Table of contents

This special issue presents a collection of refereed plenary and invited papers presented at the International Conference on Plasma Physics held in Nagoya, Japan, September 9 - 13, 1996 (ICPP96).

The ICPP is the only existing international conference covering the whole range of plasma science. Fusion plasma science and space and astrophysical plasma science have grown to big independent trees stemming from a tiny seed of research into plasma waves and instabilities. Another tree is now developing with the rapid growth in the field of industrial applications and plasma processing. Additionally, recent advances in supercomputer technology have enabled us to simulate non-linear and non-equilibrium processes and open up a new trend in science, science of complexity.

Each speciality of plasma science is pushing forward, and is maturing as an independent discipline. Under these circumstances, this conference has facilitated a forum for identifying common threads that cut across boundaries of growth and newly growing plasma fields.

The primary aim of this conference is to integrate and unify the diverse specialized disciplines into a single global discipline of science, and to play a leading role in promoting a shift in science.

We would like to express our sincere appreciation to the authors and to the members of Program Committee (M Fujiwara, T Hayashi, Y Kawai, K Mima, M Mori, K N Sato, N Sato, R Sugihara, T Yamanaka and Y Yasaka), who worked diligently to select and organize the presentations. Special thanks also go to all members of the International Advisory Committee, the Organizing Committee, the Steering Committee and the staff of the Japan Society of Plasma Science and Nuclear Fusion Research, for their continuous efforts in making the conference successful.

Atsuo Iiyoshi Chairperson, Organizing Committee

Tetsuya Sato Chairperson, Program Committee

In two dimensions (2D), a guiding-centre plasma and an inviscid fluid can be described by a continuum model or by quasi-particles (filaments) with Coulomb interaction. Other 2D continuum models are equivalent to quasi-particles with a screened Coulomb interaction.

Such 2D systems of quasi-particles have negative temperature equilibria, characterized by large fluctuations. In the continuum models this corresponds to the spontaneous appearance of macroscopic clumps of charge or vorticity and may represent a stage of 2D turbulence. However, this raises the basic question; can a limited number of particles ever represent the behaviour of a continuous fluid? At first sight the two models are incompatible. The particles are conservative, have finite degrees of freedom and few isolating invariants. The fluid has infinite degrees of freedom and is conservative only if dissipation is ignored - when it has an infinity of invariants.

Despite these fundamental differences, the two systems may be reconciled if the fluid is viscous and the quasi-particles are chosen appropriately. Roughly speaking, a small viscosity destroys invariants of an ideal fluid and limits its degrees of freedom, while preserving its essentially conservative behaviour.

The x-ray emission from a star like the sun comes from plasma at 10⁶-10⁷ K and 10¹⁰ ions cm^-3- trapped in the 100 Gauss bipolar magnetic fields of active regions. The x-ray intensity reaches 10⁷ ergs cm^-2s^-1, which is a measure of the heat input required to maintain the temperature of the emitting gas. The basic energy input is photospheric convection, which swirls and intermixes the photospheric footpoints of the bipolar magnetic fields, thereby creating an interlaced topology of the field lines in the bipolar field above the photosphere. The basic theorem of magnetostatics asserts that almost all field topologies develop internal tangential discontinuities as the field relaxes to equilibrium. The discontinuities are a necessary and sufficient condition for the Maxwell stresses to achieve static balance. It appears that rapid reconnection across these internal current sheets provides the principal heat source for the x-ray corona.

Energetic plasmas are found in all astrophysical environments. However, beyond the solar system, our understanding of cosmic plasma dynamics has remained quite poor. This paper presents an attempt to explore a new description of various high-energy astrophysical phenomena, observed from galaxies and galaxy clusters, in terms of plasma magnetohydrodynamics. In this context, three specific topics are described; diffuse hot plasmas filling the Milky Way, plasma acceleration by giant black holes, and hot plasmas of galaxy clusters with relevance to cosmology.

In this paper we shall highlight some topics of specific concern for both concepts for magnetic confinement: 2D (tokamaks) and 3D (helical) systems. Section 2 will address the primary goals of fusion: density, confinement and temperature. We shall summarize the density limit, the global confinement time scaling with engineering parameters and some physical elements of improved confinement regimes with edge (H-mode) and core transport barriers (reversed shear cases); progress in relation to temperature originates from the evidence for α-particle heating from TFTR and the successes in drift optimization in helical systems with some confirmation from W7-AS. In the section 3 on the secondary goals of fusion, β will be discussed in terms of equilibrium (including the necessary improvements in helical systems with examples from W7-AS) and in terms of stability (including fast particle-driven Alfvén waves with toroidally induced eigenmodes in cases with high shear and global eigenmodes in cases with low shear). Finally, we shall summarize the status of exhaust physics, with reference to the poloidal field divertor in tokamaks and the island divertor in helical systems. The intention of this report is to emphasize the complementarity of the two concepts of toroidal confinement.

This paper presents recent experimental results that clearly demonstrate the complex interplay between parametric instabilities in a laser-produced plasma. These results include: (a) the interplay between ion acoustic waves (IAW) and electron plasma waves (EPW) associated with stimulated Brillouin scattering (SBS) and stimulated Raman scattering (SRS), (b) the interplay between IAW from two independent SBS decays when two interaction beams are present and (c) the observation of a secondary process, the Langmuir decay instability (LDI), which is the further decay of EPWs driven by SRS, into an IAW and a secondary EPW.

Dust phenomena in processing plasmas are reviewed from the new viewpoint of birth of material in the plasma. The gas-phase growth of particles has been extensively studied for Si₄ RF plasmas. The Si particles usually grow through three distinctive stages: an initial growth phase up to about 10 nm, whose size is between size ranges dominated by plasma properties and their electron affinity; a rapid growth phase and a growth saturation phase. In the first phase, particles are localized around the plasma/sheath boundary in spite of a large fraction of neutral particles. In the second phase, extremely fast coagulations between particles appear, which can be explained by the generation of positively charged particles due to high-energy electrons. In the third phase, most of the particles are charged negatively and hence repel each other.

Experiments on the formation of Coulomb solids in carbon RF plasmas and production of K-C₆₀ plasmas are attractive since they indicate great potential for dusty plasmas. In the former experiments, almost perfectly spherical and monodisperse particles are successfully formed and transitions from liquid to solid phases and between body-centred cubic and hexagonal structures are clearly observed. In the latter experiments, plasmas composed of K⁺ and C₆₀^- ions are almost realized. Film properties change by controlling the energies of charged particles impinging on substrates and, under the specified plasma conditions, C₆₀ with the K atom in its cage is successfully formed.

A self-consistent set of MHD equations accounting for finite photon mass has been presented. General properties of these equations have been discussed. A reduced set describing the motions with a length scale exceeding the Compton length of the photon has been derived. Some specific solutions in this reduced set have been analysed in the context of possible improvement in the estimate of the photon mass on the basis of various astrophysical observations.

Computer simulation of a magnetohydrodynamic (MHD) dynamo is performed. A magnetic field is generated by the thermal convection motion of an electrically conducting fluid in a rotating spherical shell. This study is motivated by the mysterious origin of the earth's dipole field. It is shown that a strong dipole field is spontaneously generated by the MHD dynamo. The dipole field generation is closely related to the columnar structure of the convection motion, which is characteristic of the rotating spherical convection.

Three aspects of the physics of pair plasmas are discussed. The main emphasis is on the properties of waves in cold, magnetized pair plasmas, where it is shown that the inclusion of an admixture of positrons introduces an additional `cyclotron mode' into the magneto-ionic theory. Thermal effects on these waves are described. The quadratic nonlinear response and the fluid description of a pair plasma are discussed briefly.

Alfvén waves are ubiquitous in space plasmas and are the means by which information about changing currents and magnetic fields are communicated. Shear Alfvén waves radiated from sources with cross-field scale size of the order of the electron inertial length, δ=c/ω_pe, have properties which differ considerably from planar magneto-hydrodynamic waves. Currents of cross-field size δ are common in space plasmas. A series of experiments in the large plasma device (LAPD) at UCLA is presented which illustrates that waves generated by small scale fluctuating currents radiate across magnetic field lines and are associated with complex three-dimensional currents. Waves generated by two sources are observed to constructively interfere to produce large magnetic fields in spatial regions away from source field lines. Volume data sets will be presented and discussed in the light of space plasma applications.

We have experimentally found that ion - neutral collisions in a magnetized plasma cause a drastic change in the dispersion relation of the Alfvén wave at frequencies below the ion - cyclotron frequency. In a consecutive experiment, we have also found that the measured average velocity of the wave packet does not obey the conventional group velocity that exceeds the velocity of light, but obeys a new defined group velocity that is much lower than the velocity of light. Furthermore, when the initial central frequency of the wave packet is just located on the anomalous dispersion range of frequency, the wave packet splits, when propagating along the magnetic field, into two new wave packets with central frequencies being higher and lower than the initial central frequency. One of the new packets with lower central frequency is confirmed to be the shear Alfvén wave and the other the compressional Alfvén wave; the behaviour of these new wave packets can also be explained by our new theory.

Plasma potential formation is investigated in a fully ionized collisionless plasma flow in the presence of an electron cyclotron resonance (ECR) point under simple mirror configurations of magnetic field. When ECR takes place at the magnetic-well bottom, a potential structure is observed to consist of a negative potential depression around the ECR point and a subsequent positive potential peak. This potential structure gradually collapses in the timescale determined by the ion-flow speed. In the case of a converging magnetic field, such a potential structure is observed to persist in the steady state, when the ECR point is located in the region of good magnetic-field curvature. However, this structure is formed only transiently when the ECR point is located in the region of bad magnetic-field curvature. When ECR takes place in a diverging magnetic field, on the other hand, there appears a strong potential drop along the field lines, which results from a field-aligned electron acceleration, being accompanied by an effective ion acceleration.

A review of our theoretical studies and computer simulations with PIC codes of the interaction of ultrashort, high-intensity laser pulses with plasmas is presented. The mutual attraction of the currents inside the self-focused channels, arising from the pulse filamentation, makes them interact magnetically and coalesce into a single channel with strongly enhanced electromagnetic energy density. Pulses of finite length and width propagate in the shape of a `bullet' and produce a wake consisting of magnetic dipoles correlated with an electron vortex row behind the pulse. The vortices evolve into an antisymmetric configuration which is shown to be stable when the distance between its vortices is greater than the electron skin depth.

The ignition condition and fusion gain for isochoric implosion with off-centre ignition are evaluated based on 2D hydrodynamic simulations. We show that the ignition energy can be reduced by increasing the compressed fuel density, although it is accompanied by a reduction in the fusion gain. Experimental results on channel formation in an overdense plasma are presented. It is shown that the laser beam propagates in a channel over a 200 μm distance in an overdense plasma and deposits its energy close to the original target surface. For the study of fast ignition, ultrashort-pulse high-energy lasers are constructed with GEKKO MII and GEKKO XII at ILE.

Investigations into laser beam spraying, deflection and induced scattering in plasmas are presented. Recent calculations and experiments on beam spraying due to filamentation are discussed. A simple model is presented for an enhanced beam deflection associated with nearly sonic plasma flow transverse to the beam. This model provides useful insights on the laser beam deflection, its scaling and the importance of self-consistent profile modifications. Finally, some discussion is given of recent experiments demonstrating the interplay between stimulated Raman and Brillouin scattering.

Since the pioneering work of Tajima and Dawson on the laser electron accelerator, experiments on the excitation of high-amplitude plasma waves by beat-wave, photon wakefield and self-resonant wakefield have been performed in plasmas and give evidence of electron acceleration in the range 1 - 100 MeV for distances of less than one centimetre. Electrons, either from the plasma itself, or injected from outside, are accelerated to these relativistic energies by the longitudinal electric field of the laser-excited plasma wave whose phase velocity is close to the speed of light. After a short survey of the excitation mechanism of this relativistic plasma wave, the scaling laws which link the electric field of the plasma wave, the energy gain of accelerated electrons and the acceleration length to the parameters of the laser will be recalled. Then, limits for the energy of accelerated electrons will be given in the case of passing and trapped electrons as well as in the case of wavebreaking. These classical results will be compared with recent experimental results of electron acceleration in CO₂ and Nd, beat-wave, as well as for self-resonant wakefield. Future experiments on photon wakefield will be discussed. Relevant work performed to improve the energy gain of acceleration schemes such as light channelling by preformed plasmas or by relativistic effects will be discussed.

Active galactic nuclei are interesting astrophysical objects which contain a super-massive black hole at the central region as an engine of their activity. Plasma accretion onto the black hole is the fundamental process for producing the high and variable x-ray luminosity from the central region, and the magnetic field is expected to be a crucial component in the accretion models to explain various observed features. This motivates us to study general-relativistic effects in a black-hole magnetosphere, in particular, in relation to the important role of the spinning motion. The stationary and axisymmetric magnetospheric structure based on the general-relativistic magnetohydrodynamic treatment under the assumption of infinite conductivity is briefly reviewed. We then turn to analyse the time evolution of axisymmetric magnetic fields in the vicinity of the black hole, taking into account the contribution of a finite magnetic diffusivity .

Our approach is to solve the general-relativistic Maxwell equations supplemented by the generalized Ohm's law for a given velocity field of accreting and rotating plasma. Numerical examples of time-dependent solutions which show self-excitation of dipolar and quadrupolar modes of the poloidal magnetic field were recently given by Khanna and Camenzind. However, the validity of the dynamo action remains controversial. Here, we develop the analytical perturbation method under the approximation of small diffusivity , where is the black-hole radius, and our analysis is limited to the narrow boundary region very close to the black-hole surface. The condition for growing multipolar seed fields is expressed by rotation parameters. We discuss our simple analytical results in comparison with the numerical calculations.

We show that the recent observations of x-rays from comet Hyakutake can be explained by the acceleration of electrons along the magnetic field by lower hybrid waves excited by the interaction of the cometary ions with the solar wind.

Electrodynamical features of the black aurora, and its optical counterpart, the aurora, are discussed on the basis of electric field, magnetic field and particle observations by Freja. Extraordinarily intense (1-2Vm_-1) and small-scale (1 - 5 km) electric fields have been observed by the Freja satellite mostly in association with black aurora (vortices and east - west dark filaments) but also in association with auroral structures such as spirals and surges. The former structures are characterized by diverging electric fields, dropouts of energetic electron precipitation, precipitating or transversely energized ions and downward field-aligned currents carried by upward fluxes of ionospheric electrons, opposite to the characteristics of the aurora. The black auroral structures are associated with localized ionospheric density depletions below that of the ambient density and the field magnitude is found to be anticorrelated to the conductivity. During conditions of minimum conductivity small-scale diverging electric fields of 1Vm_-1 occasionally exist down to 800 km. We suggest that the diverging electric fields observed by Freja are associated with low-altitude and narrow (≈1-2Km) potential structures similar to the auroral potential structures at higher altitude but associated with a positive space charge and a downward parallel electric field. This is supported by Freja observations of intense upward electron beams and positive potential peaks as well as by low-altitude electric field observations by the S3-3 and Viking satellites.

The electric field characteristics of auroral surges and large-scale auroral spirals are very different from those of the black aurora. This will be exemplified by high-resolution electric field, magnetic field, particle and UV-imager observations from oblique Freja crossings of the eveningside auroral oval. The ambient electric field is found to intensify in the direction towards the centre of the auroral spiral which confirm previous findings that the surge and spiral head is associated with negative space charge and an intense upward field-aligned current. A pass directly through the surge head, however, reveals a very inhomogeneous and complicated picture of the surge such as narrowly structured, intense (up to 700 mV m_-1) converging electric fields, intense electron precipitation and balanced field-aligned currents (up to 30μAm_-2) embedded within an extended region of intense high-energy electron precipitation, weak electric fields and field-aligned currents. According to some surge models, a pronounced westward electric field component and a southward polarization electric field is expected within the entire high-conductivity region but evidence in support of this was not found in the data. Rather, these suggest that a significant part of the upward surge current is closed by distributed downward field-aligned currents from the near surroundings. The surge electric field is much more intense than previously observed or anticipated at these altitudes, having characteristics rather similar to those observed in the auroral acceleration region.

Localized electrostatic wave packets in the frequency region of lower-hybrid waves have been detected by instrumented rocket payloads and recently also by the instruments on the FREJA satellite. These waves are often associated with local density depletions, indicating that the structures can be interpreted as wave-filled cavities which are strongly elongated along the magnetic field. The basic features of the observations are discussed. Based on simple statistical arguments, we attempt to present some characteristics which have to be accommodated within an ultimate theory describing the observed wave phenomena.

We review observations of nonlinear plasma waves detected by interplanetary spacecraft. For this paper we will focus primarily on the phase-steepened properties of such waves. Plasma waves at comet Giacobini - Zinner measured by the international cometary explorer (ICE), at comets Halley and Grigg - Skjellerup measured by Giotto, and interplanetary Alfvén waves measured by Ulysses, will be discussed and intercompared. Owing to the page limitations of this paper we will keep the cometary wave discussion brief (but references for the interested reader will be provided), and we will focus more on the recent Ulysses observations of phase-steepened Alfvén waves. For the latter, classical `interplanetary rotational discontinuities' are explained as phase-steepened edges of Alfvén waves. Both the Alfvén waves and their phase-steepened edges will be shown to be arc-polarized. A method of determination of the direction of propagation of these waves will be discussed.

Sheared plasma flows between the magnetosheath and magnetosphere exist in vast regions of the magnetopause and contain much more free energy than other energy sources. It is of importance to consider the presence of the velocity shear while studying reconnection at the magnetopause. This paper briefly reviews our current knowledge of vortex-induced magnetic reconnection obtained from two-dimensional MHD simulations with emphasis on the dynamic features of the vortex-induced reconnection process.

The progress of the experimental LHD program, the largest superconducting helical device in the world, is described. The LHD device, which is now under construction, will start experimentation in 1998 to study the confinement of high-temperature plasmas relevant for thermonuclear fusion conditions and will develop the physics and technology of steady-state operation. Experimental research on the LHD also contributes to physics understanding of toroidal plasmas by the study of high-temperature plasmas confined in the externally defined magnetic structure. New research directions for advanced helical systems are also discussed.

A summary is presented of recent alpha-particle experiments on the tokamak fusion test reactor. Alpha particles are generally well confined in MHD-quiescent discharges, and alpha heating of electrons has been observed. The theoretically predicted toroidicity-induced Alfvén eigenmode has been seen in discharges of of alpha power, but only in plasmas with weak magnetic shear.

JET has conducted extensive experiments in H-mode plasmas focusing on two broad aims. The first, involving the characterization of reactor-relevant aspects of the regime, encompasses the study of steady-state H-modes at plasma currents of up to 6 MA. In these plasmas, energy confinement is typically a factor of 1.8 - 2 greater than predicted by L-mode scalings and local transport varies with the ion Larmor radius, at least for input powers well above the H-mode threshold power. Using nitrogen impurity seeding, a regime has been developed in which good core confinement, with an enhancement of 1.5 relative to L-mode scaling, is combined with radiated power fractions of up to 80%. The second major goal is to optimize fusion power production by exploiting high-confinement ELM-free H-modes, which exhibit energy confinement enhancement factors of three or more relative to L-mode plasmas. In this regime, fusion power quotients have been achieved, though the highest fusion performance is transient.

In a magnetically diverted tokamak, the scrape-off layer (SOL) and divertor plasma separates the first wall from the core plasma, intercepting impurities generated at the wall before they reach the core plasma. The divertor plasma can also serve to spread the heat and particle flux over a large area of divertor structure wall using impurity radiation and neutral charge exchange, thus reducing peak heat and particle fluxes at the divertor strike plate. Such a reduction will be required in the next generation of tokamaks, for without it the divertor engineering requirements are very demanding. To successfully demonstrate a radiative divertor, a highly radiative condition with significant volume recombination must be achieved in the divertor, while maintaining a low impurity content in the core plasma.

Divertor plasma properties are determined by a complex interaction of classical parallel transport, anomalous perpendicular transport, impurity transport and radiation, and plasma - wall interaction. In this paper we will describe a set of experiments on DIII-D designed to provide detailed two-dimensional documentation of the divertor and SOL plasma. Measurements have been made in operating modes where the plasma is attached to the divertor strike plate and in highly radiating cases where the plasma is detached from the divertor strike plate. We will also discuss the results of experiments designed to influence the distribution of impurities in the plasma using enhanced SOL plasma flow. Extensive modelling efforts will be described which are successfully reproducing attached plasma conditions and are helping to elucidate the important plasma and atomic physics involved in the detachment process.

The theory of far-nonequilibrium transport of plasmas is described. Analytic as well as simulation studies are developed. The subcritical nature of turbulence and the mechanism for self-sustaining are discussed. The transport coefficient is obtained. The pressure gradient is introduced as an order parameter, and the bifurcation from collisional to turbulent transport is shown. The generation of the electric field and its influence on the turbulent transport are analysed. The bifurcation of the radial electric field structure is addressed. Hysteresis appears in the flux-gradient relation. This bifurcation causes the multifold states in the plasma structure, driving the transition in the transport coefficient or the self-generating oscillations in the flux. The structural formation and dynamics of plasma profiles are explained.

The topic of plasma flows and their influence on the equilibrium and stability properties of tokamak discharges is briefly reviewed. The stability of high-n ballooning modes in the presence of sheared toroidal rotation is discussed to highlight recent advances in our understanding of the spatial structure and temporal evolution of these modes. The eikonal solution, which is mathematically equivalent to the normal mode solutions at the lowest order in 1/n, may not be physically relevant for a variety of reasons. An analytic formulation of the normal mode problem for high-n ballooning modes in the presence of sheared toroidal flow in the high-β large-aspect-ratio limit is presented. In contrast to earlier studies, the present formulation allows for large plasma shifts as well as accounting for changes in the shape of flux surfaces due to rotation effects.

The first experimental study of a mixed slab - toroidal ITG (ion temperature gradient) driven mode in the Columbia linear machine (CLM) is reported. The effects of the modest levels of magnetic curvature are seen to be further destabilization and reduction of the real frequency of the mode. However, with a further increase in the mirror ratio, the mode amplitude decreased. The mode has the appearance of a hybrid ITG-dissipative trapped electron mode.

We have measured the reduction of anomalous fluctuation driven particle transport due to feedback via cross-correlation of Langmuir probe and capacitive probe data for a centrifugal mode driven by E×B rotation of the plasma column. The particle flux is obtained from cross power spectra of density and potential fluctuation. The resulting diffusion coefficient clearly indicates that feedback can reduce the transport by up to a factor of three.

To investigate the question of non-linear saturation of the mode we have performed a bispectral analysis of fluctuation data. The results clearly indicate strong mode coupling in the spectral width of the mode. Furthermore, we resolve the spectral width to reveal that it is composed of radial harmonics. These observations suggest that the nonlinear mode coupling between these radial harmonics is the saturation mechanism of the instability, as well as the underlying cause of the consequent anomalous transport.

Sawtooth activity is a prominent example of a global plasma instability which is observed in virtually all tokamak devices. Despite numerous experimental and theoretical investigations, the phenomenon is still barely understood. As far as experimental effort is concerned, much attention has been paid to soft x-ray emission from the plasma and to its analysis in terms of two-dimensional contour plots, because it is thought to reflect the shape and temporal behaviour of magnetic flux surfaces during a sawtooth cycle. Recently, more direct methods of detecting sawtooth-related changes in the magnetic field structure have become available and have added new facets to the general picture. In this paper, some observations made on the Jülich tokamak TEXTOR by means of a Faraday rotation diagnostic technique will be reported. First, in correlation with the sawtooth collapse a localized periodic perturbation of the magnetic field with principal mode numbers m = 1 and n = 0 has been detected which, in the presence of an m = n = 1 island, may give rise to magnetic field line stochastization and thereby contribute significantly to a rapid expulsion of electronic energy from the plasma core region. Second, the so-called precursor oscillations prior to a sawtooth crash have been investigated and estimates have been obtained for the growth rate and width of a magnetic island forming immediately before the collapse.

The injection of hydrogen/deuterium pellets into a tokamak leads to a sudden increase in the electron density, and subsequently to a profile peaking of the density and an increase in the stored energy. Immediately after the injection, different types of oscillations are excited. On TEXTOR, the first type immediately follows the injection and the second one is excited with a delay of more than 10 ms. The oscillations show a `snake-like' structure and occur close to the q = 1 surface with a frequency of 0.7 - 2 kHz. The radial location of the second oscillation is slightly shifted with respect to the first one. A fast-cooling phenomenon (`pre-cooling') in the core region of a plasma is often observed at pellet ablation phase. A study on the relation between the `pre-cooling' and sawtooth oscillations suggests that the central value of safety factor of plasmas, q(0), is kept sufficiently below unity even just after the sawtooth crash. During pellet injection, the ablation rate is strongly modulated; these modulations cause so-called `striations' in the ablation cloud. One model relates the striations to the energy reservoir on the plasma flux surfaces and describes the possibility of deriving the q-profile; the question of whether this method provides reliable results cannot yet be answered conclusively. The trajectory of the pellet in the plasma is in general not straight but deflected in the electron drift direction (OH discharges) or in the ion drift direction (CO-NBI discharges). The cloud develops a helically structured tail in the electron flow direction (toroidally) and in the electron diamagnetic drift direction (poloidally). The tail structure is attributed to charge-exchange processes and to plasma rotation.

The possibility of a high-performance/low-cost fusion reactor concept which can simultaneously satisfy (1) high beta, (2) high bootstrap fractio (self-sustaining) and (3) high confinement is discussed. In CDX-U, a tokamak configuration was created and sustained solely by internally generated bootstrap currents, in which a 'seed' current is created through a nonclassical current diffusion process. Recent theoretical studies of MHD stability limits in spherical tori [e.g. the National Spherical Torus Experiment (NSTX)] produced a promising regime with stable beta of 45% and bootstrap current fraction of . Since the bootstrap current is generated by the pressure gradient, to satisfy the needed current profile for MHD stable high beta regimes, it is essential to develop a means to control the pressure profile. It is suggested that the most efficient approach for pressure profile control is through the creation of transport barriers (localized regions of low plasma transport) in the plasma. As a tool for creating the core transport barrier, poloidal-sheared-flow generation by ion Bernstein waves (IBW) near the wave absorption region appears to be promising. In PBX-M, application of IBW power produced a high-quality internal transport barrier where the ion energy and particle transport became neoclassical in the barrier region. The observation is consistent with the IBW-induced-poloidal-sheared-flow model. An experiment is planned on TFTR to demonstrate this concept with D - T reactor-grade plasmas. For edge transport control, a method based on electron ripple injection (ERI), driven by electron cyclotron heating (ECH), is being developed on CDX-U. It is estimated that both the IBW and ERI methods can create a transport barrier in reactor-grade plasmas (e.g. ITER) with a relatively small amount of power .

Physics issues of a tokamak plasma with a hollow current profile produced by a large bootstrap current are discussed based on experiments in JT-60U. An internal transport barrier for both ions and electrons was obtained just inside the radius of zero magnetic shear in JT-60U. Analysis of the toroidal ITG microinstability by toroidal particle simulation shows that weak and negative shear reduces the toroidal coupling and suppresses the ITG mode. A hard beta limit was observed in JT-60U negative shear experiments. Ideal MHD mode analysis shows that the n = 1 pressure-driven kink mode is a plausible candidate. One of the methods to improve the beta limit against the kink mode is to widen the negative shear region, which can induce a broader pressure profile resulting in a higher beta limit. The TAE mode for the hollow current profile is less unstable than that for the monotonic current profile. The reason is that the continuum gaps near the zero shear region are not aligned when the radius of q_min is close to the region of high . Finally, a method for stable start-up for a plasma with a hollow current profile is described, and stable sustainment of a steady-state plasma with high bootstrap current is discussed.

Hot ions with a strongly anisotropic distribution are created by radio-frequency waves and are sustained in the GAMMA 10 tandem mirror. The main power drain from the hot ions is accounted for by classical transport processes. The dominant loss is electron drag, which is confirmed in auxiliary electron heating experiments. The anisotropic distribution of hot ions is also examined in terms of neutron yield from D - D fusion reactions. The ion pressure and the anisotropy drive unstable waves near the ion cyclotron frequency. A dispersion relation and spatial distributions of the unstable waves are examined by use of magnetic probes and reflectometry cross-polarization scattering measurements. The hot-ion plasma contains a multi-shell structure of impurities creating concentric layers of pressure-driven azimuthal velocity with strong shears.

Results of self-organization of the magnetic field and associating plasma loss are reviewed in the reversed field pinch (RFP) experiments on TPE machines. It is shown that the RFP plasma has a strong tendency to relax to a certain magnetic configuration similar to the energy minimum state predicted by Taylor. Thus the RFP configuration is self-organized and self-sustained with appropriate control of experimental conditions. Interestingly, however, it is observed in some cases that the relaxation can take place without the conservation of total magnetic helicity.

The mechanisms of the self-organization and associated loss are discussed in some detail. In the low pinch parameter region (< 1.6), the self-organization is continuous and the main loss mechanism seems to be the electron motion along the stochastic magnetic field line caused by the overlapping of multiple modes of the magnetic fluctuations being excited simultaneously. In the high pinch parameter region, self-organization is characterized by the pulse-like relaxation (intermittent in many cases), which seems to destroy the magnetic surface in a certain position of the torus and causes rapid loss of plasma energy. Attempts to improve RFP confinement by controlling the relaxation are described. One example is the improved high-theta mode in TPE-1RM20 and another is the pulsed poloidal current drive in MST.

The irradiation uniformity of the GEKKO XII laser has been improved by implementing the state-of-the-art beam smoothing and laser control techniques in the system. Non-uniformity modes higher than 20 are improved to less than 1% RMS by introducing partially coherent light, while the lower modes are improved to 3 - 4% RMS. Implosion experiments were carried out with the new GEKKO XII: a fast-rise square pulse preceded by a pre-pulse was used to suppress the Rayleigh - Taylor instability at the acceptable expense of implosion efficiency. The experimental results are compared with a pusher-fuel mixing model combined with a one-dimensional hydrodynamic simulation. The degradation of experimentally obtained neutrons is mainly attributed to the lower-mode asymmetry. Recent results of planar target experiments on hydrodynamic instabilities are also presented. Particular emphasis is given to the noticeable reduction of fluid instability by the flushing of an external x-ray over the target surface prior to the main drive pulse, leading to suppression of the initial imprint amplitude by the pre-formed plasma

Operation of helicon discharges at magnetic fields B₀ below 100 G is of interest for plasma etching and deposition reactors if high ion flux can be maintained with reduced field requirements. The theory of coupled helicon and Trivelpiece - Gould modes is summarized for uniform B₀. Initial results from two experiments are reported. The first has a single 5 cm diameter tube with B₀=0-100G injecting plasma into a field-free region. The second contains a two-dimensional array of seven such tubes covering a large area. Densities and density profiles are measured for various fields, RF powers and gas pressures. The highest density generally occurs at zero field. Because of the non-uniformity in B₀, direct comparison with theory cannot yet be made.

Laboratory experiments on waves in dusty plasmas performed at the University of Iowa are described. All of the experiments deal with low-frequency electrostatic waves and include studies of (i) electrostatic dust ion cyclotron (EDIC) waves, (ii) dust ion acoustic (DIA) waves and (iii) dust acoustic (DA) waves.

The film growth process of hydrogenated amorphous silicon (a-Si:H) from a monosilane glow-discharge plasma as been investigated as a representative case of thin film growth from a glow-discharge plasma. On the basis of our understanding of the gas phase as well as surface-reaction processes obtained from the results of a variety of process diagnoses such as infrared laser-absorption spectroscopy and infrared reflection - absorption spectroscopy, a reaction model explaining the determination step of defect density in the resulting a-Si:H has been proposed. The defect (dangling bond) density of a-Si:H has been controlled through the surface-diffusion length of silyl radicals produced in the monosilane plasma. This allows us to obtain low defect density a-Si:H having a range of optical band gaps and to avoid the use of alloying with elements other than hydrogen. At high temperature, precursor assisted defect suppression (PADS) produces a low band gap and defect density. At low temperature, heating of precursors in the gas phase allows defect reduction by energized precursors (DREP) to attain lower defect density. Furthermore, the electron and hole mobilities in the resulting a-Si:H have been controlled by making use of the kinetic energy of ionic species impinging on the film-growing surface from the plasma.

ICPs have become one of the dominant high-density plasma sources for etching. More variations of ICP etch sources are currently being developed and inductive systems for high-density plasma deposition are also being developed. In addition, helicon inductive sources are being developed in single loop and spiral antenna configurations. In the future these two fields may merge to combine the optimum features of both systems. The physics of low-pressure ICP systems will be discussed. Experimental effects of multipole magnetic fields near the rf coil will be described as they relate to low pressure and electronegative gases.

Large-diameter high-density plasmas such as electron-cyclotron-resonance (ECR), helicon wave, inductively coupled, and surface-wave plasmas are currently being developed for plasma-assisted thin-film processes in the next generation. Actual applications of such high-density plasmas require a deeper understanding of discharge physics as well as advanced techniques for plasma control. In this paper, new findings on antenna - plasma couplings are reported. One is resonant directional excitation of helicon waves and a mechanism of density jump in a helicon RF discharge. The other is the identification of long-wavelength surface-wave modes with observation of the short-wavelength mode in a planar microwave discharge. In addition, comprehensive diagnostics of a pulsed inductively coupled plasma in chlorine is presented, which explains the pronounced effect of pulsed power discharges on charge-up suppression.