Table of contents

The full text of this preface is available in the PDF.

The full text of this introduction is available in the PDF.

The wire-array z-pinch has in a very short time achieved remarkable performance as a powerful (>200 TW), pulsed soft x-ray source of high efficiency (∼15%) and of great relevance to inertial confinement fusion. The underlying physics involves the transformation of wire cores to a plasma corona, the occurrence of uncorrelated axial instabilities, inward flowing low magnetic Reynolds number jets, sometimes an accumulated stable and dynamically confined precursor column, an almost constant velocity implosion when gaps occur in the wire cores and finally at stagnation a fast-rising soft x-ray pulse of typically 5 ns FWHM. Nested arrays improve the performance and can operate in several modes. Three hohlraum designs have been tested; one of these, the dynamic hohlraum, has achieved a radiation temperature of ∼230 eV and has compressed a capsule from 2 to ∼0.8 mm diameter with a neutron yield of > 10¹⁰ thermal DD neutrons. Lower mass stainless steel wire arrays are used for K_α radiation sources. Generally implosions lead to more energy radiated than the implosive kinetic energy, and this is hypothesized as being due to ion viscous heating, as fast-growing short wavelength nonlinear MHD instabilities are dissipated; record ion temperatures of 200–300 keV are predicted and have been measured for the stainless steel array on Z at Sandia.

The interplay between sheared E × B mean flows and turbulence is reviewed while highlighting the common features observed in three different magnetic configurations: tokamak, stellarator and reversed field pinch. By comparing results from different experiments spanning a wide range of parameters, a universal relationship between the small scales of plasma turbulence and the large scales of the plasma mean flow is observed. In particular the generation of E × B flows by plasma turbulence in the edge region of fusion experiments is addressed by discussing the driving and damping action exerted by turbulence. In this context recent measurements of Reynolds stress and estimates of plasma viscosity are presented. The energy transfer from turbulence to mean flow is also addressed on presenting the latest experimental results. Finally the findings are discussed highlighting the elements supporting a turbulence self-regulation process.

A two-dimensional array of saddle coils at M_c poloidal and N_c toroidal positions is used on the EXTRAP T2R reversed-field pinch (Brunsell P R et al2001 Plasma Phys. Control. Fusion43 1457) to study active control of resistive wall modes (RWMs). Spontaneous growth of several RWMs with poloidal mode number m = 1 and different toroidal mode number n is observed experimentally, in agreement with linear MHD modelling. The measured plasma response to a controlled coil field and the plasma response computed using the linear circular cylinder MHD model are in quantitive agreement. Feedback control introduces a linear coupling of modes with toroidal mode numbers n, n' that fulfil the condition |n − n'| = N_c. Pairs of coupled unstable RWMs are present in feedback experiments with an array of M_c × N_c = 4 × 16 coils. Using intelligent shell feedback, the coupled modes are generally not controlled even though the field is suppressed at the active coils. A better suppression of coupled modes may be achieved in the case of rotating modes by using the mode control feedback scheme with individually set complex gains. In feedback with a larger array of M_c × N_c = 4 × 32 coils, the coupling effect largely disappears, and with this array, the main internal RWMs n = −11, −10, +5, +6 are all simultaneously suppressed throughout the discharge (7–8 wall times). With feedback there is a two-fold extension of the pulse length, compared to discharges without feedback.

Using resonant magnetic perturbations with toroidal mode number n = 3, we have produced H-mode discharges without edge localized modes (ELMs) which run with constant density and radiated power for periods up to about 2550 ms (17 energy confinement times). These ELM suppression results are achieved at pedestal collisionalities close to those desired for next step burning plasma experiments such as ITER and provide a means of eliminating the rapid erosion of divertor components in such machines which could be caused by giant ELMs. The ELM suppression is due to an enhancement in the edge particle transport which reduces pedestal current density and maximum edge pressure gradient below the threshold for peeling–ballooning modes. These n = 3 magnetic perturbations provide a means of active control of edge plasma transport.

The influence of the magnetic topology on transport and stability has been investigated in four stellarators: an almost shearless medium size flexible heliac (TJ-II), a medium size and a large heliotron (CHS and LHD) with shear, and a quasihelically symmetric device (HSX) with moderate shear. All of these have variable rotational transform profiles and magnetic ripples. Using these capabilities, bifurcated states can appear and plasma can jump from one to another with subsequent changes in the transport properties. Low rational values of ι/2π can create transport barriers in LHD and TJ-II when they are located close to the plasma core or at the edge. The key ingredient for transport barriers is a positive and sheared electric field. Internal transport barriers also appear in CHS, but the role of rationals is not clear yet in this device. The time evolution of the electric field shows the onset of a bifurcation triggered either by the rational or by the presence of the ion and electron roots. The electric potential inside ITBs follows the ECE-temperature profile in a fast time scale. The plasma stability properties and its effect on the viscosity are also studied in the HSX, and the influence of the dynamics of rational surface is studied in the LHD and TJ-II stellarators.

Global gyrokinetic particle simulations and nonlinear gyrokinetic theory indicate that electron temperature gradient (ETG) instability saturates via nonlinear toroidal coupling. In such nonlinear interactions, the wave energy at the unstable high toroidal-mode number domain cascades towards the more stable lower toroidal-mode number domain via scatterings off the driven low-mode number quasi-modes. During the saturation process, there is little zonal flow generation and the radial fluctuation envelopes maintain extended structures. The nonlinear coupling process depends critically on the toroidal geometry and, as such, represents a new paradigm for the spectral cascade of drift wave turbulence in toroidal systems.

The confinement of low temperature, non-equilibrium plasmas to cavities having characteristic spatial dimensions <1 mm is providing new avenues of inquiry for plasma science. Not only is a previously unexplored region of parameter space now accessible, but the interaction of the plasma with its material boundaries raises fascinating questions and opportunities. Other scientific issues that come to the fore include scaling relationships and the collisional processes that become prevalent in a high pressure environment. The general characteristics of microplasmas, as well as several emerging applications, are briefly described here. With regard to the latter, emphasis will be placed on photonics and, specifically, the demonstration of large (500 × 500) arrays of microcavity plasma devices in Si, the observation of photodetection in the visible, near-infrared and ultraviolet by a microplasma, and the measurement of optical gain in the blue (λ ∼ 460 nm) from a linear array of microplasmas in a ceramic structure.

Long pulse, multi-MW operation on Tore Supra, with active control of the main plasma parameters and new diagnostics, has allowed the investigation of several important issues relevant to the physics of steady-state plasmas, sometimes leading to the discovery of new physics effects. Results obtained in the most recent experimental campaigns are reported. Discharges lasting several resistive times, with combined LHCD and minority ICRH at high power level (>10 MW) in plasmas close to the Greenwald limit, at T_e ∼ T_i, were realized. In these discharges, toroidal rotation velocities of the order of 50 km s⁻¹ induced by ICRH were measured by charge exchange spectroscopy. Doppler backscattering was used to complete these measurements and to provide turbulence spectra. MHD phenomena characteristic of steady-state plasmas, such as double tearing modes, were studied using a new 32-channel ECE radiometer and correlation ECE. After the discovery of stationary regimes characterized by non-linearly coupled temperature and current oscillations, a new regime was observed, characterized by spontaneous transitions between two cycles of electron temperature oscillations: low-amplitude (ΔT_e/T_e ∼ 0.1) and giant (ΔT_e/T_e > 0.5).

On the Tokamak à Configuration Variable (TCV), electron internal transport barriers (eITBs) can be formed during a gradual evolution from a centrally peaked to a hollow current profile while all external actuators are held constant. The formation occurs rapidly (<τ_eE) and locally and, according to ASTRA modelling, is consistent with the appearance of a local minimum in the safety factor (q) profile. The eITB is sustained by non-inductively driven currents (including the off-axis bootstrap current) for many current redistribution times while the current in the tokamak transformer is held constant. The maximum duration is limited by the pulse length of the gyrotrons. The transformer coil can be used as a counter (or co-) current source with negligible accompanying input power. In established eITBs the performance can be enhanced (degraded) by altering solely the central current or q-profile. New experiments show that the same stationary eITB performance can be reached starting from discharges with centrally peaked current. A fine scan in surface voltage shows a smooth increase in performance and no sudden improvement with voltage despite the fact that q_min must pass through several low-order rational values. The appearance, in some cases, of magnetic islands simultaneously with that of the eITB provides a new constraint on the simulation of the q-profiles.

Clear observations of early triggering of neo-classical tearing modes by sawteeth with long quiescent periods have motivated recent efforts to control, and in particular destabilize, sawteeth. One successful approach explored in TCV utilizes electron cyclotron heating in order to locally increase the current penetration time in the core. The latter is also achieved in various machines by depositing electron cyclotron current drive or ion cyclotron current drive close to the q = 1 rational surface. Crucially, localized current drive also succeeds in destabilizing sawteeth which are otherwise stabilized by a co-existing population of energetic trapped ions in the core. In addition, a recent reversed toroidal field campaign at JET demonstrates that counter-neutral beam injection (NBI) results in shorter sawtooth periods than in the Ohmic regime. The clear dependence of the sawtooth period on the NBI heating power and the direction of injection also manifests itself in terms of the toroidal plasma rotation, which consequently requires consideration in the theoretical interpretation of the experiments. Another feature of NBI, expected to be especially evident in the negative ion based neutral beam injection (NNBI) heating planned for ITER, is the parallel velocity asymmetry of the fast ion population. It is predicted that a finite orbit effect of asymmetrically distributed circulating ions could strongly modify sawtooth stability. Furthermore, NNBI driven current with non-monotonic profile could significantly slow down the evolution of the safety factor in the core, thereby delaying sawteeth.

The 'improved H-mode' regime, realized in ASDEX Upgrade in 1998 and confirmed by other devices, demonstrates the integration of advanced performance beyond the standard H-mode for confinement (confinement enhancement factor H_98(y,2) ⩽ 1.6), stability (normalized beta β_N ∼ 3–3.5) and, at densities close to the Greenwald density, exhaust in stationary discharges longer than 40 confinement times or up to six resistive diffusion times. The q-profile is characterized by low central magnetic shear and axis safety factor q₀ > 1 that is obtained by particular discharge and heating ramp-up scenarios and maintained via fishbones or benign higher (m, n) instabilities without using elaborate current control. Core transport is still governed by drift-wave turbulence with stiff temperature profiles, but density profiles are more strongly peaked and contribute to the increase in global confinement. A further contribution manifests itself by enhanced pressures at the edge barrier pedestal top and at the ρ = 0.9 surface both increasing with the input power. (3, 2) NTMs remain small, enabling routine operation up to β_N ∼ 3 (limited by (2, 1) NTMs) at ITER relevant collisionalities, for normalized Larmor radii down to four times the ITER value and for a broad range of q₉₅ = 3–5. Tailored heat deposition including central wave heating allows for a compromise in density peaking for enhanced confinement and tolerable high-Z impurity concentrations even with tungsten coated structures.

As far as the ITER relevance of this regime is concerned, its compatibility with significant central electron heating, low collisionality and even densities close to the Greenwald density combined with type-II ELMs and β_N ∼ 3.5 is of importance. The GLF23 turbulence model still predicts peaked density profiles (R/L_n ∼ 3) and sufficient transport to avoid impurity accumulation. At low q₉₅ ∼ 3 the fusion performance in terms of is more than doubled compared with the ITER baseline scenario (performance factor ∼ 0.2) extrapolating to long Q ≫ 10 pulses on ITER. At medium q's bootstrap current fractions up to 50% and performance factors close to 0.2 can be achieved resulting in long inductive pulse lengths of ∼1 h allowing ITER 'hybrid' operation at Q ⩽ 9.

Non-inductive current drive causes cross-field neoclassical transport in a tokamak, in much the same way that the toroidal electric field used to drive the plasma current produces the so-called Ware pinch. This transport can be either inwards or outwards, depending on the current drive mechanism, and can be either larger or smaller than the analogous Ware pinch. A Green's function formalism is used to calculate the transport produced by wave-driven currents, which is found to be inwards for electron–cyclotron and lower-hybrid current drive. Its magnitude is proportional to the collisionality of the current-carrying electrons and therefore smaller than the Ware pinch when the resonant electrons are suprathermal. In contrast, neutral-beam current drive produces outward particle transport when the beams are injected in the same toroidal direction as the plasma current, and inward particle transport otherwise. This transport is somewhat larger than the corresponding Ware pinch. Together, they may explain an observation made on several tokamaks over the years, most recently on MAST, that density profiles tend to be more peaked during counter-injection.

The ideal MHD model of peeling–ballooning modes for the onset of the edge localized mode (ELM) is reviewed with some of the previous results that lead to the present understanding of the MHD stability of the edge transport barrier. Extensions to the ideal MHD theory are discussed followed by recent developments in the observations, theory and simulations of the non-linear phase of the ELM and the peeling and ballooning modes.

Current diffusion, heat transport modelling, and linear gyrokinetic stability analysis have been carried out on a set of seven hybrid discharges from AUG, DIII-D, JET and JT-60U, in order to gain better understanding of the physics underlying this promising candidate scenario for ITER. Within this dataset, the GLF23 model has a higher accuracy than the Weiland model in predicting the temperature profiles in the region 0.3 < ρ < 0.8. The core heat transport appears to be similar between hybrid discharges and standard H-modes, and also among hybrid discharges with very different H factors. Projections to ITER show that Q = 10 can be obtained with the hybrid scenario using an alternative scaling without β degradation. However, additional off-axis current drive and current profile control might be needed for the ITER hybrid scenario, in order to achieve its full potential for high β_N on extended duration.

Nonlinear and quasilinear gyrokinetics are used in tandem to address two key open questions in the area of turbulent transport in magnetized fusion plasmas. These are, first, the qualitative and quantitative properties of electron thermal transport caused by trapped electron modes and, second, the existence and nature of an anomalous particle pinch. Both of these issues are examined in a multispecies, fully kinetic framework.

Plasma operation with high-Z plasma facing components is investigated with regard to sputtering, core impurity contamination and scenario restrictions. A simple model based on dimensionless quantities for fuel and high-Z ion sources and transport to describe the high-Z concentration in the plasma core is introduced. The impurity release and further transport is factorized into the sputtering yield, the relative pedestal penetration probability and a core confinement enhancement factor. Since there are quite large uncertainties, in particular, in the sputtering source and the edge transport of high-Z impurities, very different scenarios covering a wide parameter range are taken into account in order to resolve the experimental trends. Sputtering of tungsten by charge exchange neutrals in the energy range 0.5–2 keV is comparable to the effect of impurity ion sputtering, while the impact of thermal fuel ions is negligible. Fast ions produced by neutral beam injection as well as sheath acceleration during ICR heating may cause considerable high-Z sources if the limiters on the low-field side have high-Z surfaces. The critical behaviour of the central high-Z concentration in some experimental scenarios could be attributed to edge and core transport parameters in the concentration model. The improved H-mode with off-central heating turns out to be the most critical one, since a hot edge is combined with peaked density profiles.

A key role in the evolution of the neoclassical tearing modes (NTMs) belongs to the radial profiles of the perturbed plasma flow, temperature and density which are determined by the conjunction of the longitudinal and cross-field transport arising from thermal conduction, particle diffusion and viscosity. In a tokamak, the perpendicular transport of particles, heat and momentum is typically anomalous. In this paper the results of theoretical studies on the influence of anomalous perpendicular heat transport and anomalous ion perpendicular viscosity on early stages of NTM evolution are presented. Several parallel transport mechanisms competitive with anomalous cross-island heat transport in the formation of the perturbed electron and ion temperature profiles within the island are considered. The perturbed electron temperature profile is established in competition between anomalous perpendicular electron heat conductivity and parallel electron heat convection. The formation of the ion perturbed temperature profile was found to be dependent on the island rotation frequency. The perpendicular ion heat conductivity is balanced by the parallel transport associated with the ion inertia for an island rotating with subsonic frequency or with island rotation with respect to the plasma for supersonic islands. The partial contributions from the plasma electron and ion temperature perturbations in the bootstrap drive of the mode and magnetic curvature effect were taken into account in construction of a generalized transport threshold model of NTMs. This model gives more favourable predictions for NTM stability and qualitatively modifies the scaling law for β_onset. The anomalous perpendicular ion viscosity is shown to modify the collisionality dependence of the polarization current effect, reducing it to the low collisionality limit. In its turn a viscous contribution to the bootstrap drive of NTMs is found to be of the same order as a conventional bootstrap drive for the islands of width close to the characteristic one of the transport threshold model. A viscous contribution to the perturbed bootstrap current is destabilizing for the island rotating in the ion diamagnetic drift direction. In this case, an alternative threshold mechanism should be considered.

The concept of the dynamic ergodic divertor (DED) is based on plasma edge ergodization by a resonant perturbation. Such a divertor concept is closely related to helical or island divertors in stellerators. The base mode of the DED perturbation field can be m/n = 12/4, 6/2 or 3/1. The 3/1 base mode with its deep penetration of the perturbation field provides the excitation of tearing modes. This topic was presented elsewhere. In this contribution we concentrate on the divertor properties of the DED. We report on the characterization of the topology, transport properties in ergodic fields, impurity transport and density limit behaviour.

The 12/4 base where the perturbation is restricted to the plasma edge is suitable for divertor operation. With increasing perturbation field island chains are built up at the resonance layers. Overlapping islands lead to ergodization. The plasma is guided in the laminar region via open field lines of short connection length to the divertor target. The magnetic topology is not only controlled by the coil current but especially by the edge safety factor. For appropriate edge safety factor we observe a strong temperature drop in the plasma edge, indicating an expansion of the laminar region, which is necessary to decouple the divertor plasma from the core plasma. The modifications of the magnetic topology can be directly seen, for example, from carbon emission lines. The magnetic structure is calculated by the ATLAS code and shows good agreement with the experimental findings.

JET's recent experimental programme proved that 'burning plasma diagnostics', i.e. neutron, alpha particle, He ash, and fuel mixture measurements, can provide very useful information about crucial physical aspects of great reactor relevance. First of all, several of these diagnostics can improve the diagnostic capability of the ion fluid significantly. During TTE spatially resolved neutron measurements at JET were essential in obtaining the isotopic composition and the transport of the hydrogen isotopes, allowing a direct comparison between the measured transport coefficients and the neoclassical theory. The neutron emission profiles can also give crucial indications for assessing the merits of various heating schemes and their current drive capability. Neutron spectroscopy in its turn provides a clear and direct measurement of the temperature and the velocity distribution of the fuel ions. For example, the dependence of the toroidal velocity from the ion cyclotron radiofrequency heating phasing was clearly seen during TTE. The requirements of accurate neutron measurements are also promoting considerable research in detector technology, in particular in the fields of compact spectrometers and solid state detectors. 'Burning plasma' diagnostics can also strongly contribute to the physics of energetic particles and their interaction with the main plasma. γ-Ray spectroscopy is now an established method to determine the spatial localization, to visualize the trajectories of the alpha particles and the fast deuterons and to obtain estimates of their slowing down. A completely new method to detect the energetic particles, exploiting the line intensity ratio of extreme ultraviolet radiation emitted by suitable extrinsic impurities, is also being pursued. This technique allows investigating the energy range below 600 keV, extremely interesting for the study of wave–particle interactions.

Recent experiments (Synakowski et al2004 Nucl. Fusion43 1648, Lloyd et al2004 Plasma Phys. Control. Fusion46 B477) on the Spherical Tokamak (or Spherical Torus, ST) (Peng 2000 Phys. Plasmas7 1681) have discovered robust plasma conditions, easing shaping, stability limits, energy confinement, self-driven current and sustainment. This progress has encouraged an update of the plasma conditions and engineering of a Component Test Facility (CTF), (Cheng 1998 Fusion Eng. Des.38 219) which is a very valuable step in the development of practical fusion energy. The testing conditions in a CTF are characterized by high fusion neutron fluxes Γ_n ≈ 8.8 × 10¹³ n s⁻¹ cm⁻² ('wall loading' W_L ≈ 2 MW m⁻²), over size-scale >10⁵ cm² and depth-scale >50 cm, delivering >3 accumulated displacement per atom per year ('neutron fluence' >0.3 MW yr⁻¹ m⁻²) (Abdou et al 1999 Fusion Technol.29 1). Such conditions are estimated to be achievable in a CTF with R₀ = 1.2 m, A = 1.5, elongation ∼3, I_p ∼ 12 MA, B_T ∼ 2.5 T, producing a driven fusion burn using 47 MW of combined neutral beam and RF heating power. A design concept that allows straight-line access via remote handling to all activated fusion core components is developed and presented. The ST CTF will test the lifetime of single-turn, copper alloy centre leg for the toroidal field coil without an induction solenoid and neutron shielding and require physics data on solenoid-free plasma current initiation, ramp-up to and sustainment at multiple megaampere level. A systems code that combines the key required plasma and engineering science conditions of CTF has been prepared and utilized as part of this study. The results show high potential for a family of relatively low cost CTF devices to suit a range of fusion engineering and technology test missions.

A tokamak plasma with internal transport barriers (ITBs) is the best candidate for a steady ITER operation, since the high energy confinement allows working at plasma currents (I_p) lower than the reference scenario. To build and sustain an ITB at the ITER high density (⩾10²⁰ m⁻³) and largely dominant electron (e⁻) heating is not trivial in most existing tokamaks. FTU can instead meet both requests, thanks to its radiofrequency heating systems, lower hybrid (LH, up to 1.9 MW) and electron cyclotron (EC up to 1.2 MW). By the combined use of them, ITBs are obtained up to peak densities n_e0 > 1.3 × 10²⁰ m⁻³, with central e⁻ temperatures T_e0 ≈ 5.5 keV, and are sustained for as long as the heating pulse is applied (>35 confinement times, τ_E). At n_e0 ≈ 0.8 × 10²⁰ m⁻³T_e0 can be larger than 11 keV. Almost full current drive (CD) and an overall good steadiness is attained within about one τ_E, 20 times faster than the ohmic current relaxation time. The ITB extends over a central region with an almost flat or slightly reversed q profile and q_min ≈ 1.3 that is fully sustained by off-axis lower hybrid current drive. Consequent to this is the beneficial good alignment of the bootstrap current, generated by the ITB large pressure gradients, with the LH driven current. Reflectometry shows a clear change in the turbulence close to the ITB radius, consistent with the reduced e⁻ transport. Ions (i⁺) are significantly heated via collisions, but thermal equilibrium with electrons cannot be attained since the e⁻–i⁺ equipartition time is always 4–5 times longer than τ_E. No degradation of the overall ion transport, rather a reduction of the i⁺ heat diffusivity, is observed inside the ITB. The global confinement has been improved up to 1.6 times over the scaling predictions. The ITB radius can be controlled by adjusting the LH power deposition profile that is affected mostly by the q value of the discharge, while the ITB strength can be varied through central EC heating. FTU experiments have shown that ITER-like e-ITBs are achievable.

The issue of first wall and divertor target lifetime represents one of the greatest challenges facing the successful demonstration of integrated tokamak burning plasma operation, even in the case of the planned next step device, ITER, which will run at a relatively low duty cycle in comparison to future fusion power plants. Material erosion by continuous or transient plasma ion and neutral impact, the susbsequent transport of the released impurities through and by the plasma and their deposition and/or eventual re-erosion constitute the process of migration. Its importance is now recognized by a concerted research effort throughout the international tokamak community, comprising a wide variety of devices with differing plasma configurations, sizes and plasma-facing component material. No single device, however, operates with the first wall material mix currently envisaged for ITER, and all are far from the ITER energy throughput and divertor particle fluxes and fluences. This paper aims to review the basic components of material erosion and migration in tokamaks, illustrating each by way of examples from current research and attempting to place them in the context of the next step device. Plans for testing an ITER-like first wall material mix on the JET tokamak will also be briefly outlined.

Spherical tokamaks (STs) have attractive features for fusion, and there is considerable interest in understanding their transport properties which depend on the underlying microinstabilities. STs are capable of operation with low magnetic fields and exhibit large inhomogeneity in the toroidal magnetic field. These factors strongly affect particle dynamics and the potency of magnetic perturbations, which correspondingly impact on the microstability properties of STs. This paper reviews previous microstability studies in ST plasma configurations and presents gyrokinetic microstability calculations for a range of ST equilibria, using the gyrokinetic code GS2. Microstability properties of L-mode and H-mode equilibria, from the MAST experiment at Culham, are compared. In MAST the shearing rates of equilibrium E × B flows usually exceed the growth rates of microinstabilities with k_⊥ρ_i < 1 (including ion temperature gradient, ITG, driven drift waves) and are generally smaller than the growth rates of shorter wavelength modes with k_⊥ρ_i > 1 (electron temperature gradient, ETG, driven drift waves). Electromagnetic effects are significant at mid-radius in these MAST equilibria, where the local β ⩾ 0.1. At k_⊥ρ_i < 1, strongly electromagnetic modes dominate over ITG instabilities, and these modes are found to have tearing parity in the H-mode plasma and twisting parity in the L-mode plasma. Numerical experiments have been carried out to assess the properties of the tearing parity modes and to probe the underlying physical drive mechanism. At shorter wavelengths the electromagnetic effects can significantly stabilize the ETG instabilities. Nonlinear electron scale microturbulence calculations for two surfaces of a MAST H-mode plasma suggest that significant electron heat transport can be carried via this mechanism. In an extremely high β ST equilibrium, which has been proposed as the basis of a conceptual ST power plant, electrostatic instabilities are fully stabilized, but tearing parity modes are predicted to be unstable over-wide range of length scales.

Recent experimental results are reported on control issues involved in long timescales and enhanced performance in JT-60U. The control issues in neoclassical tearing mode (NTM) suppression in the weak shear plasma regime include background optimization through decreasing β_p(L_q/L_p) at the rational surface and active stabilization of NTMs using ECCD. By optimizing β_p(L_q/L_p), a condition of β_N ∼ 2.5 was sustained for 10 times the current profile relaxation time and one of β_N ∼ 2.4 with q_min ∼ 1.5 was sustained for 2.8 times the current profile relaxation time, with nearly full non-inductive current drive. In addition, a condition of β_N ∼ 3 was sustained for 5.5 s through stabilization of NTMs using ECCD, and an EC driven current nearly equal to the bootstrap current was required for complete stabilization. In the reversed shear plasma regime, the issue is the existence of the steady state solution with a large f_BS value. By controlling the pressure gradient at the internal transport barrier through toroidal rotation to avoid the disruption, a large f_BS value of approximately 75% was sustained for 2.7 times the current profile relaxation time, with nearly full non-inductive current drive, and a steady-state solution with a large f_BS value is confirmed. The control issues for the edge pedestal and edge localized modes (ELMs) are control of the pedestal pressure and the energy loss through ELMs. The pedestal pressure increases by >40% through the change in toroidal rotation. The type of ELM can be controlled by toroidal rotation from type-I to grassy.

The roadmap to a feasible fusion reactor based on the tokamak line is already established. However, options for further improvement are very desirable in order to be ready to meet the future requirements of the energy market. In this respect, stellarators have a significant role to play as candidates for steady state operation reactors.

Like tokamaks, stellarators are toroidal confining devices but they show two fundamental differences: the fact that the confining magnetic field is generated by external coils; and the lack of toroidal symmetry. The history of tokamaks shows that from the original concept, solutions have converged into a given range of configurations, and, in fact, all large tokamak are relatively similar. On the other hand most of the stellarators are apparently very different (coil structure, plasma shape, size, ...).

The advantage of the higher dimensionality of stellarators has been perceived as a shortcoming, but recent developments, both theoretical and computational, have permitted us to develop improved concepts and design new stellarators with outstanding physics properties. Moreover, most of the devices presently under construction are stellarators.

This work is devoted to discuss, firstly, the present role of stellarators in the understanding of basic physical processes in fusion devices, secondly the role of future devices, based on different concept improvements, which will significantly expand the stellarator plasma parameter range, and finally, the potential of the concept as a fusion reactor.

The recent success in coupling lower hybrid (LH) waves in high performance plasmas at JET together with the first demonstration on FTU of the coupling capability of the new passive active multijunction launcher removed major concerns on the possibility of using LH on ITER. LH exhibits the highest experimental current drive (CD) efficiency at low plasma temperature thus making it the natural candidate for off-axis CD on ITER where current profile control will help in maintaining burning performance on a long-time scale. We review recent LH results: long internal transport barrier obtained in JET with current profile sustained and controlled by LH acting under real time feedback together with first LH control of flat q-profile in a hybrid regime with T_e ∼ T_i. Minutes long fully non-inductive LH driven discharges on Tore Supra (TS). High CD efficiency with electron cyclotron in synergy with LH obtained in FTU and TS opening the possibility of interesting scenarii on ITER for MHD stabilization. Preliminary results of LH modelling for ITER are also reported. A brief overview of ITER LH system is reported together with some indication of new coming LH experiments, in particular KSTAR where CW klystrons at the foreseen ITER frequency of 5 GHz are being developed.

Over the past few years, the emphasis in heavy ion target design has moved from the distributed radiator target to the 'hybrid' target because the hybrid target allows a larger beam focal spot than the distributed radiator (∼5 mm radius rather than ∼2 mm radius). The larger spot relaxes some of the requirements on the driver, but introduces some new target physics issues. Most notable is the use of shine shields and shims in the hohlraum to achieve symmetry rather than achieving symmetry by beam placement.

The shim is a thin layer of material placed on or near the capsule surface to block a small amount of excess radiation. While we have been developing this technique for the heavy ion hybrid target, the technique can also be used in any indirect drive target. We have begun testing the concept of a shim to improve symmetry using a double-ended z-pinch hohlraum on the Sandia Z-machine. Experiments using shimmed thin wall capsules have shown that we can reverse the sign of a P₂ asymmetry and significantly reduce the size of a P₄ asymmetry. These initial experiments demonstrate the concept of a shim as another method for controlling early time asymmetries in ICF capsules.

The progress of the construction of very large laser facilities LMJ and NIF enables the prediction of inertial fusion achievement. These facilities will open new fields for research: the high energy density physics. Pressures of several 100 Mbars and temperatures of several 100 eV will be reached. Measurements of material properties (EOS and opacities) which have been demonstrated on current or former facilities will be possible at these never reached conditions. Pure hydrodynamics (instabilities) and radiative hydrodynamics astrophysical issues will be addressed.

However, ignition and gain as a first proof of Inertial Confinement Fusion is a primary goal. The indirect drive route to inertial fusion has been prepared for many years by CEA (Commissariat à l'Energie Atomique). The last ten years were imprinted by a close collaboration between CEA and US-DOE in both the areas of facilities R&D and ignition target physics.

The scientific issues are well known: the propagation of laser light through the very long plasma created inside the hohlraum has to be understood and mastered to be sure that less than 10% of laser energy will be backscattered by parametric instabilities. On the other hand, the stability of the capsule implosion has to be matched with the fabrication surface finish so as to avoid shell destruction and extinction of the central hot spot. Recent advances at CEA have allowed a better confidence of reaching ignition using the facility previously specified. These works used the CEA computing capability combined with plasma experiments on existing lasers facilities. Ignition achievement also supposes the realization of suitable cryogenic targets.

CEA began the construction of the Laser Megajoule (LMJ), a 240-beam laser facility, at the CEA Laboratory CESTA near Bordeaux. The LMJ is designed to deliver 2 MJ of 0.35 µm light to targets for high energy density physics experiments.

Four beams were operated for plasma experiments on the Ligne d'Integration Laser (LIL) at CESTA, for the end of 2004, meeting the specifications for LMJ. The realization phase of the LMJ facility was initiated in March 2003 with the start of construction of the building and the target chamber.

Recently the first laser–plasma interaction and hohlraum experiments have been performed at the National Ignition Facility (NIF) in support of indirect drive inertial confinement fusion designs. The effects of laser beam smoothing by spectral dispersion and polarization smoothing on the intense (2 × 10¹⁵ W cm⁻²) beam propagation in gas-filled tubes has been studied at up to 7 mm plasma scales as found in indirect drive gas filled ignition hohlraum designs. These experiments have shown the expected full propagation without filamentation and beam break up when using full laser smoothing. In addition, vacuum hohlraums have been irradiated with laser powers up to 6 TW, 1–9 ns pulse lengths and energies up to 17 kJ to activate several diagnostics, to study the hohlraum radiation temperature scaling with the laser power and hohlraum size, and to make contact with hohlraum experiments performed at the Nova and Omega laser facilities. Subsequently, novel long laser pulse hohlraum experiments have tested models of hohlraum plasma filling and long pulse hohlraum radiation production. The validity of the plasma filling assessment using in analytical models and radiation hydrodynamics calculations with the code LASNEX has been proven in these studies. The comparison of these results with modelling will be discussed.

High-energy–density systems and astrophysical systems both involve hydrodynamic effects, including sources of pressure, shock waves, rarefactions and plasma flows. In the evolution of such systems, hydrodynamic instabilities naturally evolve. As a result, a fundamental understanding of hydrodynamic instabilities is necessary to understand their behaviour. This paper discusses the validity of a hydrodynamic description in both cases, and, from the common perspective of the basic mechanisms at work, discusses the instabilities that appear in astrophysics and at high energy density. The high-energy–density research facilities of today, built to pursue inertial fusion, can accelerate small but macroscopic amounts of material to velocities above 100 km s⁻¹, can heat such material to temperatures above 100 eV and can produce pressures far above a million atmospheres (10¹² dyn cm⁻² or 0.1 TPa). In addition to enabling inertial fusion research, this enables these facilities to do experiments under the conditions that address basic physics issues including issues from astrophysics. One can devise experiments aimed directly at important processes such as the Rayleigh Taylor instability at an ablating surface or at an embedded interface that is accelerating, the Richtmeyer Meshkov evolution of shocked interfaces and the Kelvin–Helmholtz instability of shear flows. The paper includes examples of such phenomena from the laboratory and from astrophysics.

In the last few years, high power lasers have demonstrated the possibility to explore a new state of matter, the so-called warm dense matter. Among the possible techniques utilized to generate this state, we present the dynamic compression technique using high power lasers. Applications to planetary cores material (iron) will be discussed. Finally new diagnostics such as proton and hard-x-ray radiography of a shock propagating in a solid target will be presented.

The interaction of high intensity laser pulses with plasmas is an efficient source of megaelectronvolt ions. Recent observations of the production of directional energetic ion 'beams' from the front and rear surfaces of thin foil targets upon irradiation by intense laser pulses have prompted a renewed interest into research in this area. In addition, other recent observations have shown that high energy ions can be observed from intense laser interaction with low density plasma as a result of ponderomotive shock acceleration. The source characteristics and acceleration mechanisms for these ions have been extensively investigated, and there have also been a number of proposed applications for these ion beams, such as for injectors into subsequent conventional acceleration stages, for medicine, for probing of dense plasmas and for inertial confinement fusion experiments.

We present results of high energy density laboratory experiments on the production of supersonic radiatively cooled plasma jets with dimensionless parameters (Mach number ∼30, cooling parameter ∼1 and density contrast ρ_j/ρ_a ∼ 10) similar to those in young stellar objects jets. The jets are produced using two modifications of wire array Z-pinch driven by 1 MA, 250 ns current pulse of MAGPIE facility at Imperial College, London. In the first set of experiments the produced jets are purely hydrodynamic and are used to study deflection of the jets by the plasma cross-wind, including the structure of internal oblique shocks in the jets. In the second configuration the jets are driven by the pressure of the toroidal magnetic field and this configuration is relevant to the astrophysical models of jet launching mechanisms. Modifications of the experimental configuration allowing the addition of the poloidal magnetic field and angular momentum to the jets are also discussed. We also present three-dimensional resistive magneto-hydrodynamic simulations of the experiments and discuss the scaling of the experiments to the astrophysical systems.

The recent and continuous development of powerful laser systems has permitted the emergence of new approaches for generating energetic electron beams. By focusing light pulses containing a few joules of energy in a few tens of femtoseconds onto gas jets, extremely large electric fields can be generated, reaching the terravolts per metre level. Such fields are 10 000 times greater than those produced in the radio-frequency cavities of conventional accelerators. As a result, the length over which electrons extracted from the target can be accelerated to hundreds of MeV is reduced to a few millimetres. The reduction of the size and the cost of laser-plasma accelerators is a promising consequence, but these electron beams also reveal original properties, which make them a wonderful tool for science. By adjusting the interaction parameters, the electron energy distribution can be tuned from a maxwellian-like distribution to a quasi-monoenergetic one. The new properties of these laser-based particle beams are well suited to many applications in different fields, including medicine (radiotherapy), chemistry (ultrafast radiolysis), material science (non-destructive material inspection using radiography) and, of course, for accelerator physics.

We review the potential of x-ray scattering as a dense plasma diagnostic and present data taken from experiments in which x-ray scattering from dense plasmas is developed as a diagnostic tool. In one type of experiment the scattered photons are detected as a function of angle using direct detection onto a CCD chip. Such experiments are designed primarily to observe the static ion–ion structure factor, which is expected to dominate the scattering for moderate to high Z plasmas at a few electronvolts temperature. In a second type of experiment we have used a curved crystal to observe spectrally resolved x-ray scattering at a fixed angle. This experiment was designed to observe the dynamical structure factor of the plasma.

If a laser plasma generated on a slab target with a high Z is left to expand it becomes a very efficient source of highly charged ions. Depending on the parameters of the laser driver, ions with charge states from 1+ up to more than 50+ can be produced, with ion energies ranging from tens of electronvolts up to tens of megaelectronvolts, with no external acceleration. The ion current density may reach tens of milliamperes per square centimetres in a distance of 1 m from the target. They can be used either for a direct accelerator injection, for a hybrid ion source based on the coupling of a laser with an electron cyclotron resonance ion source for an easier evaporation and a pre-ionization of the target material with a subsequent charge state enhancement or for a direct ion implantation, as substrates for the implantation metallic and polymer materials are usually exposed to the laser produced ion streams with an appropriate tuning of the implantation regime to modify their surface properties. Although the interaction of the laser beam with the plasma is a fairly complex process certain fundamental phenomena have been identified based on a careful analysis of the charge–energy spectra of the outgoing ions.

Stable glow-type discharge plasmas at elevated pressures can be generated and maintained easily when the plasma is spatially confined to cavities with critical dimensions below 1 mm ('microplasmas'). We studied the properties of several atmospheric-pressure microplasmas and their use in the remediation of volatile organic compounds (VOCs) and biological decontamination. The VOCs studied include individual prototypcal aliphatic and aromatic compounds as well as mixtures such as BTEX (benzene, toluene, ethylbenzene and xylene). The biological systems under study included individual bacteria as well as bacterial biofilms, which are highly structured communities of bacteria that are very resistant to antibiotics, germicides, and other conventional forms of destruction.

A series of different discharge configurations suitable for surface treatment at atmospheric pressure is discussed, including a non-thermal modular radio frequency (13.56, 27.12 or 40.78 MHz) jet plasma.

The capacitively coupled configuration allows the operation with both rare gases (e.g. Ar) and reactive gases (N₂, air, reactive admixtures of silicon-containing compounds). Several capillaries are arranged in an array to allow plasma assisted treatment of surfaces including non-flat geometries. Optical emission spectroscopy, mass spectrometry and measurements of the axial and radial temperature profiles are used to characterize the discharge.

The surface energy of different polymer materials is significantly enhanced after plasma treatment. Many applications are possible, such as plasma activation of surfaces for adhesion control, surface cleaning, plasma enhanced CVD, plasma cleaning, plasma activation and biomedical applications.

The PK-4 experiment is a continuation of the successful dusty plasma experiments PK-1, PK-2 and PK-3 conducted on board of the orbital space stations Mir and International Space Station. For all these experiments it is important to avoid the strong influence of gravity, exerting an external stress on the system. Whereas PK-3 and PK-3 Plus experiments are using a planar rf capacitive discharge, PK-4 studies complex plasmas in a long cylindrical chamber with a combined dc/rf discharge. Such a configuration of the chamber will provide a particular advantage for investigation of different dynamical phenomena in complex plasmas such as sheared laminar flow of a highly nonideal dusty liquid and its transition to the turbulent regime, nozzle flow, boundary layers and instabilities, shock waves (solitons) formation and propagation, dust particle lane formation, and space dust grain separation according to their size.

Results are given of the experimental investigation of three-particle correlation for liquid plasma-dust structures formed in the electrode layer of a capacitive rf discharge. The obtained three-particle correlation functions for experimental and numerical data are analysed and compared with the superposition approximation. The forming of clusters of macroparticles in plasma-dust systems being analysed is revealed. The experiments in heat transfer were performed in plasma of a capacitive radio-frequency (rf) discharge in argon (P ≈ 20 Pa) with particles 4 µm in mean diameter. The results are given of an experimental investigation of processes of heat transfer for fluid dust structures in rf-discharge. The analysis of steady-state, and unsteady-state heat transfer are used to obtain the thermal conductivity and diffusivity constants.

Starting with the theory of ion generation, extraction, acceleration and focusing, the different types of gas discharges and ion sources based on them are described. Then hardware of single-beamlet sources, neutral beam injectors and ion thrusters are discussed.

The electrical characteristics of homogeneous dielectric barrier discharges in helium, argon and nitrogen are presented and discussed. From the evolution of the discharge current as a function of the voltage applied to the gas it is shown that (i) in helium and argon, during the current increase, the discharge transits from a non-self-sustained discharge to a Townsend discharge and then a subnormal glow discharge (atmospheric pressure glow discharge) (ii) in nitrogen the ionization level is too low to induce a localization of the electrical field and the glow regime cannot be achieved. The discharge is a Townsend discharge (atmospheric pressure Townsend discharge). The characteristics of this specific discharge are described including the time variation of the density of electron, ion, metastable state and electrical field.

Atmospheric pressure microwave discharge methods and devices used for producing non-thermal plasmas for control of gaseous pollutants are described in this paper. The main part of the paper is concerned with microwave torch discharges (MTDs). Results of laboratory experiments on plasma abatement of several volatile organic compounds (VOCs) in their mixtures with either synthetic air or nitrogen in low (∼100 W) and moderate (200–400 W) microwave torch plasmas at atmospheric pressure are presented. Three types of MTD generators, i.e. low-power coaxial-line-based MTDs, moderate-power waveguide-based coaxial-line MTDs and moderate-power waveguide-based MTDs were used. The gas flow rate and microwave (2.45 GHz) power delivered to the discharge were in the range of 1–3 litre min⁻¹ and 100–400 W, respectively. The concentrations of the processed gaseous pollutants were from several to several tens of per cent. The results showed that the MTD plasmas fully decomposed the VOCs at a relatively low energy cost. The energy efficiency of decomposition of several gaseous pollutants reached 1000 g (kW-h)⁻¹. This suggests that MTD plasmas can be useful tools for decomposition of highly concentrated VOCs.

Experiments were conducted to study a continuously working e-beam ionized dusty plasma at atmospheric pressure in argon and nitrogen. The formation of highly ordered dusty structures of glassy carbon spherical particles in argon was observed at a gas ionization rate of about 10¹⁴ cm⁻³ s⁻¹. A numerical simulation of charging dust particles and non-self-sustained discharge structures was performed. The dust particle trap formation in the cathode sheath of a non-self-sustained discharge was identified theoretically.

Propagation of electromagnetic waves in several types of microplasmas has been examined experimentally in a frequency range 10–75 GHz. Firstly, the fundamental characteristics of the propagation were investigated using a planar geometry of microplasma assembly, and the electron density was derived by a comparison of the transmittance with the theoretical analyses using a Drude type model with collisional effects. Secondly, an extraordinary propagation phenomenon such as the focusing effect was observed in a two-dimensional periodical microplasma array. This kind of anomalous refraction cannot be interpreted only by predictions based on the dielectric property of bulk plasma, and it is suggested that a photonic-crystal-like periodical dielectric structure may play a significant role. Thirdly, it was demonstrated that the T-junction formed by a microplasma connected to a microstrip line can control the transmission of microwaves. An attenuation (or modulation) depth of about 35% was obtained with a series of two T-junctions connected to the strip line at the right-angled corners. All the above features come from (a) the relatively high electron density of the microplasmas near 10¹³ cm⁻³, (b) the complex dispersion relation with collisional effects and (c) the spatial arrangement with a characteristic scale of the same order of the wavelength of microwaves.

Here we present a novel use of fine dust for diagnostic measurements in the sheath of a planar rf discharge and in the inertial electrostatic confinement (IEC) plasma. Since dust charge is a function of a number of plasma parameters, and the dust charge adjusts itself to changes in plasma conditions instantaneously, dust particles may be used as an ideal diagnostic tool. A major advantage of such a diagnostic approach is its simplicity as the only measurement of a position of a dust particle or its motion after a perturbation is necessary. This technique only requires access to a discharge chamber, dust particles, a laser to illuminate them and a camera to capture the motion of the dust. The dependences of the sheath profiles on the discharge pressure and power have been determined. The radial potential profiles in a cylindrically-symmetrical capacitively-coupled rf discharge and in an IEC plasma were obtained. The direction of the ion flux in the radial IEC discharge was found.

The spatially resolved cross-correlation spectroscopy (CCS) was used for systematic investigations of the barrier discharge (BD) in N₂/O₂ mixtures at atmospheric pressure. The spatio-temporal distributions of the microdischarge (MD) radiation intensities were recorded for the spectral bands of the (0–0) transitions of the 2nd positive (λ = 337.1 nm) and 1st negative system of molecular nitrogen (λ = 391.4 nm). The velocities of the cathode-directed ionization waves were evaluated from the CCS data. In the middle of the gap, the MD channel diameter was found to be about 0.3 mm and to expand towards both electrodes. On the dielectrics, outward propagating discharges were observed. A computational model of the BD is proposed, to explain the MD formation in short (1–2 mm) air gaps by a Townsend mechanism. The two-dimensional dynamics of the MD development and the channel radiation for the second positive system of nitrogen are simulated. The proposed model explains satisfactorily the experimental results on the velocity of the cathode-directed ionization wave and the MD radiation of the 2nd positive system.

The microwave (mw) plasma torch at atmospheric pressure has been studied for carbon nanotube (CNT) synthesis. The depositions were carried out on silicon substrates with 5–15 nm thin iron catalytic overlayers from the mixture of argon, hydrogen and methane. The optical emission spectroscopy of the torch showed the presence of C₂ and CH radicals as well as carbon and hydrogen excited atoms. The vicinity of the substrate influenced the relative intensities and increased the emission of C₂. For fixed mw power, the temperature of the substrate strongly depended on its position with respect to the nozzle electrode and on the gas mixture, particularly the amount of H₂. The speed of the substrate heating during an early deposition phase had a significant effect on the CNT synthesis. An abrupt increase of the temperature at the beginning increased the efficiency of the CNT synthesis. Areas of dense straight standing CNTs, 30 nm in average diameter, with approximately the same sized iron nanoparticles on their tops were found in accordance with the model of growth by plasma enhanced chemical vapour deposition. However, the deposit was not uniform and a place with only several nanometres thick CNTs grown on much larger iron particles was also found. Here, taking into account the gas temperature in the torch, 3100–3900 K, we can see similarities with the 'dissolution–precipitation' model of the CNT growth by high temperature methods, arc or laser ablation.

We review the theory of cosmic ray transport and acceleration with an emphasis on the underlying plasma physics and examine how that theory can be applied to sources such as supernova remnants and giant radio galaxies. Starting with Fermi's original model for scattering off moving magnetized clouds, we discuss quasilinear transport theory and its application to the acceleration of particles at shock fronts. We discuss problems of injection and the excitation of MHD turbulence by the accelerated particles. In the diffusive limit and at strong shocks this mechanism produces a differential energy spectrum of N(E) ∝ E⁻². Recent observations of supernova remnants suggest that their spectra may be steeper than this value. We discuss the transport and acceleration of energetic particles in highly correlated magnetic field structures. In this case particles have an enhanced probability of escape from the shock as they are trapped on field lines and the resulting spectrum is steepened up to a value of 2.5. Fast particle transport also seems to be required by observations of the structures of giant radio galaxy lobes as a function of frequency.

Upon the choice of an Eulerian observer adapted to a 3 + 1 spacetime foliation and suitable fluid and magnetic field variables, it is possible to cast the equations of both general relativistic (inviscid) hydrodynamics (GRHD) and (ideal) magneto-hydrodynamics (GRMHD) as first-order, hyperbolic systems of conservation laws for state-vectors comprising the densities of mass, momentum, energy and magnetic field components. Hyperbolicity allows the use of the flux-vector Jacobians wave structure to build up stable and accurate numerical schemes for their solution. In recent years, the so-called Godunov-type schemes, based upon approximate Riemann solvers, have been successfully extended from classical to relativistic fluid dynamics (both special and general). While such advances also hold true in the case of the MHD equations, the development still awaits here for a thorough numerical exploration. This paper reports formulations of the GRHD/GRMHD equations amenable to numerical investigations using Godunov-type schemes. A number of relevant applications in the field of relativistic astrophysics is also covered.

It is well known that the magnetic and the velocity field fluctuations in the solar wind possess many features expected of fully developed magnetohydrodynamic (MHD) turbulence. However, for frequencies higher than 0.1 Hz the situation is different and clear discrepancies have been recently observed with, for example, a steepening of the magnetic fluctuation power law spectrum. In order to give a satisfactory description of the high frequency solar wind it is necessary to adopt a new description like the electron MHD approximation in which the dynamics is entirely governed by electrons. In that context, we review and discuss recent theoretical results obtained in anisotropic whistler wave turbulence for electron MHD in the presence of a strong and uniform external magnetic field. Using helicity decomposition, the wave kinetic equations for energy and magnetic helicity are derived at the level of three-wave interactions between whistler waves. It is shown that nonlinear interactions of whistler waves transfer energy and magnetic helicity mainly in the direction perpendicular to the external magnetic field. The anisotropic turbulence thus generated has exact stationary power law solutions which scale as for the energy spectrum and for the magnetic helicity spectrum. A strong analogy is found with the problem of rotating turbulence for incompressible neutral flows which share almost all the same properties.

Spacecraft measurements in the solar wind offer the opportunity to study magnetohydrodynamic (MHD) turbulence in a collisionless plasma in great detail. We review some of the key results of the study of this medium: the presence of large amplitude Alfvén waves propagating predominantly away from the Sun; the existence of an active turbulent cascade; and the presence of intermittency similar to that in neutral fluids. We also discuss the presence of anisotropy in wavevector space relative to the local magnetic field direction. Some models suggest that MHD turbulence can evolve to a state with power predominantly in wavevectors either parallel to the magnetic field ('slab' fluctuations) or approximately perpendicular to it ('2D'). We review the existing evidence for such anisotropy, which has important consequences for the transport of energetic particles. Finally, we present the first results of a new analysis which provides the most accurate measurements to date of the wave-vector anisotropy of wavevector power in solar wind MHD turbulence.

Theoretical work aimed at understanding the electrodynamic structure of pulsar magnetospheres and winds has received new impetus from detailed imaging in the optical and x-ray wavelength bands of these objects, and especially of the Crab Nebula. This has motivated two-dimensional relativistic MHD models that have confirmed the low magnetization state of the nebular plasma. The fundamental problems this raises concerning the dissipation of energy and particle acceleration in relativistic plasmas are summarized. It is indicated how recent progress in teraelectronvolt gamma-ray astronomy may help resolve these issues by placing constraints on the sites of dissipation and acceleration in the binary system containing the pulsar PSR B1259 −63.

The history, benefits, suitability and limitations of laboratory simulation of space plasma processes are reviewed. Aspects of waves, instabilities, nonlinearities, particle transport, reconnection and hydrodynamics are addressed in terms of interrelated experiments performed in space and in the laboratory. Over time, the degree to which the interrelated experiments can be compared has increased, thanks to improved diagnostic techniques in space and closer attention to matching dimensionless space parameters in the laboratory.

Opacity experiments that have been conducted to date using long-pulse (nanosecond) high-power lasers have reached conditions similar in density and temperature to those found in the outer part of the radiative flow region of the Sun. Experiments using short-pulse (picosecond) lasers are beginning to give access to conditions halfway to the Sun's centre. In this paper we describe a preliminary design of an opacity experiment that uses a combination of long and short-pulse laser beams to compress and heat the plasma and thereby access conditions similar to those at the centre of the Sun.

Diffusion is essentially the macroscopic manifestation of random (Brownian) microscopic motion. This idea has been generalized in the continuous time random walk formalism, which under quite general conditions leads to a generalized master equation (GME) that provides a useful modelling framework for transport. Here we review some of the basic ideas underlying this formalism from the perspective of transport in (magnetic confinement) plasmas.

Under some specific conditions, the fluid limit of the GME corresponds to the Fokker–Planck (FP) diffusion equation in inhomogeneous systems, which reduces to Fick's law when the system is homogeneous. It is suggested that the FP equation may be preferable in fusion plasmas due to the inhomogeneity of the system, which would imply that part of the observed inward convection ('pinch') can be ascribed to this inhomogeneity.

The GME also permits a mathematically sound approach to more complex transport issues, such as the incorporation of critical gradients and non-local transport mechanisms. A toy model incorporating these ingredients was shown to possess behaviour that bears a striking similarity to certain unusual phenomena observed in fusion plasmas.

Plasma transport in the presence of turbulence depends on a variety of parameters such as the fluctuation level, δB/B₀, the ratio between the particle Larmor radius and the turbulence correlation length, and the turbulence anisotropy. In this paper, we present the results of numerical simulations of plasma and magnetic field line transport in the case of anisotropic magnetic turbulence, for parameter values close to those of the solar wind. We assume a uniform background magnetic field B₀ = B₀e_z and a Fourier representation for magnetic fluctuations, which includes wavectors oblique with respect to B₀. The energy density spectrum is a power law, and in k space it is described by the correlation lengths l_x, l_y, l_z, which quantify the anisotropy of turbulence. For magnetic field lines, transport perpendicular to the background field depends on the Kubo number R = (δB/B₀) (l_z/l_x). For small Kubo numbers, R ≪ 1, anomalous, non-Gaussian transport regimes (both sub- and superdiffusive) are found, which can be described as a Lévy random walk. Increasing the Kubo number, i.e. the fluctuation level, δB/B₀, or the ratio l_z/l_x, we find first a quasilinear regime and then a percolative regime, both corresponding to Gaussian diffusion. For particles, we find that transport parallel and perpendicular to the background magnetic field depends heavily on the turbulence anisotropy and on the particle Larmor radius. For turbulence levels typical of the solar wind, δB/B₀≃ 0.5–1, when the ratio between the particle Larmor radius and the turbulence correlation lengths is small, anomalous regimes are found in the case l_z/l_x ⩽ 1, with a Lévy random walk (superdiffusion) along the magnetic field and subdiffusion in the perpendicular directions. Conversely, for l_z/l_x > 1 normal Gaussian diffusion is found. A possible expression for generalized double diffusion is discussed.

Fast ignited inertially confined fusion targets have potentials for high gain at moderate laser energy. Gain estimates are based on simulations of separate aspects of target evolution and on gain models, and depend critically on ignition requirements and assumptions concerning coupling of the igniting beam to the compressed fuel. In this paper, we review and discuss ignition requirements, burn studies, and gain models. We present selected gain results, illustrating the dependence of the gain on the parameters of the ignition beam. We discuss the requirements for very large gain, as well as for substantial gain at small driver energy.

We present the results of some recent experiments performed at the LULI laboratory using the 100 TW laser facility concerning the study of the propagation of fast electrons in gas and solid targets. Novel diagnostics have been implemented including chirped shadowgraphy and proton radiography. Proton radiography images did show the presence of very strong fields in the gas probably produced by charge separation. In turn these imply a slowing down of the fast electron cloud as it penetrates in the gas and a strong inhibition of propagation. Indeed chirped shadowgraphy images show a strong reduction in time of the velocity of the electron cloud from the initial value, which is of the order of a fraction of c. We also performed some preliminary experiments with cone targets in order to verify the guiding effect and fast electron propagation in presence of the cone. Finally we compared results obtained by changing the target size.

Here we only give a first presentation and preliminary analysis of data, which will be addressed in detail in a following paper.

Integrated fast-ignition experiments on the combined OMEGA/OMEGA EP laser systems have been simulated with the multidimensional hydrodynamic code DRACO. The OMEGA laser system provides up to 30 kJ of compression energy, and OMEGA EP will provide two short-pulse beams, each with energies up to 2.6 kJ. In the electron transport model included in DRACO, the relativistic electrons are introduced at the pole of a two-dimensional (2D) simulation and transported in a straight line toward the target core. The electron's energy is calculated from the laser irradiance using a semi-empirical formula. An OMEGA cryogenic DT target designed to reach a one-dimensional fuel ρR of 0.5 g cm⁻² has been simulated in 2D, with and without non-uniformities, to assess the sensitivity to energy, timing, and irradiance of the Gaussian fast-ignitor beam. For the uniform case, the neutron yield is predicted to be in excess of 10¹⁵ (compared to ∼10¹⁴ without an ignitor beam) over a synchronization range of ∼80 ps. Implosions with the ignitor beam show little decrease in the neutron yield with increasing inner-ice non-uniformity, in contrast to implosions without the ignitor beam, which show a significant decrease in neutron yield with increasing non-uniformity.

A possible new mechanism for anomalous ion heating in ultra-intense laser plasmas is considered here. This mechanism is based on the excitation of an electron two-stream instability that is driven by the fast electron beam and resonantly decays into ion-acoustic waves. These low frequency waves are then strongly damped by the ion collisions in the dense plasma. The model gives a simple explanation for the preferential heating of the bulk ion population observed in recent laser experiments in the petawatt (PW) regime.

In this contribution, we address laser-driven electron transport in fast ignition research and the need for ultra-fast (<10 fs) pump–probe diagnostics to study, on the microscopic level, current filamentation and related anomalous energy deposition. Rates of instability growth consistent with realistic fast ignition parameters are discussed. A 3D-hybrid-PIC simulation is presented as an illustrative case. The development of 5 fs PW-range laser pulses at MPQ is reported, which may generate ultra-bright VUV-, x-ray, electron and ion beam pulses to be used for ultra-fast plasma probing.

A new ignition scheme, impact fast ignition (IFI), is studied, in which the compressed DT main fuel is to be ignited by impact with another fraction of separately imploded DT fuel, which is accelerated in the hollow conical target. The first and distinct milestone in the IFI scenario is the demonstration of such a hyper-velocity, of the order of 10⁸ cm s⁻¹. Two-dimensional hydrodynamic simulation results obtained in full geometry using plastic instead of DT fuel are presented, in which some key physical parameters for the impact shell dynamics, such as an implosion velocity of 10⁸ cm s⁻¹, a compressed density of 300–400 g cm⁻³ and a converted temperature greater than 5 keV, are demonstrated. A preliminary experimental result with a planar target is presented to show the highest velocity, 6 × 10⁷ cm s⁻¹, ever achieved.

Experimental study on energy transport in ultra-high intensity laser plasma was made. X-ray emission from a triple-layer target irradiated at 10¹⁹ W cm⁻² was observed with x-ray spectrographs, monochromatic imagers and an x-ray polarimeter to provide a temperature profile in the depth of the target, lateral extension of the heated region and the velocity distribution function of hot electrons. For PW plasma, a very shallow region (∼0.5 µm from the target surface) was heated up to 650 eV but the temperature of deeper region (up to 5 µm) was around 100 eV. These depths are much shorter than those expected from the classical penetration of the hot electrons. The localized energy deposition is also found for the plasma generated at 10¹⁷ W cm⁻², and the degree of polarization of the helium-like Cl–Heα line (1s^2 1S₀–1s2p ¹P₁) from the surface region is polarized parallel to the surface direction whereas that from a deeper region is in perpendicular to it. The experimental result is analysed using a two-dimensional Maxwellian distribution function for hot electrons. Beam-like distribution was found in the depth of plasma.

We review a recent experimental campaign to study the interaction physics of petawatt laser pulses incident at relativistic intensities on solid targets. The campaign was performed on the 500 J sub-picosecond petawatt laser at the Rutherford Appleton Laboratory. An extensive suite of optical, x-ray, and particle diagnostics was employed to characterise the processes of laser absorption, electron generation and transport, thermal and K-alpha x-ray generation, and proton acceleration.

The acceleration of ions by ultra-intense lasers has attracted great attention due to the unique properties and the unmatched intensities of the ion beams. In the early days the prospects for applications were already studied, and first experiments have identified some of the areas where laser accelerated ions can contribute to the ongoing inertial confinement fusion (ICF) research. In addition to the idea of laser driven proton fast ignition (PFI) and its use as a novel diagnostic tool for radiography the strong dependence on the electron transport in the target was found to be helpful in investigating the energy transport by electrons in fast ignitor scenarios. More recently an additional idea has been presented to use laser accelerated ion beams as the next generation ion sources, and taking advantage of the luminosity of the beams, to develop a test bed for heavy ion beam driven inertial confinement fusion physics. We review our recent experiments and simulations relevant to ICF research presenting a possible scenario for PFI as well as the prospects for next generation ion sources.

Sandia National Laboratories is developing a combination of experimental and theoretical capabilities useful for the study of fast ignition physics. Pulsed power machines such as the present Z machine have demonstrated the ability to drive inertial fusion implosions. The Z-beamlet laser is currently being used to create multi-kiloelectronvolts photons for backlighting, which can be used to diagnose the compression of deuterium/tritium to high densities. A high-energy petawatt capability is presently being added to the Z-beamlet to extend the backlighting x-ray energy up to 10–50 keV and to enable integrated fast ignition experiments. We are also developing a capability to implode capsules filled with liquid deuterium/tritium, which avoids the stringent temperature control required for β-layered capsules.

In preparation for such experiments, the theory group at Sandia is modelling various aspects of fast ignition physics. Numerical simulations of laser/plasma interaction, electron transport and ion generation are being performed. Simulations of the compression of deuterium/tritium fuel in various geometries are being performed. Analytic and numerical modelling has been performed to determine the conditions required for fast ignition breakeven scaling. These results indicate that breakeven will require about 5% of the laser energy needed for ignition and might be an achievable goal with an upgraded Z-beamlet laser in short pulse mode.

The fuel assembly of gas-filled, cone-in-shell, fast-ignitor targets is being studied experimentally at the OMEGA Laser Facility. The spherical plastic (CH) shells with a ∼24 µm wall thickness had a 70° or 35° opening-angle gold cone inserted. The targets, filled with ∼10 atm of D³He or D₂, were imploded by direct illumination with up to 21 kJ of 351 nm laser light. Some experiments used a backlighter to study the fuel assembly and mixing of cone material into the dense core. No significant mixing of the cone and core material was observed. Using both fusion products and backlit images, an areal density of ∼60 to 70 mg cm⁻² was inferred for the dense core assembly. The filling of the inside of the cone, where the ultrafast laser propagates in integrated fast-ignitor experiments, was studied using a streaked optical pyrometer. No plasma was seen inside the cone before the assembled core reached peak compression. These results are promising for successful integrated fast-ignitor experiments on the OMEGA EP Facility, scheduled to be completed in 2007.

We discuss the physical processes, which take place in a multi-component plasma set in expansion by a minority of energetic electrons. The expansion is in the form of a collisionless rarefaction wave associated with three types of electrostatic shocks. Each shock manifests itself in a potential jump and in the spatial separation of plasma species. The shock front associated with the proton–electron separation sets the maximum proton velocity. Two other shocks are due to the hot–cold electron separation and the light–heavy ion separation. They result in the light ion acceleration and their accumulation in the phase space. These structures open possibilities for control of the number and the energy spectrum of accelerated ions. Simple analytical models are confirmed in numerical simulations where the ions are described kinetically, and the electrons assume the Boltzmann distribution.

It is suggested that bulk acceleration of ions can occur in a target with density discontinuities when hot electrons are transported through the target. A foam target just belongs to such a kind of target, which is composed of irregular lamellar layers distributed randomly. To simplify the problem, we study the interaction of a high intensity laser pulse with a target consisting of regular micro-thin layers separated with a thickness of around a micrometre. Particle-in-cell simulations suggest that localized electrostatic fields with multi-peaks around the surfaces of the thin layers inside are induced when fast electrons produced are transported through such a target. These fields inhibit hot electron transport and simultaneously accelerate ions from the thin layers inside the target, forming a bulk acceleration in contrast to the surface acceleration at the front and rear sides of a thin solid target. Bulk acceleration can produce a large number of ions of moderate energy, which may be useful for applications such as fast ignition by fast protons. Experimental evidence of bulk acceleration is found with low-density foams irradiated by ultra-intense laser pulses.