discussed was the progress at creating collisionless shocks on the Omega laser, and physical constraints that need to be satisfied by such experiments at high-energy density facilities.

COMPUTATIONAL PLASMA PHYSICS GROUP
S. Jardin attended the workshop "Computational Challenges in Magnetized Plasma" at the UCLA Institute for Pure and Applied Mathematics (IPAM). He presented an invited overview lecture: "Multiple Timescale Simulations of Global Macroscopic Dynamics of Magnetized Plasma". Most of the material from this talk can be found in the PPPL report: PPPL-4717 (2012) which has now been accepted for publication in the IOP Journal "Computational Science and Discovery".

Apr 13

Theory Group

On April 5, Dr. Maxime Lesur gave a seminar on "Phase-Space Turbulence, and Nonlinear Instabilities Driven by Self-Organized Structures." A new theory was described in which the growth of coherent phase-space structures, called holes and clumps, can drive the wave by direct momentum exchange due to dissipation. Also discussed was numerical evidence of breakdown of the quasi-linear theory in the presence of structures, which showed that coalescing holes survive much longer than the classical quasilinear diffusion time and dominate the nonlinear
evolution.

On April 6, Prof. Zhihong Lin gave a special theory seminar on "Gyrokinetic Particle Simulations of Kinetic-MHD processes." The presentation summarized the status of firstprinciple simulations of kinetic-MHD processes using the gyrokinetic particle code (GTC). Recent progress in the studies of nonlinear wave-particle interactions underlying the transport processes was also outlined, including: convective flux driven by the constraint of the longitudinal invariant in the trapped electron mode turbulence; nonlinear frequency oscillation of Alfven eigenmodes induced by phase space coherent structures; and scaling of energetic particle transport due to wave-particle decorrelation and orbit-averaging.

The paper "Magnetohydrodynamic Simulations of Edge Poloidal Flows" by L. Guazzotto and R. Betti was accepted for publication in Nuclear Fusion.

Allan Reiman wrote, and submitted to DOE, the Second Quarter Report on the FY 2012 FES Theory Milestone. The work in the second quarter focused on equilibrium calculations for two DIII-D shots by four codes: VMEC, IPEC, MARS and the linearized version of M3D-C1. Significant differences were seen between the nonlinear and linearized equilibrium solutions, and progress was made in developing an understanding of the sources of those differences. The participants in the second quarter research were N. Ferraro (GA), M. Lanctot (LLNL), E. Lazarus (ORNL), S. Lazerson (PPPL), J.K. Park (PPPL), A. Reiman (PPPL), and A. Turnbull (GA).

COMPUTATIONAL PLASMA PHYSICS GROUP
The split-implicit algorithm in the3D nonlinear extended MHD code M3D-C1 involves three independent large sparse-matrix solves each time step for (1) the velocity matrix, (2) the pressure matrix, and (3) the magnetic field matrix. When the normalized ion skin depth, d_i, is non-zero (i.e. two-fluid MHD) the magnetic field matrix becomes non-normal (non-symmetric with complex eigenvalues) and the iterative solve becomes time-consuming and may not converge at all for certain parameters. We have computed all the eigenvalues of the magnetic field matrix for some typical cases with non-zero d_i and found that while the same block-Jacobi preconditioner used in the velocity matrix solve reduces the spectral radius of a problem with over 3500 degrees-of-freedom (DOF) from over 10^8 to around 30, the matrix solve using GMRES will not converge for d_i larger than about 0.05. We have now introduced another preconditioner stage (based on the differential approximation form of the equations) that further reduces the spectral radius somewhat (about a factor of 2), but more importantly, introduces large symmetric terms into the matrix, making it less non-normal. With this new preconditioner, all cases of physical interest will now converge in an acceptable number of iterations.

At the request of the UKAEA MAST team, the `get_fbm' program (TRANSP post-processing software) has been extended to include an option to produce simulated 2-D distributions of neutron production rates over the plasma cross-section. These are now output in convenient NetCDF format on the irregular spatial grid used in the Monte Carlo NUBEAM package, with the cylindrical (R,Z) coordinates of the grid vertices being specified. This new capability was requested to help with the interpretation of data from the most recent experimental campaign in which a new neutron camera was installed.

Mar 30

Theory Group

On March 21, Dr. Gregory Kagan gave a special theory seminar on "Neoclassical Theory in a Tokamak Pedestal." Kinetic calculation to find pedestal modification to the conventional neoclassical prediction for the banana regime main ion flow, as well as the zonal flow residual, was discussed in the context of evaluation of the main ion orbits while accounting for the radial electric field inherently present in a subsonic H-mode tokamak pedestal. Theoretically deduced impurity flow and the bootstrap current from the kinetic calculations was compared with the boron flow measurements in the banana regime C-Mod pedestals. The results show that accounting for the radial electric field effect on the main ion orbits improves agreement between theory and experiment. The comparison supports the conclusion that the bootstrap current is enhanced in banana regime pedestal and verifies the role of the radial electric field in modifying the main ion flow.

On March 22, Dr. Edward Startsev gave a theory seminar on "Finite-Beta Simulation of Microinstabilities." A new scheme for gyrokinetic particle simulation of high beta plasmas was discussed. In the new scheme the perturbed particle response, associated with the quasi-static bending of the magnetic field lines, is separated out from the total particle response. The finitebeta stabilization of drift-waves and ion temperature gradient modes was demonstrated using a simple gyrokinetic particle code based on realistic fusion plasma parameters. The proposed scheme is most suitable for studying shear-Alfven physics in general geometry using straight field line coordinates, for microturbulence and magnetic reconnection problems. Problems associated with the resolution of electron skin depth when beta is much larger than the electron to ion mass ratio were also addressed, together with possible solutions.

A paper titled "Nature of Energetic Ion Transport by Ion Temperature Gradient Driven Turbulence and Size Scaling” by a PPPL former visiting student Jugal Chowdhury (from Institute for Plasma Research, India), and W. X. Wang, S. Ethier, J. Manickam and R. Ganesh (IPP, India) has been accepted for publication in Physics of Plasmas. The paper reports interesting results regarding ITG turbulence driven energetic particle transport obtained from global gyrokinetic nonlinear simulations using the GTS code. The measured transport and its nature show dependence on the system size of the tokamak expressed as the ratio of plasma minor radius to the thermal ion Larmor radius. It increases with system size initially and then tends to saturate at larger system size. The nature of transport, on the other hand, exhibits nondiffusive character for smaller system size, and gradually becomes diffusive as the system size becomes larger. This work was mainly carried out during Jugal Chowdhury's visit at the PPPL theory department.

On March 27, Professor Scott Parker gave a special theory seminar on "Bootstrap Current Destabilization of the Kinetic Ballooning Mode in the Tokamak Edge Pedestal." Results obtained using global gyrokinetic simulations were used to demonstrate that the high-n kinetic ballooning mode (KBM) is the dominant instability in the tokamak edge pedestal assuming the bootstrap current locally flattens the q-profile in the steep pressure gradient region. In addition to KBM, an intermediate-n electromagnetic mode was found to be unstable whose features are similar to the MHD peeling-ballooning mode. This mode, called "kinetic peeling ballooning mode" (KPBM) was stabilized when the magnetic shear was weak in a small region near the steep pressure gradient, while the high-n KBM became unstable. Collisions decreased the KBM's critical beta and increased the growth rate. The simulations were carried out using the global electromagnetic GEM code, including kinetic electrons, electron-ion collisions and the effects of realistic magnetic geometry.

COMPUTATIONAL PLASMA PHYSICS GROUP
Stephane Ethier gave an invited talk entitled "Computational Advances in Gyrokinetic Particlein Cell Simulations" at the "Accelerating Computational Science Symposium 2012" held in Washington, DC, on March 29-30. The purpose of this Symposium, which was sponsored by the Department of Energy and by the National Science Foundation, was to advance our understanding of how extreme-scale hybrid-computing architectures are accelerating progress in scientific research.

Mar 23

Theory Group

A paper entitled "Coupled Core-Edge Simulations of H-mode Buildup Using the Fusion Application for Core-Edge Transport Simulations (FACETS) Code" by A. Hakim et. al was published in Physics of Plasmas, 19 032505 (2012) (URL: http://dx.doi.org/10.1063/1.3693148). This paper reports results from coupled core-edge simulations of density and temperature evolution of a DIII-D discharge and studies the impact of gas puff on pedestal profile evolution

COMPUTATIONAL PLASMA PHYSICS GROUP
Stephane Ethier gave a mini-course entitled "Introduction to Parallel Debugging" as part of the series of mini-courses hosted by the Princeton Institute for Computational Science and Engineering (PICSciE). The lecture introduced the Princeton University participants to different approaches in tackling the task of debugging parallel applications, including the use of advanced parallel debuggers, such as Totalview and DDT.

Mar 16

THEORY GROUP
On March 13, Daren Stotler, Bill Davis and Greg Tchilinguirian ran the Robot Arm competition at the New Jersey Science Olympiad State Tournament. Each of the 23 high school teams participating in Robot Arm designed and built a stationary robot capable of picking up various objects and placing them in one of four "goal" boxes.

On March 15, Feng Wang and Wei Shen gave a joint seminar on 'Simulation of Non-Resonant King Mode and Fast Ion Transport in NSTX' and 'M3D-K Simulations of Sawteeth and Energetic Particle Transport in Tokamaks', respectively. In the first part of the seminar, Feng Wang discussed results obtained from nonlinear MHD simulations using the M3D code, initialized with measured NSTX plasma profiles. It was found that when the rotation velocity is near the experimental level, its effect on equilibrium and linear stability is small. In the nonlinear regime the (m=1,n =1) mode is also weakly affected. However, with finite rotation the width of the saturated (m=2,n=1) island is greatly reduced as compared to that of non-rotating case. The simulated Soft X-Ray (SXR) emission from M3D agreed qualitatively with the measurement. The fast ion transport was also investigated using the M3D-K code. The initial simulation results show that the fast beam ion distribution is flattened radially due to the saturated (1,1) mode. In the second part of the seminar, Wei Shen discussed nonlinear simulations of sawteeth and energetic particle transport simulations using the M3D-K code. MHD simulations showed repeated sawtooth cycles due to a resistive (1,1) internal kink mode for a model tokamak equilibrium with a broader pressure profile. However, sawteeth did not occur for a peaked pressure profile. In this case, a steady state saturated (1,1) mode follows the initial sawtooth crash. Test particle simulations were also carried out to study the energetic particle transport due to a sawtooth crash. The results show that energetic particle distribution is flattened depending on pitch angle and energy. For trapped particles, the redistribution occurs for particle energy below a critical value.

COMPUTATIONAL PLASMA PHYSICS GROUP
A new visualization has been developed of the 3D vector field of plasma flow computed by global gyrokinetic simulations with the GTS code. The visualization shows the direction, magnitude and structure of turbulence-generated intrinsic rotation in tokamak for each time step. Color-coded vectors are drawn at each grid point on poloidal planes. Technical highlights include transferring large amounts of data from NERSC to PPPL using multiple streams, parallel rendering by the VisIt software, and multiple nx client sessions connecting to a persistent server session. The animated movies were assembled with software developed at PPPL. A workflow is in place for producing visualizations of new simulations. Work performed by E. Feibush, S. Ethier, and W. Wang.

Stephane Ethier attended the Nu-FuSE project meeting in Edinburgh, UK. Nu-FuSE is an international project (funded through the G8 Research Councils Initiative on Multilateral Research Funding) looking to significantly improve computational modeling capabilities to the level required by the new generation of fusion reactors. The project, still in its first year and making good progress, is led by Professor Graeme Ackland at The University of Edinburgh, and includes research teams in Cadarache (France, and also the location of the next generation of Fusion reactors, ITER), Edinburgh (UK), Princeton (USA), Garching and Juelich (Germany), Keldysh Institute of Applied Mathematics (Russia), and Tsukuba (Japan). Dr. Ethier made two presentations during the workshop: "Computational Advances in Large-Scale Gyrokinetic Particle-in-cell Codes for Tokamak Fusion Simulations", and "Confinement Physics of Fusion
Grade Plasmas: Scientific and Associated 'Path to Exascale' HPC Challenges".

Mar 9

THEORY GROUP
On March 6, Dr. Slava Lukin of Naval Research Laboratory gave a seminar on "Simulations of magnetic reconnection in laboratory and space with the HiFi multi-fluid modeling framework." The methods and underlying principles used in the development of this implicit spectral element multi-fluid framework were described together with its various domains of applicability. Results for 2D reconnection in single-fluid MHD with anomalous resistivity, in Hall MHD, in plasmaneutral two-fluid MHD, as well as fully 3D reconnection in single-fluid and Hall MHD through a moving reconnection region were discussed. The 3D simulations were presented in context of two interacting spheromaks constrained by a perfectly conducting cylindrical boundary and oriented to form a single magnetic field null at its center. Correlation between the reconnection region and the magnetic null were observed, together with stronger reconnection in plasmas with lower thermal to magnetic pressure ratio.

COMPUTATIONAL PLASMA PHYSICS GROUP
S. Jardin attended the combined US-Japan Workshop on MHD and ITPA/MHD meeting at the NIFS Institute in Japan. He made a presentation at the former on "Transport timescale calculations of sawteeth and other m=1 activity in tokamaks" and gave a report at the latter on Working Group WG-10: "Halo Current Modeling", for which he is the Chair.

Stephane Ethier attended the "Industry-National Laboratory Workshop on Modeling and Simulation" held in Austin, Texas. This is the second in a series of workshops organized by DOE to increase industry awareness of relevant research capabilities within the DOE national laboratory system, deepen the national laboratories’ understanding of the technical challenges facing industry, and identify and improve paths forward for collaboration. Dr. Ethier presented a poster describing PPPL's major activities in the realm of modeling and simulation, from engineering design and industrial plasma simulations to advanced physics calculations with large-scale parallel codes.

Feb 24

THEORY GROUP

Professor Jim Drake of University of Maryland gave a seminar on February 22 entitled, "Particle Acceleration During Magnetic Reconnection." Models for heating and production of energetic particles were discussed in two distinct environments: impulsive flares, and the sectored regions of the heliosheath (regions with radially periodic reversal of azimuthal magnetic field). Multiisland reconnection model was suggested to be necessary to explain the observations in both cases. For the case of solar flares the model was used to explain the recent observations that all electrons in the flaring region undergo acceleration, and that the acceleration of high mass-tocharge ions is more efficient. As for the sectored regions of the heliosheath, the Voyager spacecraft has revealed that the termination shock is not the source of anomalous cosmic rays. Here the model suggests that the source of the anomalous cosmic rays may be the compressed sectored structure of the heliospheric magnetic field which undergoes collisionless reconnection. The high beta environment of the heliosheath leads to magnetic island dynamics that differ greatly from the low beta environment of flares but in both cases Fermi reflection in contracting islands drives the plasma towards the marginal firehose condition, which ultimately controls the spectra of energetic particles.

COMPUTATIONAL PLASMA PHYSICS GROUP
An Audio Web conference call was held this week between the TRANSP code developers (and some users) at PPPL, MIT, and JET. Jim Conboy from JET gave a web presentation on"TRANSP at JET". A number of issues were discussed regarding the deployment of parallel TORIC and NUBEAM within TRANSP at the two sites and maintaining compatibility. Also discussed were the obstacles encountered when trying to use TRANSP at PPPL from JET. Among the actions taken were steps to install the exact same TRANSP version used at JET as a branch of the repository at PPPL so that compatibility can be tested on the PPPL unix system before being committed into the PPPL TRANSP SVN repository.

Feb 17

THEORY

On February 13, Dr. Robert Hager gave a Special Theory Seminar entitled "Nonlinear dispersion relation of the geodesic acoustic modes." He discussed results obtained using two-fluid turbulence simulation code NLET. Some topics covered were frequency shift of geodesic acoustic modes (GAMs) and energy input induced by turbulence; dependence of radial phase velocity, in nonlinear stage, on up-down asymmetric equilibrium magnetic field; variation of GAM and turbulence intensity with the direction of radial phase velocity with respect to magnetic drifts. If the equilibrium magnetic field is up-down asymmetric, his simulations showed periodic bursts of turbulence and GAM activity which he used to interpret the experimentally observed pulsations during the I phase in ASDEX Upgrade and the periodic turbulence suppression in NSTX.

COMPUTATIONAL PLASMA PHYSICS GROUP

A major restructuring of the nonlinear operating mode of the M3D-C1 code was accomplished this week. The normal algorithm used in the code is the "split implicit method" where the velocity field at the new time level is first solved for, and then the magnetic field and pressures are advanced. We have added new options to allow the magnetic field and pressure advance to be either combined or done separately, and an option to allow equations to advance the electron and ion temperatures instead of the pressures. The first option resulted in a 20% speedup in many cases because it is more efficient to invert two smaller matrices than one large one. The second option was necessary to obtain a numerically stable time advance in the presence of a nonuniform density and large parallel electron thermal conductivity. These improvements were done in close collaboration with N. Ferraro of General Atomics.

S. Jardin participated in a conference call with representatives from ITER, India and the ITER Organization (IO) regarding the use of the TSC code to calculate the generation of runaway electrons during an impurity-pellet induced plasma disruption and the subsequent control of the runaway beam. The team from ITER, India has begun a series of TSC simulations using a high resolution model of the ITER plasma and surrounding structure and coils and the existing TSC pellet and runaway electron models. Plans for an IAEA paper on this work, and on subsequent experimental calibrations using JT-60 and DIII-D data were discussed.

Jin Chen attended SIAM Parallel Processing for Scientific Computing (Feb 14-17, 2012) and reported on an algorithm and numerical implementation for calculating eigenvalues from the M3D-C1 3 dimensional nonlinear MHD simulation, both before and after the pre-conditioner is applied. In the near future, she has plans to test the SS method using contour integration to speed up the convergence.

Feb 10

COMPUTATIONAL PLASMA PHYSICS GROUP

Stephane Ethier attended the annual NERSC Users' Group (NUG) meeting held at the NERSC facilities in Oakland, California. The three-day meeting consisted of two days of user training on all topics related to computing at NERSC, followed by a one-day business meeting during which participants learned about NERSC's current and future activities. It is also an opportunity for the users to give direct feedback to the NERSC staff and administrators about their overall experience with working on the NERSC systems. The business meeting was very informative and well attended by most members of the NUG Executive Committee, of which Dr. Ethier is the chair.

Feb 3

COMPUTATIONAL PLASMA PHYSICS GROUP

Several modifications and improvements have been made to the Tokamak Simulation Code (TSC) as requested by different research groups. A capability has been added to switch from "non surface-averaged transport" to "surface-averaged" transport at restart time. This was requested for modeling NSTX discharges that begin as CHI discharges and transition to OH discharges. A capability was restored to model impurity pellet injection [as done in Jardin, et al, "A fast shutdown technique for large tokamaks, NF 40 923 (2000)] as requested by ITER, India. A capability was added to import LHCD heating and current drive profiles from GENRAY via the SWIM framework.

J. Chen succeeded in calculating all the eigenvalues of the implicit M3D-C1 matrix for a small nonlinear 3D application both before and after the preconditioner was applied. This allows us to quantitatively measure the effectiveness of the preconditioner. We were thus able to show how dramatically effective the SuperLU_dist-based block Jacobi preconditioner is in reducing the condition number (ratio of the largest to smallest eigenvalue) from over 10^15 to approximately 30. After preconditioning, the matrix equation is solved in just 10's of iterations using the PETSc iterative solver GMRES.

X. Yuan visited G. Stabler and others at General Atomics regarding the TGLF transport module and its incorporation into transport codes. As a result of that visit, both sides now agree that there is adequate agreement in the fluxes computed by TGLF as installed in XPTOR (GA) and PTSOLVER (PPPL). PPPL and GA also jointly modified TGLF to make it parallel in the wavenumber evaluations. There were also discussions regarding transferring the capability of PTSOLVER to be parallel over flux surfaces to XPTOR

Jan 27

THEORY GROUP

On January 26, Dr Leonid Malyshkin of the Department of Astronomy and Astrophysics at the University of Chicago presented a Theory Department seminar entitled "Two-fluid Physics of Magnetic Reconnection". Magnetic reconnection is sometimes fast and sometimes slow in laboratory and astrophysical plasmas, depending on the plasma collisionality and the mechanism by which the magnetic field lines are broken. Highlighting the importance of understanding what drives the fast and slow regimes, Dr. Malyshkin presented a theoretical model he developed of two-fluid MHD reconnection to explain the physical processes which are central for distinguishing the two regimes. In the slow reconnection regime the resistive and Hall terms are important, while the electron inertia does not play any role. In the fast reconnection regime the electron inertia terms are important. He summarized how it is possible that the presence of the two reconnection regimes can provide an explanation for the initial slow build up and subsequent rapid release of magnetic energy observed in cosmic and laboratory plasmas.

COMPUTATIONAL PLASMA PHYSICS GROUP

The M3D-C1 code is now running on the recently installed PPPL computer Greene (named after the late PPPL physicist John Greene). Production runs are being performed using the 256 processors that each have 8GB of memory. Since the memory per processor is approximately 3 times that of the Hopper computer at NERSC, this allows us to run the same jobs that require 768p on Hopper. Typical runs use 16 planes (16 Hermite cubic elements in the toroidal direction) with the mesh points on each plane spread across 16 processors, with good efficiency and load balancing. The availability of this resource is proving invaluable in further M3D-C1 development, in particular in the optimization of the nonlinear implicit algorithms when density evolution and two-fluid effects are present. Dr. Jai Sachdev from Combusion Research and Flow Technology presented a CPPG Seminar on: "Numerical Solution of a Dilute and Disperse Gas-Particle Flow". In this seminar, the equations governing the motion of a dilute and disperse gas-particle flow, the mathematical characteristics of these equations, and methods for their solution were presented. Numerical results were also described that demonstrate the capabilities of different approaches for the solution of coupled gas-particle flows.

January 20

THEORY GROUP

There were four Theory seminars during the week of January 16. On January 16, Dr. William Daughton of LANL presented a seminar entitled "Spontaneous Generation of Turbulence in Magnetic Reconnection" in which he discussed results of 3D simulations of magnetic reconnection using the VPIC simulation code and highlighted the complex differences the extra dimensionality introduces over the 2D system.

On January 17, Dr Xin Wang of the Institute for Fusion Theory and SImulations of Zhejiang University gave a seminar entitled "Nonlinear dynamics of Beta-induced Alfven eigenmode excited by Energetic particles in Tokamak plasmas" in which she discussed nonlinear BAE simulations excited by purely circulating energetic particles in Tokamak plasmas using an extended version of the Hybrid Magetohydrodynamic Gyrokinetic Code (XHMGC).

On the same day, Dr Zhiyong Wang (also of Zhejiang University) presented a seminar entitled "Spatialtemporal evolution of the nonlinear DW-GAM system" which highlighted work investigating the effects of plasma nonuniformities such as the GAM (Geodesic Acoustic Mode) continuum, nonuniform DW frequency and nonuniform pump wave amplitude on GAM/KGAM excitation.

The final seminar entitled "Capitalizing on a better understanding of plasma turbulence" was presented by Professor Frank Jenko of the Max Planck Institute for Plasma Physics in Garching on January 18. During the seminar he discussed the basic character of plasma turbulence in comparison with that of fluid turbulence. He discussed the limits of quasilinear theory in describing the characteristics of the plasma turbulence and how this useful approach must be complimented by simulations (such as from the GENE code) in order to elucidate nonlinear effects and how the insight gained can be used to develop advanced transport models.

COMPUTATIONAL PLASMA PHYSICS GROUP
The ADAS310_FORTRAN_DRIVER has now been implemented and validated in the energetic ion program NUBEAM. It is based on the ADAS310 program from ADAS (Atomic Data and Analysis Structure) library. The program calculates the exited population structure, effective ionization and charge-exchange coefficients of the hydrogen atom(ions) and its isotopes in an impure plasma. A very many n-bundle-n approximation is used. The ADAS310_FORTRAN_DRIVER provides users with atomic physics data computed using the excited state model. To use this option in NUBEAM together with TRANSP, the variables LEV_NBIDEP=2 and NSIGEXC=3 need to be set in the namelist.

January 13

As the POC for the FES FY 2012 Theory Milestone, Allan Reiman submitted the first quarter report on the milestone to DOE on January 6. The work on the milestone focuses on understanding experiments on the DIII-D tokamak in which relatively small nonaxisymmetric fields are imposed for the purpose of suppressing edge localized modes (ELMs). There are 17 participants involved in the project from four different DOE laboratories and three universities. ELMs pose a threat to ITER, and the only method that has been demonstrated to suppress ELMs without reducing global plasma confinement is the imposition of a nonaxisymmetric field as pioneered by DIII-D. There is as yet no first-principles understanding of this effect. A firstprinciples understanding would greatly increase our confidence in extrapolating these results to ITER. The milestones for the first quarter have three elements: 1) Develop a preliminary coordinated research plan among the groups that will be contributing to this research; 2) Assemble data from at least two DIII-D ELM suppression experiments in a form that can be used by 3D equilibrium codes; and 3) Provide a theoretical assessment of measurement options for ITER as-built field errors.

January 6

The paper entitled "Electron acceleration in a geomagnetic Field Line Resonance" by P.A. Damiano and J.R. Johnson has been accepted for publication in the journal Geophysical Research Letters. In this paper, a 2D hybrid MHD-kinetic electron model in dipolar coordinates was used to conduct the first kinetic simulation of a global scale Field Line Resonance system (standing shear Alfven wave along the Earth's closed dipolar magnetic field lines) for realistic magnetospheric electron temperatures of a keV. It was found that mirror force effects self consistently lead to the formation of field aligned potential drops that accelerate electrons to greater than the keV energies needed to power auroral emissions. It was also determined that the wave energy dissipated in the electron acceleration was sufficient to damp an undriven resonance structure in under two Alfven periods.

2011 highlights

Dec 16

Theory Group

Dr. Harry Mynick presented the Plasma Physics Seminar at the MIT PFSC, entitled "Reducing Turbulent Transport in Stellarators and Tokamaks Through Shaping", on work done in collaboration with Neil Pomphrey, Pavlos Xanthopoulos (IPP-Greifswald), and Matt Lucia.

Mr. Zhirui Wang from RFX Padua visited PPPL from November 23 to December 6. He had broad discussions with Weixing Wang, Jong-Kyu Park and other theorists and experimentalists at the PPPL on kinetic effects on MHD instabilities. He presented a Theory Seminar entitled“Physical understanding of resistive wall modes in reversed field pinch plasmas and comparison with tokamak”. The physics of kinetic effects on the Resistive Wall Mode (RWM) stability arestudied and compared between Reversed Field Pinch (RFP) and Tokamak configurations. The toroidal kinetic hybrid stability code MARS-K is upgraded with an extensive energy analysis module for the study. In tokamak, the kinetic stabilization on the pressure driven RWM is provided by the precession resonance of thermal trapped particles, which requires very slow, or vanishing plasma rotation. In RFP, instead, the stabilization of RWM comes from the transit resonance of passing ions (i.e. the ion acoustic Landau damping) in high beta plasmas. The required rotation for kinetic stabilization is in the ion acoustic speed range or even higher (depending on the toroidal mode number and wall position). The trapped particles do not play a significant role in the kinetic stabilization in RFP plasmas. Detailed numerical analysis based on the perturbed potential energy components have been performed to obtain the understanding of kinetic physics in the two different systems. The physics of the active stabilization of feedback control on RWM in RFP is also studied based on the cylindrical MHD model. The study, via both analytical and numerical analysis, provides the physical understanding on the response of the unstable plasmas to the feedback control in RFPs. The linear solutions of time evolution of RWM instability under various feedback scenarios are given out.

A lecture paper "Excitation of Alfvén Modes by Energetic Particles in Magnetic Fusion" by N.N. Gorelenkov will soon appear in the proceedings of this year international ITER summer school held in Cadarache, France on June 2011. The paper deals with the most recent advances in the area of energetic particles spanning from the TFTR first DT campaign to the ITER planned experiments. Both theory and computation progresses are reviewed.

On November 25 - December 7, Leonid E. Zakharov visited Peking University, Graduate University of Chinese Academy of Sciences in Beijing and Institute of Plasma Physics in Hefei. He was shown three special divertor tiles on EAST tokamak installed following his suggestion in July for Hiro current measurements during vertical disruptions. Full information was presented to him on different kinds of damages of in-vessel components by disruptions: several broken water cooling pipes, tiles supports, and unexpected hair-thin holes in water pipes due to runaway
electrons from the previous experimental campaign on EAST. A schedule for flowing liquid lithium system for HT-7 and EAST was determined.

Leonid Zakharov gave a talk on "Understanding Disruption in Tokamaks" to ASIPP and to physics seminar at Peking University. He also described the scope of work on simulations of the free surface liquid lithium flow to the Prof. Ming-Jiu Ni group in College of Physical Sciences of Beijing Graduate University.

COMPUTATIONAL PLASMA PHYSICS GROUP

Stephane Ethier gave a two-part tutorial entitled "Introduction to Mixed Parallel Programming with MPI and OpenMP" at Princeton University as part of the technical mini-course series from the Princeton Institute for Computational Science and Engineering (PICSciE). Over two practiceoriented sessions, participants learned to exploit the power of parallel computing on a distributed memory system using message passing (MPI), as well as on a shared memory multicore node using OpenMP. Attendees were also introduced to mixing these two methods together in the same application, which is the favored approach for achieving high scalability on the largest computers currently available. This popular mini-course was attended by Princeton University's graduate students and research scientists from diverse science departments.

The NERSC computer time allocations were announced this week and PPPL scientists were awarded a total of 63,115,000 hours of supercomputer time for FY 2012. The awards went to the following principle investigators: S. Cohen, “FRC simulations with the LSP PIC code” (15,000 hr); R. Davidson, “Simulations of field-reversed configuration and other compact tori plasmas” (50,000 hr); R. Davidson, “Nonlinear delta-f particle simulation of collective effects for heavy ion fusion drivers and high intensity particle accelerators” (50,000 hrs); D. Mikkelsen,“Experimental tests of gyrokinetic simulations of microturbulence” (8,000,000 hrs); C.S. Chang,“Center for plasma edge simulation” (37,500,000 hrs); G. Fu, “Center for simulation of energetic particles” (4,500,000 hrs); S. Jardin, “3D extended MHD simulation of fusion plasmas” (4,000,000 hrs); and W. Lee, “Turbulent transport and multiscale gyrokinetic simulation” (9,000,000 hrs).

Dec 2

Theory Group

Dr. H. Strauss (HRS Fusion) and R. Paccagnella (RFX Padua) visited PPPL this week to consult with J. Breslau and others in the PPPL Theory Division on the modeling of the forces produced by plasma disruptions on the vessel wall, and projection of these forces to ITER using the M3D code. Dr. Strauss is the principal on a subcontract to address the ITER proposal “3D MHD simulation of VDEs for detailed evaluation of toroidal peaking factor (TPF) and associated
electro-magnetic load”. Dr. Paccagnella is the principal on a similar subcontract being handled by the European agency F4E. The focus of this work is to (1) utilize the massively parallel capabilities of M3D to greatly increase the resolution and time-scale separation of earlier work performed on a workstation version of M3D, and to (2) increase the physical realism of the halo model in M3D.

Roselyne Tchoua (ORNL) visited PPPL November 29-30 to learn about MDSplus and to plan extensions to the eSiMon simulation monitoring system (developed by the Center for Plasma Edge Simulation) to provide access to MDSplus data and tools. During her stay, she met with Eliot Feibush, Daren Stotler, Stan Kaye, and Devon Battaglia. She also had discussions on the ADIOS adaptable I/O system with Guo-Yong Fu, Stephane Ethier, and Jin Chen.

On November 22, Dr. Gregory von Nessi of the Australian National University presented a Theory Department Seminar entitled "Tokamak Equilibria Validation: Reconciling Theory and Observation via Bayesian Inference and BEAST". During his presentation, Dr von Nessi, gave an overview of a new technique for reconciling force-balance models with diagnostic observations via the statistical theory of Bayesian analysis. This method forms the backbone of a
new data analysis code called BEAST (Bayesian Equilibrium Analysis and Simulation Technique) and is based on refactoring the force-balance relation into two different forward models, each associated with a set of 'fractional' observations. These pairs of models and sets of observations are subsequently used in the Bayesian inference of the plasma equilibrium. By using a variant of the nested sampling algorithm, the evidence of the inferred posterior
distribution is calculated and provides a value which is sensitive to how much the inferred equilibrium differs from a force-balance solution. Results of the model were presented for discharges on the Mega-Ampere Spherical Tokamak (MAST), which are calculated using pickup coil, flux loop and Motional-Stark Effect (MSE) diagnostic observations. These results encompass inferences made utilizing both standard Grad-Shafranov (GS) and flow-modified GS
force-balance models. Finally, the interpretation of these results were discussed in the context of Bayesian model comparison and BEAST outputs.

On December 1, Dr, Timothy Stoltzfus-Dueck of the Max Planck Institute, Greifswald, Germany presented a Theory Department seminar entitled " Transport-driven Toroidal Rotation in the Tokamak Edge". The edge of H-mode tokamak plasmas without external momentum input almost always rotates toroidally in the co-current direction, which has prompted a theoretical search for non-diffusive momentum transport mechanisms. In contrast to these efforts, Dr.
Stoltzfus-Duek presented work that treats a model kinetic ion equation for the pedestal and SOL (Scrape Off Layer) containing only parallel free streaming, magnetic drifts, and spatially inhomogeneous but purely diffusive transport. The solution demonstrates that passing-ion orbits and spatially inhomogeneous diffusion interact to cause a variation of the orbit-averaged diffusivities that depends on the sign of the parallel velocity, typically resulting in preferential transport of counter-current ions. If the plasma at the boundary with the core is allowed to rotate toroidally to annihilate toroidal momentum transport, the resulting pedestal-top rotation reaches experimentally relevant values and exhibits several features in qualitative agreement with experiment. It is almost always in the co-current direction, with a rate that is proportional to pedestal-top ion temperature over poloidal field strength and the gradient scale length for the
intensity of potential fluctuations, thus inversely proportional to plasma current in accord with Rice scaling. It is independent of the toroidal velocity and its radial gradient, representing a residual stress. The given scaling implies co-current spin-up at the transition to H-mode, as ion temperature increases and the gradients steepen. Untested predictions of the model include a sensitivity of the rotation to the major-radial position of the X-point, with a more inboard Xpoint leading to stronger co-current rotation. Beyond intrinsic rotation predictions, comparison of
heat and momentum transport reveals that neutral beam injection must be significantly unbalanced in the counter-current direction to cause zero toroidal rotation at the pedestal top.

COMPUTATIONAL PLASMA PHYSICS GROUP

Both the tshare and pshare versions of TRANSP now have a new option for a shielded beam current calculation in the Monte Carlo neutral beam package NUBEAM. The model was recently derived and submitted for publication in Nuclear Fusion by Honda M., Kikuchi M., Azumi M. "Collisionality dependence on shielding factor of beam driven current" Eqs. 16 – 18 (submitted October 2011). The fit takes into account the collisionality dependence of the shielding factor at arbitrary aspect ratio. The new shielding calculation is invoked by setting NMCURB=4 in the TRANSP namelist. Initial preliminary studies by Budny of a D3D discharge shows that the NMCURB=4 setting gives a result for the beam-driven current that is intermediate between the values obtained with NMCURB=2 and NMCURB=3 (LinLiu).

The article “Review of Implicit Methods for the Magnetohydrodynamic Description of Magnetically Confined Plasmas” by S. C. Jardin is now available online: J. Comput. Phys. 231 (2012) pp. 822-838. This appears in a special JCP issue on Computational Plasma Physics. The article traces the development of modern algorithms for obtaining implicit solutions of the MHD equations to their roots in algorithms used in the 1960s and 1970s. These modern algorithms
make possible the simulation of slowly growing (compared to the Alfven time) global instabilities in today’s fusion experiments.

Nov 25

Theory Group

Many researchers from the Theory Department attended the Annual Meeting of the APS Division of Plasma Physics in Salt Lake City, UT Nov. 14-18. Allen Boozer presented a tutorial talk " Physics of Tokamak Disruption Simulations". Leonid Zakharov gave an invited talk " Understanding Disruptions in Tokamaks" and Andrey Zhmoginov, Theory department graduate student presented an invited talk "Alpha-channeling in Mirror Machines".

Dr. Jay Johnson gave a space plasma seminar at Dartmouth College about "Modeling Wave Propagation, Dissipation, and Mode Conversion in Space Plasmas" on November 8. While there, he collaborated with Richard Denton on hybrid simulations of whistler waves and Jim Labelle on mode conversion of Langmuir waves in ionospheric density gradients.

Nov 18

COMPUTATIONAL PLASMA PHYSICS GROUP

A TRANSP Users Group meeting was held at the annual APS-DPP meeting. The meeting discussed new options available in TRANSP such as the ability to invoke the RF routines in parallel, improved atomic physics packages in NUBEAM, a new nonlinear solver for TGLF, and a new free-boundary equilibrium capability. Approximately 25 users from the US and abroad attended the meeting. Presentations by the TRANSP team were followed by a discussion with the users regarding what new capabilities were desired and the best way of implementing them.

R. Andre, X. Yuan, and S. Jardin attended the DPP-APS meeting and made presentations on:"Status of TRANSP/PTRANSP", "A modular, parallel, multi-region predictive transport equation solver, installed and available in PTRANSP", and "Multiple Timescale Calculations of Sawteeth in Tokamak Plasmas" respectively.

Stephane Ethier attended the SC'11 conference on supercomputing held in Seattle, WA. He was co-author on three papers presented at the conference: "Gyrokinetic Toroidal Simulations on Leading Multi- and Manycore HPC Systems", K. Madduri et al., "ISABELA-QA: Query-driven Data Analytics over ISABELA-compressed Extreme-Scale Scientific Data",
S. Lakshminarasimhan et al., and "Multithreaded Global Address Space Communication Techniques for Gyrokinetic Fusion Applications on Ultra-Scale Platforms", R. Preissl et al.

Nov 4

THEORY GROUP
On November 3, Leonid E. Zhakarov gave a Theory Department seminar entitled "Scalable Flowing Liquid Lithium System (FLiLi) for Tokamaks" where he presented a practical system for the implementation of a flowing liquid lithium layer in tokamaks. The suggested scheme satisfies all requirements for FLiLi for development of the LiWall Fusion regime on existing devices and for its utilization in future tokamaks with a burning plasma. The thin LiLi layer is
insensitive to MHD effects and yet unknown j �� B forces, transparent to the heat flux an sufficient for plasma pumping. The FLiLi system is scalable in both poloidal and toroidal directions, compact, and has minimal necessary LiLi inventory in the machine.

COMPUTATIONAL PLASMA PHYSICS GROUP
Stephane Ethier gave an invited talk at the 2011 international Conference on Computational Physics held in Gatlinburg, Tenessee. This annual conference is dedicated to presenting an overview of the most recent developments and opportunities in computational physics across a broad range of topical areas and from around the world. Dr. Ethier's presentation was entitled"Computational advances in large-scale gyrokinetic particle-in-cell simulations of tokamak fusion devices".

Oct 28

Theory Group

Jay Johnson and the space physics group hosted a NASA LWS team meeting on "Integrating kinetic effects in global models" at PPPL on October 24-25. The group discussed collaborative projects considering comparative studies of reconnection in global systems using MHD, Hall- MHD, and fully kinetic models. The team also considered how to best include wave-particle interactions that control temperature anisotropies in global models of the solar wind and
planetary magnetospheres. Sixteen members of the team attended from University of New Hamsphire; University of Maryland; Johns Hopkins University, APL; NASA Goddard Space Flight Center; NASA Headquarters; Naval Research Laboratory; University of California, Berkeley; and PPPL.

On October 27, Prof Huishan Cai of the Department of Modern Physics, University of Science and Technology, Heifei, China presented a Theory Department seminar entitled "Influence of energetic ions on tearing modes". During his presentation, Professor Cai summarized the disruptive effects of tearing modes on tokamak plasmas. He discussed how fast ions generated by neutral beam injection are being investigated as a method of controlling these modes. In
contrast with the stability effects of trapped energetic ions on tearing modes, the effects of circulating energetic ions (CEI) on tearing modes depend on the toroidal circulating direction, and are closely related to the momentum of the energetic ions. CEI provide an additional source or sink of momentum to affect tearing modes. For co-circulating energetic ions (co-CEI), where the circulating direction is the same as the plasma toroidal current, tearing modes can be stabilized if the momentum of energetic ions is large enough. On the other hand, the growth of tearing modes can be enhanced by countercirculating energetic ions. Professor Cai additionally pointed out the possibility to suppress the island growth of neoclassical tearing modes by co- CEI. (Cai et al., PRL, 106, 075002, 2011).

Computational Plasma Physics Group

The M3D-C1 nonlinear simulation code has been used to calculate a sequence of sawtooth events in a tokamak of the shape of CMOD with a Spitzer resistivity profile and an analytic transport model. A loop voltage is applied to the boundary to sustain the current. After about 10^5 Alfven times the system becomes perfectly periodic with the sawtooth occurring at regular intervals. E. Feibush (CPPG) and A. Sanderson (U. Utah) have used the Visit visualization tool
to create a movie of the evolution of a pressure contour during one of these periods. The movie can be viewed at (requires QuickTime): https://w3.pppl.gov/cemm/Highlights/iso006b.mov. This sequence is being used as a baseline for which to study the effects of the transport model, 2- fluid terms, and kinetic effects on the sawtooth period, crash time, and other characteristics. Marina Gorelenkova held an informal design review on her project to include 3D geometry for
the halo neutrals in the NUBEAM Monte-Carlo neutral-beam simulations used in TRANSP. Participating were Sid Medley, Stan Kaye, Daren Stotler, Bill Heidbrink (remotely), and Robert Akers (remotely).

Oct 21

Computational Plasma Physics Group

Marina Gorelenkova hosted Dr.Stuart Loch (Assistant Professor at Auburn University) and Dr.Martin O'Mullane (Professor of Physics, University of Strathclyde) on October 17-18. Dr.Loch and Dr.O'Mullane had presentation on The Atomic Data and Analysis Structure (ADAS) activities and future development. Stuart and Martin worked with Daren Stotler, C.S Chang's group, Vlad Soukhanovskii, Daniel Clayton, Marina Gorelenkova, Brentley Stratton and
few other physicists.

Oct 14

Theory Group

Greg Hammett recently spend a couple of weeks visiting Oxford and Culham, interacting with Professor Alex Schekochihin, Professor Steve Cowley, Professor Bill Dorland, Michael Barnes, Ian Abel, Edmund Highcock, and others on gyrokinetic theory and simulations of plasma turbulence, including topics such as momentum transport and possible transitions from gyroBohm to Bohm scaling. (This trip was supported in part by Oxford University and
Schekochihin's grant from the Leverhulme Trust.). He gave a talk for a UK Fusion theory meeting at Culham/JET summarizing recent work with PPPL graduate students on GYRO simulations of ETG turbulence in NSTX and improved algorithms for transport solvers (J. L. Peterson), improved algorithms for gyrokinetic stellarator simulations in the GS2 code (J. Baumgaertel), and on simulations of density-gradient-driven turbulence in improvedconfinement
regimes expected with lithium walls (E. Granstedt).

Graduate student Luc Peterson published an article, "Bringing Star Power to Earth," on plasma
physics, fusion energy and his graduate research in Imagine Magazine, a publication of the Johns
Hopkins Center for Talented Youth that aims to provide gifted students in grades 7-12 insight,
perspectives and opportunities in fields they may wish to explore. Mr. Peterson's article appears
in the September/October 2011 issue on Physics and Astronomy.

A paper titled ""Radial Transport of Energetic Ions in the Presence of Trapped Electron Mode
Turbulence By a PPPL former visiting student Jugal Chowdhury (from Institute for Plasma
Research, India), and W. X. Wang, S. Ethier, J. Manickam and R. Ganesh (IPP, India) has been
accepted for publication in Physics of Plasmas. In this paper the nature of transport of hot ions is
studied in the presence of microturbulence generated by the trapped electron mode in a Tokamak
using massively parallel, first principle based global nonlinear gyrokinetic simulation, and with
the help of a passive tracer method. Passing and trapped hot ions are observed to exhibit inverse
and inverse square scaling with energy, while those with isotropic pitch distribution are found to
exhibit inverse dependence on energy. For all types of hot ions, namely, isotropic, passing and
trapped, the radial transport appears to be subdiffusive for the parameters considered.

On October 13, Dr. Jacob Bortnik of NJIT and UCLA presented a Theory Department seminar entitled "Some Open Problems Related to Wave Modeling in the Context of the RBSP Mission". More than 50 years after the discovery of the radiation belts, many of the key physical processes that control the structure and dynamics of these high energy electrons remain poorly understood, poorly quantified, or both. An emerging consensus recently stated that the long-held view of electron energization by inward radial diffusion was insufficient to account for the dynamics and
phase space density gradients that are typically observed, and that the operative process most likely involves a combination of several different types of plasma waves. During his presentation, Dr Bortnik briefly reviewed the role that waves play in controlling radiation beltdynamics. He described recent work on the origin and modeling of lasmaspheric hiss, that attributes the unexpected source of the hiss to electromagnetic chorus wave emissions which can propagate into the plasmasphere from tens of thousands of kilometers away (Bortnik et al., Nature, 2008). He highlighted a few of the upcoming missions and the key problems that that will be studied over the next few years with emphasis on the Radiation Belt Storm Probes mission, scheduled for launch in the Fall of 2012.

Computational Plasma Physics Group

Marina Gorelenkova attended the ADAS international workshop at Auburn University, Alabama
on October 5-8. The Atomic Data and Analysis Structure (ADAS) is an interconnected set of
computer codes and data collections for modeling the radiating properties of ions and atoms in
plasmas. She gave a talk "TRANSP/NUBEAM: atomic physics updates and issues" and had
fruitful discussions with Dr. M.O'Mullane(CCFE), Dr. O.Marchuk(Institut fuer Klimaforschung
Plasmaphysik IV Forschungszentrum Juelich). She also participated in the ADAS steering
committee meeting.

Stephane Ethier attended the DOE-ASCR Exascale Research Principal Investigators meeting in
Annapolis, MD. Dr. Ethier is a co-PI on an ASCR-funded Exascale Software Stack (X-Stack)
project led by Dr. Sam Williams of the Future Technologies Group at LBNL. The project
explores new optimizations for the particle-in-cell (PIC) algorithm as the computing technology
moves towards exascale.

S. Jardin attended the ITPA meeting on MHD and Disruptions in Padua, Italy and made a
presentation on "Nonlinear calculations of sawteeth and helical states in tokamaks". He also
reported on the progress for Working Group-10, "Halo current modeling", for which he is the
group head.

Oct 7

Theory Group

Daren Stotler attended the XSAMS (XML Schema for Atoms, Molecules and Solids) meeting at
the National Institute of Standards and Technology in Gaithersburg, Maryland on October 3-5.
The XSAMS schema promises to provide a single, comprehensive standard for exchanging
atomic physics and plasma-material interaction data between data producers and consumers.
Igor Kaganovich was invited to be a key note speaker at the TRR24 workshop on "Physics of
Complex Plasmas" in Germany (see file http://www.tr24.uni-greifswald.de/WS/flyer-TR24-
Workshop.pdf). He was also invited to visit University of Greifswald and The Leibniz Institute
for Plasma Science and Technology, (see http://www.inp-greifswald.de/webn.
nsf/index?OpenPage&Language=eng). He was given a tour of facilities and met with the
Directors.

Jin Chen attended XE6 Workshop at SNL September 27-28. The workshop features achieving
good scaling on future Multi-Core MPP system using Hybrid OpenMP-MPI programming.

Dr. Chris Chaston of the Space Sciences Laboratory at UC Berkeley visited PPPL on September
29-30. In addition to presenting the Colloquium on September 29, he was involved in research
discussions with the Space Physics Group. These discussions led to his collaboration with Jay
Johnson and Peter Damiano on a proposal to be submitted to the National Science Foundation,s
Geospace Environment Modeling program.

Sep 23

Theory Group
On September 22, Francesca Poli of PPPL gave a Theory Department seminar entitled, "Stability
and performance of ITER steady stated scenarios with ITBs". Non-inductive steady state
scenarios on ITER will need to operate with Internal Transport Barriers (ITBs) in order to reach
adequate fusion gain at typical currents of 9 MA. The large pressure gradients at the location of
the internal barrier are conducive to the development of ideal MHD instabilities that may limit
the plasma performance and lead to plasma disruptions. In her talk, Dr Poli presented five fully
non-inductive scenarios with various combinations of heating and current drive sources and
plasma currents in the range of 7 to 10 MA. For each configuration, the linear, ideal MHD
stability was analyzed for variations of the Greenwald fraction and of the pressure peaking factor
around the operating point aiming at defining an operational space for stable, steady state
operations at optimized performance. It was shown that lower hybrid heating is desirable to
maintain the safety factor profile above 1.5 and that these plasmas have better performance and
more favorable MHD stability properties. Operating with moderate ITBs at 2/3 of the minor
radius leads to safety factor profiles with minimum above 2, which significantly improves
stability and extends the operational space up to and above beta_N~3.

The US-Japan Joint Institute for Fusion Theory Workshop on ``Turbulent and Neoclassical Flow
Generation and Associated Transport Dynamics=94 was hosted by PPPL on September 9-10 at
Long Branch, New Jersey. A number of PPPL researchers, graduate students and visitors
participated the workshop. Choong-Seock Chang, Luca Guazzotto, Hantao Ji, John A. Krommes,
Seung-Hoe Ku, Jong-Kyo Park, Wayne Solomon, Weixing Wang, Jinlin Xie and Stewart J.
Zweben from PPPL presented talks at the workshop.

The paper "Magnetohydrodynamic Mechanism for Pedestal Formation" by L. Guazzotto and R.
Betti was published in Physical Review Letters.

On September 11-16, Leonid Zakharov visited Max Planck Institute of Plasma Physics in
Garching. He discussed experiments with high pressure gas injection on ASDEX Upgrade and
gave presentations on "Progress in understanding disruptions in tokamaks" and "Basics of
LiWall Fusion". Topics of discussion after the talks included the halo current based
interpretation of vertical instability and diffusion based confinement regime. The recent version
of the ASTRA code was obtained from IPP.

Sep 16

Theory Group
PPPL was the host for the International Conference on Numerical Simulation of Plasmas
(ICNSP) [http://icnsp2011.pppl.gov], which took place at the Jersey Shore (Long Branch, NJ),
September 7-9. There were 125 papers presented (45 orals and 80 posters) at this biennial
conference, which covered six disciplines: Magnetic & Inertial Fusion Plasmas, Space &
Astrophysical Plasmas High Intensity Beams & Laser Plasma Interactions, Low-temperature
Plasmas, Numerical Methods & High Performance Computing, and Scientific Visualization.
There were 18 papers presented at the conference from PPPL (4 orals and 14 posters) with many
student observers from the Lab as well. W. W. Lee gave the John Dawson Prize presentation,
entitled: “Surfing the Waves: a Particle Pusher's Personal Perspective”. The organizers of the
conference were Riccardo Betti, Stephane Ethier, Wei-li Lee, and Ed Startsev. Jennifer Jones,
Barbara Sarfaty - Carl Scimeca provided the help before and during the conference.

Eun-Hwa Kim attended a meeting in Fukuoka, Japan on "Physical Processes in Non-Uniform
Finite Magnetospheric Systems---50 years of Tamao's Resonant Mode Coupling Theory" from
Sept 12-15. She presented an invited talk on Electromagnetic Ion Cyclotron Waves at Mercury.

Guoyong Fu and Nikolai Gorelenkov attended The IAEA Technical Meeting on Energetic
Particles in Magnetic Confinement Systems held on September 7-10 and the ITPA Energetic
Particles Topical Group Meeting held on Sept. 12-13 in Austin, Texas. Dr. Fu gave an invited
talk titled "Nonlinear Theory and Simulation of Energetic Particle-induced Qeodesic Acoustic
Mode" at the first meeting. He also gave a talk at the second meeting titled "M3D-K simulations
of beam-driven Alfven eigenmodes in DIII-D plasmas".

Seung-Hoe Ku gave an invited talk , entitled “Utilizing extreme scale HPC for full-f gyrokinetic
particle simulation in realistic diverted tokamak geometry ” at 22nd ICNSP at Long Branch, NJ.

Sep 9

Theory Group
On August 25, Leonid A. Zakharov gave the PPPL Theory Department seminar on the "MHD
energy principle for tokamaks", where he showed that the MHD energy principle can be
expressed in terms of magnetic field perturbations. The advantages of this representation are: (a)
a direct interface with the equilibrium codes and mitigation of the convergence problem, and (b)
the applicability to perturbed equilibrium studies for understanding the thermal quench and
schemes of runaway electron suppression during disruptions. Also, robust stability of the plasma
edge in the LiWF regime (predicted in 2005 and later confirmed by many experiments) is
evident from the new form of energy principle.

On September 6, Professor Zhengming Sheng of the Laboratory for Laser Plasmas and
Department of Physics, Shanghai Jaiotong University presented a special Theory department
seminar entitled "Charged particle acceleration and radiation from ultra-intense laser-plasma
nteractions". During his presentation, Professor Sheng summarized the basic concepts and
introduced some recent progress of his group on laser plasma acceleration and radiation. The
interaction of ultra-short intense laser pulses with plasmas has been a subject of significant
interest in the last 20 years with the invention of the CPA (Chirped Pulse Amplification) laser
technology. Such interaction shows highly-nonlinear features, where both kinetic and collective
effects for particles are important and the relativistic nonlinearity becomes significant. In
particular, the large laser field itself and the induced electrostatic fields (waves) are so high that
both electrons and ions can be accelerated quickly under different conditions. For example, it
becomes possible to accelerate charged particles to GeV energies over a short distance of a few
centimeters. Meanwhile, the interaction of such lasers with different targets can result in
electromagnetic radiation ranging from THz to gamma-rays. These particle beams and radiation
sources are collimated and are suitable for a range of applications.

COMPUTATIONAL PLASMA PHYSICS GROUP
Several staff attended the International Conference on the Numerical Simulation of Plasmas this
week in Long Branch, NJ. S. Ethier presented a poster on "Optimizing Gyrokinetic PIC Codes
for Extreme Scale Computing", and S. Jardin gave a talk on “Multiple Timescale Calculations of
Sawteeth and other Global Macroscopic Dynamics of Tokamaks”.

Sep 2

Theory Group
Dr. Wei-li Lee, has been awarded the 2011 John Dawson Prize for Numerical Simulation of
Plasmas. The award reflects on seminal contributions of Dr. Lee to the area of gyrokinetic
simulations in plasma physics. An award ceremony presenting the prize will be held during the
International Conference on Numerical Simulation of Plasmas in Long Branch, New Jersey,
September 7-9.

Riccardo Betti gave a review invited talk on hydrodynamic instabilities at the Turbulent Mixing
and Beyond meeting. The meeting took place at the Abdus Salam International Center for
Theoretical Physics (ICTP), Trieste, Italy, on August 21-28. Betti's invited talk reviewed the
theory of the Rayleigh-Taylor instability in the presence of mass ablation.

A paper titled "Combined Ideal and Kinetic Effects on Reversed Shear Alfven Eigenmodes" by
N.N. Gorelenkov, R. Nazikian and G. Kramer has been accepted for publication in Physics of
Plasmas. In this paper the authors elaborate on methods to find the solutions of the so called
RSAE (reversed shear Alfven eigenmodes) modes and its physics. These modes are known to be
among the most dangerous for fast ion confinement. The theory of their existence in a special
case when the magnetic shear profile is reversed and the value of q min is slightly above m/n
ratio is developed. With the help of MHD code NOVA the paper explains the relations of the
MHD theory to the oscillations, observed in experiments and modes obtained using kinetic
approximations.

The paper entitled "ULF Wave Absorption at Mercury" by E.-H. Kim, J. Johnson and K.-D. Lee
has been published to Geophysical Research Letter.
(see http://www.agu.org/pubs/crossref/2011/2011GL048621.shtml

COMPUTATIONAL PLASMA PHYSICS GROUP

A modular, parallel, multi-regional, implicit transport equation solver (PT_SOLVER) built over
the Plasma State and other publicly available (NTCC) libraries has been developed. It integrates
the highly nonlinear time-dependent equations for ion, and electron temperatures, and angular
momentum with implicit Newton iteration methods. The user controls choice of transport models
attached to the solver, with neoclassical models of Chang-Hinlton, and NCLASS and/or
turbulent models available. Implemented turbulent transport models include GLF23, TGLF and
ETG models. For the more expensive transport coefficient models such as TGLF, a multi-level,
communicator splitting method is used to parallelize the computation of transport coefficients
using MPI, which allows the code to run on parallel supercomputers. Once the verification
activities are concluded, this module (PT_SOLVER) will be installed as the primary transport
equation solver in PTRANSP.
PLASMA

Aug 26th

COMPUTATIONAL PLASMA PHYSICS GROUP

The MODPALEO model, containing the PALEO subroutine which calculates the electron thermal transport according to the Paleoclassical transport model of Jim Callen, et al. was added to the NTCC Modules Library, ttp://w3.pppl.gov/ntcc/. Authors: Lixiang Luo, Tariq Rafiq, Arnold Kritz, Lehigh University It was observed this summer, as a result of a careful benchmarking exercise by a student Intern, that the linear growth rate of a resistive instability calculated by M3D-C1 when using a full MHD model on an adapted mesh could be slightly different when different numbers of processors were used in the calculation. This discrepancy did not occur for a uniform mesh, or for any of the reduced MHD models. This week, we have identified that significant differences are present in certain velocity variables after the first iteration for the full MHD model when using an adapted mesh. Using this new data, we are now working with our meshing partners at the SCOREC center at RPI to identify and correct the source of this discrepancy.

Aug 19

Theory Group

Wenjun Deng of the Department of Physics and Astronomy at the University of California Irvine, gave a Theory Department seminar entitled "Gyrokinetic particle simulations of reversed shear Alfven eigenmode in DIII-D tokamak". In his talk, Mr. Deng presented the electromagnetic gyrokinetic simulation model used in the gyrokinetic toroidal code (GTC) which reduces to ideal MHD theory in the long wavelength limit. He summarized the application of the
code to the study of the reversed shear Alfven eigenmode (RSAE) (Deng et al, Physics of Plasmas, 2010). The damping rates measured from the antenna excitation and from the initial perturbation simulation were found to agree very well. The RSAE excited by fast ions showed an exponential growth. The finite Larmor radius effects of the fast ions were found to significantly reduce the growth rate. With kinetic thermal ions and electron pressure, the mode frequency
increased due to the elevation of the Alfven continuum by the geodesic compressibility. For simple geometries with concentric-circular flux surfaces and large aspect ratio, the non-existance of the RSAE when the parallel equilibrium current is taken into account (as derived from ideal MHD theory) was confirmed. In a realistic geometry, the RSAE can exist even in the presence of equilibrium current due to toroidal coupling and other geometric effects. Simulations of the
DIII-D discharge 142111 near 750ms showed that the RSAE exists even without fast ion drive. The mode structure, frequency, and growth rate obtained from GTC simulations are close to those given by GYRO and TAEFL simulations.

Jay Johnson led an international team meeting on "Plasma Entry and Transport in the Plasma Sheet" Aug 8-12 at the International Space Science Institute in Bern, Switzerland. The team discussed observational constraints on plasma entry processes at the Earth's magnetopause such as entropy, dawn-dusk asymmetries, and conservation of the electron to ion temperature ratio even when particles are heated by an order or magnitude. The international team has also been invited to write a review paper on plasma entry and transport processes. The paper entitled ULF wave absorption at Mercury by Kim, Johnson, and Lee, has been accepted to Geophysical Research Letter. The abstract is: The field line resonance at Mercury is expected to occur when the ion-ion hybrid (IIH) and/or Alfven resonance conditions are satisfied. However, the relative efficiency of wave energy absorption at these resonances has not been studied in the context of Mercurys magnetosphere. To understand the efficiency of wave absorption, we evaluate absorption coefficients at the IIH and Alfven resonances for variable concentrations of sodium and azimuthal and field-aligned wave numbers in 1D multi-ion plasmas. The results show that wave absorption is much more efficient at the IIH resonance than at the Alfven resonance at Mercury. Our results suggest that the mode conversion efficiency is sensitive to the azimuthal and field aligned wave numbers as well as heavy ion concentration ratio. Therefore, the radial profile of field-line resonances at Mercury can exhibit complex, discontinuous structure.

The paper "Magnetohydrodynamic mechanism for pedestal formation, L. Guazzotto and R. Betti" was accepted for publication in PRL.

COMPUTATIONAL PLASMA PHYSICS GROUP

Checkpointing (for restarting a calculation) is becoming more and more critical in large scale computation. Traditionally one file per processor is written/read, but this is not efficient or even allowable when thousands of processes are used since every process competes for IO and there are limits on the maximum number of files allowable. ADIOS (Adaptable IO System (ADIOS) has been developed at ORNL for this purpose. It efficiently coordinates all MPI processes to
write all the checkpoint data into a single file. It also provides flexibility for processing the checkpointed data outside of the running simulation. Jin Chen has now integrated ADIOS-1.3 into M3DC1. This resulted in significant improvements in convenience and performance, particularly for large jobs on Hopper at NERSC. Arnold Kritz and three others from the Lehigh U. predictive transport group visited PPPL for discussions related to simulations performed with PTRANSP and the SWIM IPS. A number of issues that arose in benchmarking the two codes were resolved. A planning session was held regarding extending the SWIM IPS to include toroidal momentum transport and extending PTRANSP to include more general particle transport. These extensions will aid in performing more extensive benchmark calculations in the future, as well as in performing more realistic simulations of ITER and other devices.

Bryan Weinstein and Jeffrey Liu completed their Science Undergraduate Laboratory Internships under the direction of Eliot Feibush and Harry Mynick. The students developed a graphical user interface to the Plasma Apprentice formulary developed by Harry Mynick. The user interface is written in Java and connects to Mathematica for expression evaluation.
Parag Srivastava completed his Science Education Internship under the direction of Eliot Feibush and Gerrit Kramer. Parag developed visualizations and analysis of the results from the SPIRAL code. The user interface enables selection of trapped or passing particles for graphing any pair of the 60 parameters computed by the code. Christina Efthimion completed her PPPL Summer Internship under the direction of Stephane Ethier and William Tang. The project consisted in exploring new algorithms to set up the poloidal grid in PPPL's Gyrokinetic Tokamak Simulation code (GTS).

A video conference was held between Jim Conboy (JET), John Wright (MIT, PSFC) and the PPPL TRANSP developers regarding the status and future developmental path for both serial and parallel TORIC within TRANSP. A TRANSP specific version of TORIC is kept in a branch of the TORIC SVN repository at IPP in Garching. The same source code is also kept in a directory of the TRANSP SVN repository at PPPL. Plans were made for Jim Conboy to initially synchronize the two versions of the TRANSP/TORIC source code, incorporating his own modifications as well, and for the PPPL TRANSP developers to have direct access to the IPP SVN repo in the future so that they could keep their SVN repo version in sync with the IPP version. Several bug fixes were also discussed. Stephane Ethier attended the five-day summer school on "Proven Algorithmic Techniques for Many-core Processors" organized by the Virtual School of Computational Science and Engineering (VSCSE). NSF-funded VSCSE is a national virtual organization whose goal is to develop and deliver a computational science curriculum that accelerates the ability of faculty, staff, and students to use emerging computational resources to advance science and engineering. The lectures were delivered to several geographically-distributed sites via audio-video broadcast. Princeton University hosted one of those sites at the Lewis library through PICSciE. The highly technical lectures focused on algorithms and methods to efficiently program and use GPU processors.

Aug 5

THEORY
Leonid Zakharov visited the People’s Republic of China (Wuhan, Hefei, Chengdu, Beijing) from
July 10-27. In HUST (Huazhong University of Science and Technology, Wuhan) he discussed
the possibility of installation of ITER-mimicking tiles on the local J-TEXT device for measuring
Hiro currents during plasma disruptions. Zakharov gave a presentation on "Progress in
Understanding and Simulations of Disruptions in Tokamaks". In ASIPP (Institute of Plasma
Physics of Chinese Academy of Sciences, Hefei) he proposed modification of 6 divertor target
tiles for the first measurement of Hiro currents during Vertical Disruption Events in tokamaks.
At the HT-7 data analysis workshop Zakharov gave a presentation entitled "Design Guidance for
Flowing Lithium Systems in Tokamaks", where for the first time a practical Flowing Liquid
Lithium System (developed with Peter Yushmanov) was disclosed. In SWIP (South West
Institute of Physics, Chengdu), Zakharov discussed the possibility or making a low voltage (10-
15 kV)/high current (15-20A)) Neutral Beam Source for LTX device in PPPL. Zakharov gave a
presentation to the SWIP Center for Fusion Studies on "Progress in Understanding Disruptions in
Tokamaks". Waveforms of magnetic measurements from HL-2A device in SWIP were providedto the visitor. In Peking University (Beijing), Zakharov discussed with Professor Ming-Jiu Ni the MHD physics of the liquid metal blanket of tokamaks. Finally, Zakharov made a presentation at the Fusion Simulation Center of Peking University on "Lithium Wall Fusion Regime and the Next Step in Chinese Magnetic Fusion".

July 22

Theory Group

Eliot Feibush and  Bill Tang attended the DOE Computational Science Graduate Fellowship Conference in Arlington, VA July 20 - 22. They presented PPPL as a DOE site where fellows can perform a laboratory practicum for students to broaden their experience beyond thesis search.

W. W. Lee gave a keynote talk in the Science Session of the TeraGrid Conference on Extreme Digital Discovery at Salt Lake City, Utah, July 18-21. The title of Lee's talk was "Kinetic Simulation of Magnetic Fusion Plasmas on High Performance Computing Platforms." The
TeraGrid project, after ten years of service to the national science and engineering community, will be succeeded by a new National Science Foundation program called XSEDE -- the Extreme Science and Engineering Digital Environment. A closer collaboration between PPPL and XSEDE in the future is to be explored.

Prof. Dmitri Uzdensky of the University of Colorado Boulder gave a Theory Department seminar entitled "Plasmoid-dominated magnetic reconnection". During his presentation, Prof. Uzdensky discussed how analytical and numerical results have illustrated that long thin
Current layers, the hallmark of the classical Sweet-Parker (SP) resistive MHD model are violently unstable  to secondary tearing instability when the systems global Lundquist number S exceeds a critical value Sc ~ 104. As a result, these layers are replaced by hierarchical multiscale chains of numerous secondary islands (plasmoids; flux ropes in 3D). He presented a new conceptual analytical model of reconnection in the plasmoid-dominated regime, as well as the related recent numerical work. The model analyses the interplay of the plasmoids' production,
growth, ejection, and coalescence with the key results that  (1) the global effective reconnection rate is fast, of order 0.01 of the Alfven rate, independent of S for S > Sc ; (2) the distribution of plasmoid fluxes is the inverse-square law; (3) there is a finite probability of emergence of anomalously large "monster" plasmoids that can disrupt the chain. Prof Uzdensky also described how the plasmoid hierarchy hastens the transition from collisional to collisionless limits, blurring the distinction between the two regimes.

July 8 & 15

Theory Group

R. Betti gave a theory seminar at General Atomics on "Two-Dimensional Magnetohydrodynamic Simulations of Poloidal Flows in Tokamaks and Pedestal Formation." He also taught three lectures on implosion hydrodynamics at the Summer School in high-energy density physics organized by the Fusion Science Center for Extreme States of Matter at the University of California San Diego.

CPPG Group

The subroutine "innerc" has been added to the NTCC modules library. This routine implements the solution of the inner region equations in the asymptotic matching solution for tearing modes as formulated in the 1970s by Glasser, Greene, and Johnson. The highly accurate routine follows the procedure given in Glasser, Jardin, Tesauro Phys. Fluids 27 (5), May 1984.This routine, originally written in the early 1980's has been resurrected and modernized by a graduate intern, Thomas Scaffidi, from the Ecole Normale Superieure, Paris, France. It has recently been verified both by comparing with published solutions, and also by comparing with linear tearing mode solutions from the M3D-C1 code in simplified geometry. A paper describing this benchmarking is in preparation.

July 1

Theory Group

Drs. N. Gorelenkov and N. Fisch gave lectures to the international ITER summer school held in Aux-en-Provence, France from June 20-24 to an audience of young researches and students. Two lectures focused on the recent achievements in the area of fast ion physics, their confinement, Alfven instability studies and the possibility to use them to make more efficient fusion reactor. Several other lecturers from UK, Europe, US, India and Russia were also present.

June 24

Theory Group

Professor Bruce Scott of the Max-Planck-IPP (and EURATOM Associates) presented two Theory Department seminars. In the first, entitled "Energy Consistency and Momentum Conservation in Gyrokinetics", Professor Scott briefly reviewed the modern theory of gyrokinetics where the dynamics is described in terms of a Lagrangian with canonical structure -Dependent field variables appear only in the time component. The gyrokinetic and associated field equations are derived from the same Lagrangian by varying gyrocenter coordinate positions and field amplitudes. Energetic consistency follows from the general symmetry implied by the support of the equations by the Lagrangian. The Noether theorem indicates what is conserved and both local and global conservation forms were given. The conversion from canonical to plasma momentum uses the charge conservation equation which follows from continuity. Professor Scott emphasized the specific role of the time-dependent polarization current and explained that in the plasma momentum form, the original motivation for considering higher-order drifts is removed exactly. The correspondence to the MHD momentum conservation law was also given. The results are automatically valid for any ordering since all ordering is done in the construction of the Lagrangian in the beginning and none is done thereafter.
 
In the second presentation "entitled "Axisymmetric Electromagnetic Gyrokinetic Computation" Prof. Scott discussed simulating the tokamak edge layer using a total-f electromagnetic gyrokinetic field theory model (FEFI) based on an "MHD Lagrangian" designed to capture shear-Alfven (reduced) MHD and equilibrium flows on a banana width scale. Dr. Scott illustrated that without collisions, the neoclassical control case determines the necessary resolution to conserve energy/momentum and also produce negligible transport. In this context "neoclassical" refers to the time scale assumption of flows in 2D divergence balance, nearly but not exactly incompressible as variations within the flux surface of all quantities is allowed. With collisions the bootstrap current is found self consistently. Professor Scott also described ongoing attempts to turn FEFI into a comprehensive equilibrium and turbulence model.
 
 
CPPG GROUP

The 2F 3D nonlinear MHD code M3D-C1 is now operational on the NERSC peta-flop system Hopper; a Cray XE6 with 153,216 computer cores. Initial benchmark run was performed with 96 processors to reproduce nonlinear sawtooth calculations performed on the PPPL computer, STIX, using a time-step dt=20 Alfven transit times. We then performed preliminary parallel scaling studies keeping the time-step fixed, but increasing both the number of finite-element nodes and the number of compute cores by a factor of 40 and the wall-clock time increased by only 40% to get to the same point in time.  We had discussions with the PETSc group, the SCOREC team, and the SuperLU-dist group this week on analyzing the performance data, and expect to improve the parallel scaling performance and extend it to 10's of thousands of compute cores in the coming weeks.   The higher resolution made possible by Hopper is enabling these sawtooth studies to be performed in a parameter regime that was previously inaccessible.

June 17

Theory Group

Dr. Gabriel Plunk of the University of Maryland gave a Theory Department seminar entitled "The generalized Fjortoft constraint for two dimensional gyrokinetics". In his presentation, Dr. Plunk initially summarized the result of the 1953 paper by Fjortoft, which established the simple argument for understanding the upscale transfer of energy that occurs in incompressible two-dimensional Navier-Stokes turbulence. This argument starts from the simple observation that the existence of a second quadratic invariant, "enstrophy," places a constraint on how energy can be redistributed spectrally under nonlinear interaction. This sets the stage for the famous "inverse cascade" of fluid turbulence. He then went on to explain how the gyrokinetic equation also conserves two quadratic quantities  under nonlinear interactions. To isolate the effects of this property, he focused on the simple limit: the electrostatic two-dimensional gyrokinetic equation, with a homogeneous equilibrium. He showed that the relationship between the two invariants establishes a simple constraint on spectral transfer, in analogy to that of 2D Navier-Stokes turbulence. However, in the gyrokinetic case, the inverse transfer of energy (appropriately defined) is only one possible outcome. If suitably driven, a constrained dual forward cascade can also occur, revealing the possibility that large scales can be "forcibly" damped by fluctuations at much smaller scales. Dr. Plunk also discussed the consequences for zonal flows and other features of realistic saturated turbulence.

W. W. Lee gave a seminar, entitled "Recent Development in Gyrokinetic Theory and Simulation," at the Department of Physics and Astronomy, University of California at Irvine, last week. In his talk, three related topics in gyrokinetic theory and simulation were addressed: 1) The recent concern related to the validity of the gyrokinetic Vlasov--Poisson equations used in most of the existing gyrokinetic codes in the community, which contained only the terms associated with the lowest order in gyrokinetic ordering parameter Ξ΅; 2) A discussion on the newly developed electromagnetic gyrokinetic formulation for finite-Ξ² plasmas; 3) A demonstration of the new two-wieght PIC scheme, which could enable us to carry out simulations using first the Ξ´f scheme to minimize the intrinsic noise in PIC codes at the start of the simulation and gradually evolve into a full-F code in nonlinearly saturated stage of the simulation.

R. Betti served on the DOE Review Committee of the National Ignition Campaign chaired by the Undersecretary for Science Dr. S. Koonin. The two-day long review assessed the progress of the ICF program toward demonstrating thermonuclear ignition on the National Ignition Facility at Lawrence Livermore National Laboratory.

June 3

Theory Group

Professor Amita Das of the Institute for Plasma Research, Gujarat, India presented a Theory Department seminar entitled "Electron - Magnetohydrodynamic (EMHD) studies in an inhomogeneous plasma". The advent of high intensity femto-second laser pulses and the availability of fast diagnostic techniques have now made it possible that a plasma can be triggered at very fast electron time scales and its response at these fast time scales can now also be watched in considerable detail. These studies have therefore opened up an entirely new regime of studies of plasma phenomena. During her presentation, Prof. Das highlighted how these phenomena can be suitably depicted by the Electron Magnetohydrodynamics (EMHD)  model, which is a simplified description for magnetized electron fluid  in the presence of  neutralizing stationary background ions. The model has been recently generalized (G-EMHD) to incorporate plasma density inhomogeneity and the salient features of his advanced version were discussed. Simulations using the G-EMHD model demonstrate the possibility of maneuvering  the propagation and evolution of  an electron current pulse  in the plasma by an appropriate choice of plasma density inhomogeneity. Specifically, regions of  high plasma density can trap electron current pulses, thereby providing a basis for a plasma tweezer like device  for electron currents. The plasma inhomogeneity also helps in collimating and guiding the  electron current pulses. Furthermore, the location of electron energy deposition  in plasma medium can be maneuvered by an appropriate choice of the plasma density inhomogeneity and the relevance of this to the fast ignition scheme for laser fusion was discussed. A variety of PIC simulations, as well as several experimental observations, provide sufficient evidence in support of the mechanism presented based on the G-EMHD studies.  

May 27

Theory Group

Riccardo Betti gave the Physics Colloquium at the University of California Los Angeles. His talk entitled "Conventional and Advanced Thermonuclear Ignition in Inertial Confinement Fusion" covered the essential physics of inertial fusion and the high-gain schemes for inertial fusion energy.
 
Professor Allen Boozer of Columbia University presented a Theory Department seminar entitled "Reconnection in Naturally Arising Magnetic Fields". During his presentation, Professor Boozer emphasized the distinction between reconnection processes in laboratory plasmas and  naturally occurring systems. He discussed how the fundamental questions of reconnection theory are why is the observed dissipation of the magnetic field sometimes enhanced by ten or twenty orders of magnitude from its characteristic value and what is the trigger for the enhanced dissipation. Electrodynamics provides an answer-the exponential increase in the separation between neighboring magnetic field lines. In toroidal magnetic configurations with small islands this exponentiation occurs only near the X-points of the islands, though it occurs almost everywhere in regions of stochastic field lines. In naturally occurring magnetic structures X-points and islands are ill defined, but neighboring magnetic field lines generically change their separation by a factor of exp(S) within a segment of the magnetic system of length L. The exponentiation S is easily calculated field-line-by-field-line in any bounded region in which a magnetic field is known-for example in a numerical simulation-and S evolves even when the magnetic evolution is dissipationless. The exponential increase in the separation implies an exponential increase in derivatives across the magnetic field of quantities that vary slowly along the magnetic field lines. Some dissipative mechanisms are proportional to these derivatives squared, so the dissipation becomes exponentially large, proportional to exp(2S). Reconnection is naturally triggered in regions in which S ~ 20.
 

May 20~

Professor Alan Brizard of Saint Michael's College gave a Theory Department seminar entitled "On the Noether Derivation of the Gyrokinetic Energy-Momentum Conservation Laws". The variational formulation of the nonlinear gyrokinetic theory is based on a Lagrangian density that depends on the gyrocenter Vlasov distribution and the perturbed electric and magnetic fields.  According to the Noether Theorem, the energy-momentum conservation laws are associated with symmetries of the gyrokinetic Lagrangian density with respect to time and space translations. During his seminar Prof. Brizard presented a tutorial on the Noether method and its applications in plasma physics illustrating the derivation of the exact energy-momentum conservation laws for the gyrokinetic Vlasov-Poisson equations with particular reference to the axisymmetric tokamak geometry (Brizard and Tronko, Physics of Plasmas, submitted 2011). Future work will extend this analysis to the electromagnetic case.

On May 15-19, 2011 Dr. Leonid E. Zakharov attended the 15th International Conference on Emerging Nuclear Energy Systems, San Francisco, CA. He gave a talk on "Fission-Fusion Research Facility
(FFRF) as a Practical Step Toward Hybrids" where potential burning plasma regimes for the next step device in China were described.

Dr. Harry Mynick presented a talk for the weekly 3D Task Force Meeting entitled "Transport simulations for the DIII-D double-null experiments". The talk is focused on transport assessments and was done in collaboration with Neil Pomphrey.

Drs. Hank Strauss of HRS Fusion and Roberto Paccagnella of Consorzio RFX concluded a two-week visit to PPPL in which they collaborated with Josh Breslau and Steve Jardin on expanding the ability of the free-boundary version of the M3D code to perform disruption studies for ITER.

May 13~

Theory Group

May 6~

Theory Group

April 30~

Theory Group

Dr. Jack Berkery from Columbia University/PPPL, gave a seminar entitled, "Resistive Wall Mode Kinetic Stability Theory Advancements for Refined Comparison With Experiments" on April 28.  He presented the theory of RWM stabilization that invokes kinetic effects as energy dissipation mechanisms. In particular, rotational resonances between the mode, the plasma rotation, and thermal particle motions have been shown to explain RWM instability at intermediate rotation in NSTX. The inclusion of energetic particle effects refined the qualitative agreement between computed RWM marginal stability points and experimental results. He showed that it is important to take into account the anisotropy of neutral beam injected energetic ions to correctly account for their stabilizing effects. He also outlined the role of collisions in both dissipating the mode energy and also in damping the resonant kinetic effects. In future devices with lower collisionality, it is important to avoid unfavorable plasma rotation profiles through rotation control or active mode control.
 
 Harry Mynick presented the Plasma Physics Seminar at the University of Wisconsin,  entitled "Reducing Turbulent Transport in Toroidal configurations through shaping", on work done in collaboration with Neil Pomphrey and Pavlos Xanthopoulos (IPP-Greifswald).

COMPUTATIONAL PLASMA PHYSICS GROUP

The PPPL CPPG supports TRANSP run production on PPPL servers, for tokamak simulation. TRANSP jobs can be submitted from numerous sites around the world; there are currently over 115 registered users. The PPPL Linux server cluster has recently seen high levels of TRANSP/PTRANSP simulation activity. On many occasions, the total count of concurrent runs has exceeded 60, which includes many low-N MPI simulations. A recent snapshot (Thursday April 28) showed 61 runs, including 1 with 16 processors, 2 with 8 processors, 46 with 4 processors, and 12 serial jobs for a total of 228 processors concurrently engaged in TRANSP/PTRANSP production computing.
 
Stephane Ethier completed a two-week visit to IPP in Garching, Germany, to work with Dr. Bruce Scott on incorporating a series of specialized diagnostics into the global gyrokinetic PIC code GTC-P. The diagnostics, developed by Dr. Scott, are being used in gyrofluid and continuum gyrokinetic codes for studying the macroscopic characteristics of plasma microturbulence under various conditions. The implementation of those diagnostics into GTC-P will allow for direct code comparison and validation with experiments. This work is being carried out as part of a PPPL INCITE project

April 15th~

Theory Group

April 8th~

Theory Group

Dr. A. Reiman gave a talk at the DIII-D task force meeting on March 30 describing Theory Department plans for applying stellarator codes to analysis of 3D DIII-D ELM suppression experiments. The STELLOPT code will evaluate a number of properties, including 3D ballooning stability and macroscopic stability. This will make use of the TERPSICHORE code, and will be done by N. Pomphrey in collaboration with A. Turnbull at GA and A. Cooper at CRPP, Lausanne, Switzerland. The ORBIT Monte Carlo code, developed by Dr. R. White,   will be used by Dr. Pomphrey and Dr. H. Mynick to calculate 3D effects on neoclassical ion confinement.

The GENE gyrokinetic code will be used by Dr. Mynick to obtain information on the effect of the 3D fields on turbulent transport. The PIES code, which can calculate equilibria with islands and stochastic regions, will be applied by S. Lazerson, A. Reiman and D. Monticello.

A paper titled "Onset and Saturation of a Non-Resonant Internal Mode in NSTX and Implications for AT modes in ITER" by J.A. Breslau, M.S. Chance, J. Chen, G.Y. Fu, S. Gerhardt, N. Gorelenkov, S.C. Jardin, and J. Manickam has been accepted for publication in Nuclear Fusion. The article describes numerical modeling work performed with M3D and Several other PPPL codes that helped to elucidate the origin of an unexplained tearing mode that had been observed in NSTX. It arises from a non-resonant ideal MHD mode with toroidal mode number n=1 that is destabilized in the low shear equilibrium as the central safety factor approaches one.  Nonlinear simulation demonstrates that this mode can lead to saturated 2,1 magnetic islands, as seen in the experiment.

April1st~

March 25th~

March 18th~

March 11th~

Professor Don Batchelor of Oak Ridge National Lab presented a special Theory Department seminar entitled "Using the SWIM Integrated Plasma Simulator for Multi-physics Fusion". The SWIM (Simulation of Wave Interaction with Magnetohydrodynamics) center has the scientific objectives of improving the understanding of interactions that both RF wave and particle sources have on extended-MHD phenomena, and improving the capability for predicting and optimizing the performance of burning plasmas. The center has built an end-to-end computational system that allows physics codes to be able to function together in a parallel environment and connects them to utility software components and data management systems.  This framework has been used to couple together state-of-the-art fusion energy codes to produce a unique multi-physics simulation capability.  In his talk, Professor Batchelor presented a physicist's overview of the Integrated Plasma Simulator (IPS)  and described applications.  For example, the IPS is being employed to support ITER with operational scenario studies.  A computational approach to coupling MHD with RF has also been developed, and initial numerical studies of RF effects on tearing modes with reduced models have been completed. Additionally, the IPS is being used to investigate the possibility of parallelization in the time domain of plasma turbulence calculations. The wide range of simulation work-flows that can be composed using the IPS was emphasized.
 
Dr. Walter Guttenfelder of PPPL presented the regular Theory Department seminar entitled, "Electromagnetic Transport From Microtearing Mode Turbulence in NSTX". In this presentation, Dr. Guttenfelder summarized  first-of-a-kind non-linear gyrokinetic simulations of microtearing mode turbulence using the GYRO simulation code. The physically comprehensive simulations (including kinetic ions and electrons, electromagnetic perturbations, collisionality, and toroidal flow and flow shear) use parameters from a high beta NSTX discharge. The predicted electron thermal transport is comparable to experimental analysis, and it is dominated by the electromagnetic contribution of electrons free streaming along stochastic magnetic field line trajectories. The structure of the turbulence is distinctly different from traditional tokamak turbulence and initial ideas for experimentally diagnosing such characteristics, and the associated transport, were additionally presented.
 
Leonid E. Zakharov participated in a two day workshop on "Stationary Fusion Neutron Sources" in Kurchatov Institute (Moscow) February 21-22. In his presentation, "Potential Burning Plasma Regimes for Tokamak Based VNS", he outlined the necessity of new plasma regimes for better confinement, stability and plasma-wall interaction for neutron sources based on tokamaks with a conventional aspect ratio 4 and 1 m thick blanket. In another presentation "ASTRA-ESC simulations for the LiWF regime in a Compact FNS-ST" he revealed a possibility of a very attractive "hot-ion" regime with Ti >3Te for a small plasma configuration R/a=0.5/0.3 with P_DT=0.2-0.8 MW.
 
Peter Damiano attended the American Geophysical Union Chapman conference on "The Relationship Between Auroral Phenomenology and Magnetospheric Processes" in Fairbanks Alaska (Feb 27 - Mar 4) and gave an oral presentation entitled "Characteristics of current saturation and electron energization in simulations of global scale Alfven waves".
 
Dr. Ilya Dodin of Princeton University, gave a Theory seminar entitled, "Ponderomotive Forces, Wave Dispersion, and Action Conservation" on March 3. His talk focused on two types of adiabatic effects. First is fundamental properties of nonlinear forces on particles due to waves and second is the particle influence on waves. He presented the derivation of a general nonlinear dispersion relation for arbitrary stationary waves in plasma, which includes the kinetic effects without solving the Vlasov equation and the frequency shifts due to trapped particles. He finally showed the application of adiabatic conservation laws to plasmas undergoing densification in various contexts. Existing results pertaining to slow transformation of linear waves in such plasmas are generalized and corrected.
 
The paper " Nonlinear Simulation of Toroidal Alfven Eigenmode with Microturbulence-induced Radial Diffusion" by Jianying Lang and Guo-Yong Fu, has been accepted for publication in the Physics of Plasmas. The abstract reads as follows: "It is shown analytically and numerically that microturbulence-induced diffusivity can affect nonlinear saturation of energetic particle driven modes in the similar way as collisional pitch angle scattering does. Introducing a simple diffusion operator to the code, our numerical results have shown that a single Toroidal Alfven Eigenmode (TAE) can saturate at a steady state with sufficiently high diffusion rate. The calculated saturation level scales with the radial diffusion rate by the same scaling of pitch angle scattering. A criterion is derived to judge the importance of microturbulence-induced radial diffusion effect comparing to the collisional pitch angle scattering effect. According to the criterion, we find that the micro-turbulence induced diffusion has a stronger effect than the Coulomb collision on the TAE saturation in present tokamak devices and future burning plasmas (ITER)".

February 18, 2011

THEORY GROUP

Riccardo Betti was a member of the DIII-D Program Advisory Committee that met at General Atomics on February 15-17. The PAC reviewed the DIII-D experimental and theory programs and made recommendations to the DIII-D team.
 
John Krommes was nominated Outstanding Referee for Physical Reviews and Physical Review Letters. Initiated in 2008, the Outstanding Referee program expresses appreciation for the essential work that anonymous peer reviewers do for APS journals. Each year a small percentage of 45,000 active referees are selected and honored with the Outstanding Referee designation. Selections are made based on the number, quality, and timeliness of referee reports as collected in a database over the last 23 years. The program recognizes about 150 referees each year.
 
Professor George Tynan of the Department of Mechanical and Aerospace Engineering at the University of California at San Diego (UCSD) gave a Theory department seminar entitled "Turbulent momentum transport in magnetized plasmas: results from linear devices and thoughts for studies on toroidal confinement devices". Turbulent momentum transport is thought to play an important role in the formation of sheared ExB flows in confined plasmas and thus may be a key piece of important macroscopic transport phenomena such as critical gradient behavior and transport barrier formation. In addition, recent experiment and theory suggest it may play a crucial role in the formation of so-called "intrinsic rotation" in confined plasmas, where the plasma acquires a net rotation in the absence of external momentum input. Thus turbulent momentum transport studies are of fundamental importance for magnetic fusion. In his presentation, Professor Tynan provided an overview of studies of turbulent momentum transport in a linear plasma device. Using a mixture of multi-point probe studies, digital signal processing techniques and fast imaging diagnostics, he showed how coherent drift waves transition into a turbulent spectrum, and how coherent sheared zonal flows arise from the turbulence during this transition. The results showed the key role that turbulent symmetry breaking plays in the formation of large scale ordered flows out of turbulence, and suggested several laboratory and confinement device experiments.
 
Five members of the Theory Department participated in the FSP planning workshop from February 8-11 in San Diego: Daren Stotler, Doug McCune, Guo-Yong Fu, Weixing Wang, and Allan Reiman.

COMPUTATIONAL PLASMA PHYSICS GROUP

Stephane Ethier participated in the "Juelich Blue Gene/P Extreme Scaling Workshop" held at the Juelich Supercomputing Center in Juelich, Germany. The purpose of the three-day workshop was to give application teams an opportunity to scale their code across the full Blue Gene/P system JUGENE, which consists of 72 racks with a total of 294,912 cores - the highest number of cores world-wide. The benchmark version of the gyrokinetic kinetic particle-in-cell code GTC successfully ran on the full system, pushing over 14 billion particles per second per time step. The test run used 95 billion particles.

February 11, 2011

THEORY

Professor Bruno Coppi of MIT gave a Theory department seminar entitled,  "Attractive Properties of the I-Confinment Regime and Interpretation: Relevance  to the Ignitor Experiment". In his presentation he summarized the theoretical model (Coppi et al., PRL, 17, 1966, 377-379) that can account for the characteristics and effects of the plasma mode around 200 kHz that is excited in connection with the onset of the I Confinement Regime. The  mode is associated with the presence of a heavy particle (impurity) population near the edge of the plasma column and the theory correctly predicts the direction of the mode phase velocity (that of the electron diamagnetic velocity). The driving factor for the mode is the temperature gradient of the main ion population combined with the finiteness of the impurity population temperature.  The main effects of the mode is to transport the impurity population outward (increasing their density gradient) and the main ion population inward while allowing an outward flow of the thermal energy of the hotter plasma. This characteristic of impurity confinement makes the  confinement regime of particular interest for experiments aimed at producing plasmas close to ignition conditions. In the last part of his talk, Professor Coppi presented an overview of the Ignitor experiment (to be constructed in Russia) that will be able to expand on experiments studying the I-Mode Confinement Regime that have been conducted with the Alcator C-Mod experiment at MIT.

February 4, 2011

THEORY GROUP

Professor C.Z. (Frank) Cheng of National Cheng Kung University in Taiwan, gave a Theory seminar entitled, "Observation Features and Physical Mechanisms of Substorms". Professor Cheng started his presentation by giving an overview of the space plasma physics related research efforts currently underway (and under development) at Cheng Kung University. He then presented his official seminar summarizing the observational characteristics and physical mechanisms behind magnetospheric substorms. In particular, he highlighted the fine structure observed in the onset arc and the associated magnetic fluctuations evident in the Pi1 and Pi2 frequency ranges (as measured from ground based magnetometers) and their exponential growing behaviors evident before the onset of the substorm expansion phase. He discussed the kinetic ballooning instability as a possible physical mechanism for substorm onset in the magnetosphere and the formation of the auroral breakup arc in the ionosphere. He also presented the nonlinear evolution of the onset arc breakup and magnetic fluctuations into turbulent states, the current disruption and magnetic field dipolarization processes and the dispersionless particle injection during the expansion phase.
   
COMPUTATIONAL PLASMA PHYSICS GROUP

A new version of ElFresco, the full-wave reflectometer simulation code, has been released by Ernest Valeo, Gerrit Kramer, and Eliot Feibush. The model of the transmitter antenna has been expanded so users can specify a 2-D array for the incident electric field. Visualizations have been added for the incoming and outgoing data in the paraxial region. Both plasma profile and wave field data can be displayed with a variety of presentation types: color map plots, contour plots, indexed plots, and surface plots. Managing and monitoring simulation runs has also been improved.
 
A new capability for modeling electron cyclotron heating and current drive ( ECRH and ECCD) has been added to the TSC code. All heating systems, NB, ICRH, LH, and ECRH can now be applied in a single simulation run. For each of these, the user has the choice of using the internal TSC model for heating and current drive, or using a more detailed model through the SWIM IPS (or a coupled LSC calculation for LH).

January 28, 2011

Theory Group

A paper titled, “Trapped Electron Mode Turbulence Driven Intrinsic Rotation in Tokamak Plasmas”, by W. X. Wang, T. S. Hahm, S. Ethier, L. E. Zakharov and P. H. Diamond, has been accepted for publication in Physical Review Letters. In this letter, recent progress from global gyrokinetic simulations in understanding the origin of intrinsic rotation in toroidal plasmas is reported with emphasis on electron thermal transport dominated regimes. The turbulence driven intrinsic torque associated with nonlinear residual stress generation by the fluctuation intensity and the intensity gradient in the presence of zonal flow shear induced asymmetry in the parallel wave number spectrum is shown to scale close to linearly with plasma gradients and the inverse of the plasma current. These results qualitatively reproduce empirical scalings of intrinsic rotation observed in various experiments. The origin of current scaling is found to be due to enhanced k// symmetry breaking induced by the increased radial variation of the safety factor as the current decreases. The physics origin for the linear dependence of intrinsic torque on pressure gradient is that both turbulence intensity and the zonal flow shear, which are two key ingredients for driving residual stress, increase with the strength of turbulence drive, which is R0/LTe and R0/Lne for the trapped electron mode.                                                                                                              

On January 25, Daniel Raburn defended his Ph.D. thesis, which was done under the supervision of A. Reiman and D. Monticello and with the collaboration of E. Fredrickson and R. Samtaney. The work implemented a Jacobian-Free Newton-Krylov scheme for converging the PIES 3D equilibrium code. The new scheme was found to be about an order of magnitude faster than the Picard algorithm that had previously been used. The modified code was applied to calculate the saturated island widths for neoclassical tearing mode (NTM) experiments in the TFTR tokamak where the island width was carefully measured. The calculations were found to agree with the experimental data within the error bars.

CPPG Group

A highly simplified model for density prediction has been demonstrated in PTRANSP, using a Plasma State based solver. The model, which uses a prescribed density profile shape and a two parameter 0d scaling, was implemented to develop and test PTRANSP's capability to receive a
density solution from a portable solver module coded without dependence on PTRANSP internals.

January 21, 2011

January 14, 2011

THEORY GROUP

December 24, 2010- January 7, 2011

THEORY GROUP

W. W. Lee, as an adjunct professor at the Department of Applied Physics and Applied Mathematics, Columbia University, has just finished teaching a graduate/upper undergraduate level course on "Kinetic Theory and Simulation of Kinetic Plasmas" for the Fall Semester of the 2010-2011 academic year. The topics included: various reduced descriptions for the Vlasov-Maxwell system, wave-particle interactions, collisional effects, gyrokinetic equations and their relationship to MHD, microinstabilities and  the various simulation methods for plasmas. A one-dimensional Particle-In-Cell code using MATLAB was used for the course.
 
Jia Zhu, a graduate student from Zhejiang University in China, has finished his eight-month long visit to PPPL Theory Department from April 23-December 23, 2010, and returned to China. During his visit, he worked with Dr. Guoyong Fu on nonlinear simulations of energetic particle-driven Alfven modes in tokamak plasmas as a part of his Ph.D. thesis studies. A reduced model code has been built where the mode structures of Alfven modes such as TAEs are fixed and mode amplitudes and phases are evolved due to wave particle interaction between energetic particles and Alfven waves. The initial simulation results compare well with NOVA-K code and M3D-K code for the linear growth rate of an n=1 TAE mode. Work is in progress to benchmark the code in the nonlinear regime and to investigate the nonlinear saturation of multiple Alfven modes driven by fast ion.