Random media emerge in several applications in reactor physics and safety analysis. Most often, models of stochastic media assume spatial homogeneity, whereas real-world complex materials, such as fuel chunks resulting from core degradation, typically display apparent heterogeneities. In a series of previous works, we have shown that stochastic tessellations can be successfully used in order to describe the material properties of several classes of random media. In this paper we extend these results to the case of heterogeneous random media by using Voronoi tessellations with space-dependent seed distributions, allowing for spatial gradients.

Recent experiments indicate that controlling the propagation of high-power laser beams through millimeter long and low-density plasmas still remains challenging. In such plasma conditions, it is equally important to consider the impact of the plasma on laser propagation and laser properties, and the impact of the laser on plasma conditions. These complex phenomena are still difficult to implement in fluid models owing to the highly non-linear physics at play. Yet, electromagnetic fields prove to be good signatures of most of these low frequency phenomena. In particular, local pressure gradients and electron transport can be inferred from the electric fields. Such in-depth plasma characterization can be achieved through proton deflectometry. For that purpose, we have developed a three-dimensional simulation capability in order to compute protons' trajectories modified by the local electric fields.

DIANE is the general Monte Carlo code developed at CEA-DAM. DIANE is a 3D multiparticle multigroup code. DIANE includes automated biasing techniques and is optimized for massive parallel calculations.

This paper discusses experimental results obtained at the 1 MV testbed at CEA Cadarache that appear to show a higher extracted D- current density from small apertures. Plasma grids with different shapes have been installed and tested. All grids had one single aperture. The tests were done in volume operation and in caesium operation. We tested four grids, two with circle divide 14 mm, one with circle divide 11 mm and one with circle divide 8 mm apertures. No aperture size effect was observed in volume operation. In caesiated operation the extracted current density for the circle divide 8 mm aperture appears to be significantly higher (similar to 50%) than for the circle divide 14 mm aperture. Simulations with a 3D Monte Carlo Trajectory Following Code have shown an aperture size effect of about 20%. Finally, as byproducts of the experiments, data on backstreaming positive ions and the temperature of the plasma grid have been obtained.

From a computing standpoint, flash X-ray radiography is much more time-consuming than traditional X-ray applications, and despite the constant increase of computing resources, methods to reduce the calculation time while preserving accuracy are highly needed. At the Commissariat a l'Energie Atomique, DIANE is the code devoted to flash X-ray calculations. After a brief description of the general features of DIANE, two selected methods implemented in DIANE to provide fast calculations are described. One concerns bremsstrahlung X-ray creation without electron transport electrons: the SSB model. The quality of this model is assessed within the framework of flash X-ray applications on two test problems with a fully photon-electron transport performed with MCNP5. The other focuses on particle tracking and Woodcock tracking. The performance of tracking within large meshes is evaluated.

Recent progress in high-gain direct-drive inertial confinement fusion with the laser Megajoule is reviewed. A new baseline direct-drive target design is presented which implodes with a two-cones irradiation pattern of indirect-drive beam configuration and zooming. Perturbation amplitudes and correlated growth rates of hydrodynamic instabilities in the compressed core of a directly driven inertial confinement fusion capsule are analyzed in planar and spherical geometries, with and without heat conduction, in the unsteady state regime of the deceleration. Shock propagation in heterogeneous media is addressed in the context of first shock. The neutron and photon emissions of high-gain direct-drive target are characterized. Numerical interpretations of directly driven homothetic cryogenic D-2 target implosion experiments on the Omega facility are presented.

In the context of the French Laser-Megajoule (LMJ) fusion-research programme, the neutron and photon emissions of a high-gain direct-drive target are characterized. The neutron spectrum accounting for primary, secondary, and tertiary nuclear reactions is determined by means of the post-processor Diane of ID calculations. Photon sources are considered over a wide range of energy from soft x-rays to gamma. Several energy windows of neutron and photon spectra are also identified in order to measure the target areal density.

A new neutron diagnostic has been designed to achieve high energy neutron spectroscopy in large γ background. The concept is based on the association of a Micromegas detector with a neutron-to-charged particle converter. Studies have been performed with a neutron generator and γ-ray sources and it has been shown that 14MeV neutrons can be detected with improved efficiency and relative γ insensitivity. This low γ sensitivity makes this concept appealing for inertial fusion experiments. Experiments have been performed on the 60 beams, 30kJ OMEGA laser system at the University of Rochester and have demonstrated the applications of this detector.

A metal membrane of macroscopic thickness can, under certain conditions, become superpermeable to hydrogen particles whose energy (kinetic, internal, or chemical) exceeds 1 eV. Such a membrane can transfer the energetic hydrogen almost like an opening of the same area. Superpermeable membranes can also compress the permeating gas by orders of magnitude and clean it from any impurities. Due to these characteristics superpermeable membranes could be employed in ion sources and neutral beam injectors: (a) for pumping and recycling hydrogen from ion sources, (b) for pumping hydrogen gas from the extraction area and recycling the gas into the ion source, (c) for pumping deuterium or hydrogen in the plasma neutralizer of a negative ion based neutral beam injector. We will describe the possible use of superpermeable membranes in these devices, the experiments which we effected demonstrating this new technique, and projects for its future use.

A code called NIETZSCHE has been developed to simulate the negative ion transport in a plasma source, from their birth place to the extraction holes. The H−/D− trajectory is calculated by numerically solving the 3D motion equation, while the atomic processes of destruction, of elastic collision with H+/D+ and of charge exchange with H0/D0 are handled at each time step by a Monte Carlo procedure. This code can be used to calculate the extraction probability of a negative ion produced at any location inside the source. Calculations performed with NIETZSCHE have been allowed to explain, either quantitatively or qualitatively, several phenomena observed in negative ion sources, such as the isotopic H−/D− effect, and the influence of the plasma grid bias or of the magnetic filter on the negative ion extraction. The code has also shown that, in the type of sources contemplated for ITER, which operate at large arc power densities (>1 W cm−3), negative ions can reach the extraction region provided they are produced at a distance lower than 2 cm from the plasma grid in the case of volume production (dissociative attachment processes), or if they are produced at the plasma grid surface, in the vicinity of the extraction holes.

The purpose of this experiment is to improve the optics of an intense H− negative-ion beam, to be used in future Tokamaks. The emittance diagnostic has been installed in the experimental setup INCA, to perform an optimization of a negative-ion injector preaccelerator. The first stage of this work consists of testing the operation of the electric-sweep scanner, to measure the beam current in the two-dimensional-phase space and to investigate the diagnostic performances. In addition to the measurement of H− angles, we can scan the angles of the fast H+ created along the beam in the transport region and test the efficiency of the electron magnetic trap installed in the second accelerator grid (extraction electrode). We also observe the presence of a neutral component. This analyzer not only provides a diagnostic of the beam divergence but also improves of the knowledge of the beam components. We studied two accelerator configurations. The first one permitted to obtain an optimal beam optics under 15 keV. Then the accelerator geometry was modified to ensure a higher optimum high voltage. We will show several experimental results performed with two accelerator configurations.

Advanced Tokamak concepts and steady state plasma scenarios require external plasma heating and current drive for extended time periods. This poses several problems for the neutral beam injection systems that are currently in use. The power loading of the ion source and accelerator are especially problematic. The Kamaboko negative ion source, a small scale model of the ITER are source, is being prepared for extended operation of deuterium beams for up to 1000 seconds. The operating conditions of the plasma grid prove to be important for reducing electron power loading of the accelerator. Operation of deuterium beams for extended periods also poses radiation safety risks which must be addressed.

Plasma heating and current drive in future machines for magnetically confined thermonuclear fusion will require neutral beam injectors based upon negative ion sources. Although the technology of filament arc ion sources is considered adequately mature for large scale injectors, such as for ITER, ion source efficiency and reliability need to be improved. The power loading of the source and accelerator are especially problematic. This is particularly important for long pulse (1000 s) operation. Investigations with the Kamaboko source, a small scale model of the ITER arc source, illustrate the importance of the conditions of the plasma grid for reducing electron loading of the accelerator.

The development in Europe of negative ion sources and accelerators is carried out principally at Cadarache, France and at Garching, Germany. Other work is in progress at the Ecolc Polytechnique, France, and CIEMAT, Spain. The main thrust of the programme so far has been driven by the requirements for the 1 MeV D0 injection on ITER. It is now being proposed to develop a new negative ion source and accelerator for use on present and future European fusion experiments. This paper summarises the present and proposed future programmes in Europe.

In a collaborative effort between JAERI and EURATOM- CEA, the KAMABOKO negative ion source, which has produced more than 30 mA/cm2 of H− in Japan, has been tested for deuterium at Cadarache (France). The ion source is of the same design concept, but at a reduced scale, as the proposed large negative ion source for ITER-NBI. The negative D− ions are produced in a cesium seeded multi-cusp plasma generator of good plasma confinement, which allows efficient operation at low pressure. Routine diagnostics such as Langmuir probes and spectrometer are installed to measure plasma characteristics and atomic population. The transverse magnetic filter of 0.088 T.cm, installed in front of the extraction system, had been optimized for electron suppression in hydrogen operation. Stable arc discharges are observed at 0.35 Pa in deuterium for arc powers of up to 78 kW. For a pressure of 0.35 Pa, the ion source produces 1.4 A of D− with an average current density of 20 mA/cm2. The extracted electron current, originally 13 A, was prohibitive for a steady state operation of the extraction grid. Stronger transverse magnetic fields (up of to 0.18 T.cm) have been installed to reduce the electron flux on the extraction grid. At the maximum field stength, we observe that the production of deuterium negative ions is identical to the original configuration, but that the extracted electron current is strongly reduced. The ratio of the electron current to the D− current is less than one, instead of 10 previously. Under these conditions, 1000 second source operation should be possible, which would satisfy the design nominal values for ITER.

NIETZSCHE (Negative Ions Extraction and Transport ZSimulation Code for HydrogEn species) is a negative ion (NI) transport code developed at Cadarache. This code calculates NI trajectories using a 3D Monte‐Carlo technique, taking into account the main destruction processes, as well as elastic collisions (H−/H+) and charge exchanges (H−/H0). It determines the extraction probability of a NI created at a given position. According to the simulations, we have seen that in the case of volume production, only NI produced close to the plasma grid (PG) can be extracted. Concerning the surface production, we have studied how NI produced on the PG and accelerated by the plasma sheath backward into the source could be extracted. We demonstrate that elastic collisions and charge exchanges play an important role, which in some conditions dominates the magnetic filter effect, which acts as a magnetic mirror. NI transport in various conditions will be discussed: volume/surface production, high/low plasmas density, tent filter/transverse filter.

Large current and high current density deuterium negative ion sources are investigated on the MANTIS test bed with the objective of producing several amperes of D− beams, at an accelerated current density in the range 10–20 mA/cm2, for possible application in future neutral beam injectors, e.g. ITER. As a first step, the DRAGON source, which was built by Culham Laboratory was tested on the MANTIS test bed in order to test this large source using only ‘‘pure volume’’ production of negative ions. The accelerated negative ion current is found to be a strong function of the source operating pressure and the arc power, and a significant isotopic effect is observed. The maximum accelerated currents are 1.3 A of H− (3.3 mA/cm2) and 0.5 A (1.3 mA/cm2) at 110 kW of arc power. Cesium injection from a non conventional dispenser together with an improved extraction system, have significantly improved the D‐current. A maximum of 14 mA/cm2 of D−1 are accelerated at 30 kV, which corresponds potentially, to more than 5 A for a full aperture extraction with an arc power of 140 kW (2250 A of arc current).

Neutral beam injection (NBI) is one of the candidates for plasma heating and current drive in the new generation of large magnetic fusion devices (ITER). In order to produce the required deuterium atom beams with energies of 1 MeV and powers of tens of MW, negative D− ion beams are required. For this purpose, multiampere D− beam production and 1 MeV electrostatic acceleration is being studied at Cadarache. The SINGAP experiment, a 1 MeV 0.1 A D− multisecond beam accelerator facility, has recently started operation. It is equipped with a Pagoda ion source, a multiaperture 60 keV preaccelerator and a 1 MV 120 mA power supply. The particular feature of SINGAP is that the postaccelerator merges the 60 keV beamlets, aiming at accelerating the whole beam to 1 MeV in a single gap. The 1 MV level was obtained in less than 2 weeks, the accumulated voltage on-time of being ∼22 min. A second test bed MANTIS, is devoted to the development of multiampere D− sources. It is capable of driving discharges with current up to 2500 A at arc voltages up to 150 V. A large multicusp source has been tested in pure volume and cesiated operation. With cesium seeding, an accelerated D− beam current density of up to 5.2 mA/cm2 (2 A of D−) was obtained. A modification of the extractor is underway in order to improve this performance. A 3D Monte Carlo code has been developed to simulate the negative ion transport in magnetized plasma sources and optimize magnetic field configuration of the large area D− sources.

A compact, safe system for directed cesium deposition into large volume sources is described. It consists of a small oven mounted on a feedthrough, which permits one to rotate and move the oven across the discharge chamber length or to remove it from the vacuum box to reload the oven. Special industrial pellets containing the cesium compound (30% of cesium chromate +70% titanium) were used for a pure cesium release. The pellets are insensitive to pollution by air and can provide the cesium release during several cycles of heating and cooling in long-term experiments. There was no ‘‘poisoning’’ impurities degassing (oxygen, water) during the standard oven operation. The Cs system was reliably operated for cesium deposition in multiampere negative-ion source experiments.

Experimental data on the enhancement of D− (H−) negative ion production due to cesium injection into a large volume multiampere negative ion source (MANTIS) are described. The directed deposition of small cesium amounts (5–100 mg) from a compact, movable oven, placed into the central part of a MANTIS gas-discharge box was used. A calorimetrically measured D− beam with an intensity up to 1.6 A and an extracted current density up to 4.2 mA/cm2 (beam energy 25 kV) was obtained. Exactly 30 mg of cesium provides at least one month of source operation (1000 pulses with a discharge pulse duration of 4 s). The effect of cesium on NI enhancement was immediately displayed after the distributed Cs deposition, but it needed some ‘‘conditioning’’ of cesium by tens of discharge pulses (or by several hours ‘‘pause’’) in the case of a localized Cs deposition. No degradation of extraction-acceleration voltage holding on within the tested range of cesium injection was observed.