[en] In future experimental fusion ITER facility, beryllium (Be) constitutes the main first wall plasma- facing material. ITER will then produce Be waste, either in the solid or in powder form. It is important to anticipate the management of the generated radioactive waste. To manage Be, direct immobilization in a cementitious matrix seems to be an option. A research project has then been undertaken to select the best matrix for a safety storage of Be. Nevertheless, Be powder is known as carcinogen and it can cause a respiratory disease, the berylliosis. It then seems interesting to find a surrogate to understand the Be reactivity without handling it. Our work focuses on comparing the reactivity of massive Be with its known substitutes Al and Zn to find the proper surrogate for prospective powder experiments. Thermodynamic data on the stability of Al and Zn in water at 25℃ are known: their passivation zone is respectively between pH 3-10.5 and pH 7.5-13 and they corrode into ionic species bellow and above these pH. On the contrary, the calculated data for Be reported in the literature differ especially in basic media: calculations of Pourbaix predict an aqueous corrosion of Be starting at pH 10.5 while recent studies shows a corrosion above pH 13.5. Owing to the incertitude of thermodynamics for Be, the conditioning of metals (Be, Al and Zn) have been studied in 5 cements with different interstitial pH: brushite, magnesium phosphate cement, calcium sulfo aluminate cement, portland cement and activated slag. The pore solution pH of these matrices ranges from very acid to very basic values. Corrosion studies have been also made in aqueous solutions in a wide range of pH (HCl-NaOH solutions). The metal reactivity is followed mainly by electrochemical impedance spectroscopy in both solutions and cements. Corrosion measurements on Al and Zn are consistent with the literature for both solution and cement experiments. Results on Be can confirm thermodynamics: Be is corroded at low pH with production of hydrogen gas (brushite cement) and Be is stable in cement having a neutral pH pore solution (magnesium phosphate cement) like Al. However, the electrochemical experiments in alkali solutions and cements having a basic pore solution are in agreement with the most recent thermo-dynamic data: Be seems to be protected against the corrosion by a stable protective layer in a pH range from 10 to 13 but strongly corrodes in high alkalis cement (slag) like Zn. (author)

[en] The experimental equipment in combination with pulsed Nd-YAG lasers was developed and applied to investigate co-deposited layer characterisation and ablation. Heating and ablation regimes were distinguished by ablation threshold fluence that was determined experimentally for graphite samples from TexTor (Germany) and TORE SUPRA (France) tokamaks. With 100 ns pulses, the ablation threshold for graphite substrate (2.5±0.5 J/cm²) was much higher than the one for co-deposited layer (0.4±0.1 J cm^-2). These threshold features are very promising to ensure self-controlled laser cleaning without substrate surface damage. The obtained optimal conditions (laser fluence F=1-2 J/cm², 10-20 kHz repetition rate) were applied for co-deposited layer cleaning. The TexTor 50 μm thickness layer was almost completely removed after a single scanning without any damage of the graphite substrate. Cleaning rate of 0.2 m²/hour was demonstrated experimentally for 20 W mean laser power. A theoretical model of a complex surface heating (graphite or metal with a co-deposited layer) was developed to explain the experimental results and to obtain laser cleaning optimisation. A good agreement of the theoretical data with the experimental results was obtained. The studies on LIBS method for co-deposited layer characterisation have determined the analytical spectral lines for hydrogen, carbon, and other impurities (B, Fe, Si, and Cu) in TexTor graphite tile. The obtained results should be regarded optimistic for co-deposited layers characterisation by LIBS method. The development of certain laser methods and their application for in-situ detritiation and co-deposited layer characterisation are presented and discussed. (authors)

[en] First results are presented in relation with experimental and theoretical studies performed at the CORIA laboratory in the general framework of the determination of the chemical analysis of Tokamak plasma facing materials by laser-induced breakdown spectroscopy (LIBS) in picosecond regime. Experiments are performed on W in a specific chamber. This chamber is equipped with a UV-visible-near IR spectroscopic device. Boltzmann plots are derived for typical laser characteristics. We show that the initial excitation temperature is close to 12 000 K followed by a quasi steady value close to 8500 K. The ECHREM (Euler code for CHemically REactive Multicomponent laser-induced plasmas) code is developed to reproduce the laser-induced plasmas. This code is based on the implementation of a Collisional-Radiative model in which the different excited states are considered as full species. This state-to-state approach is relevant to theoretically assess the departure from excitation and chemical equilibrium. Tested on aluminum, the model shows that the plasma remains close to excitation equilibrium. (paper)

[en] Highlights: • Measurements in space and time characterizing the absorption. • Electron density variation of the same order for the studied experimental conditions. • Observed absorption strongly depending on the experimental conditions. • Saturation effect in the ionization process put into light. The LIBS (Laser-Induced Breakdown Spectroscopy) method has already demonstrated its reliability and its robustness in many situations for the multi-elemental composition determination of samples. However, certain conditions prevent a totally satisfactory determination. For instance, the method is weakly efficient to measure with accuracy the light elements concentration in metallic matrices. Since the laser pulse used to produce the plasma contributes to its heating, using an additional pulse (double pulse configuration) provides the increase in electron temperature and density without additional ablation. A better signal-to-noise ratio and a lower limit of detection can be reached. The present paper reports the results of different experiments performed to quantify the modifications induced (1) on the electron density by the second laser pulse in a preformed aluminum plasma, and (2) on the second laser pulse itself. The related experiments have been done in the case where the plasma is produced by a picosecond laser pulse and the second laser pulse is of the nanosecond type. The electron density reaches a maximum resulting from the total ionization of the aluminum plasma volume irradiated by the second laser pulse.

[en] Images of wide-angle visible CCD cameras contain information on dust creation events (flaking) that occur during plasma operations. Due to the interaction with plasma, flakes entering into the plasma left straight line-like visible traces behind in the images. Analyzing these traces by image processing, the temporal evolution, spatial distribution, and statistics on dust creation events in DITS campaign in Tore Supra were obtained.

[en] The electrostatic self-charging rate of tokamak dust is investigated using Geant4, a toolkit for the simulation of the passage of particles through matter. To do so, the particles geometrical characteristics, the β disintegration energy spectrum and the deepness of tritium infusion are taken into account. The investigated materials are tungsten and beryllium, the plasma facing components (PFC) of ITER, considered as spherical particles from 20 nm to 200 μm in diameter, both tritiated. Two cases of tritium distribution in the particles are examined. On the one hand, tritium is homogeneously distributed over the whole sphere; on the other hand, tritium is homogeneously distributed within the external 100 nm layer of the sphere. The self-charging rate is assessed through the calculation of the particle exiting electron rate. Based on a tritium inventory of 10 GBq/g, relevant for ITER tokamak environment, our results show that, for a single tungsten or beryllium particle of 10 μm in diameter, the self-charging rate when the tritium is homogeneously distributed within the whole sphere is respectively 2.4 and 1.9 positive elementary charges per second. In the configuration where the tritium absorption is confined in the external 100 nm layer, the charge magnitude raises up to 37.1 and 8.4 respectively. (paper)

[en] Impact of the helium plasma exposure on the surface modification in tungsten and reduced activation ferritic/martensitic (RAFM) steel have been investigated on the linear plasma device PSI-2 assuming the condition of DEMO first wall. In tungsten, a nanoscale undulating surface structure, which has a periodic arrangement, is formed under low temperature conditions below fuzz nanostructure formation threshold similar to 1000 K. Interval and direction of the undulation shows dependence on the crystal orientation. A large variation in surface level up to 200 nm has been observed among grains at a fluence of 3 x 10(26) He m(-2) showing dependence of the surface erosion rate on the crystal orientation. The (100) plane in which the undulating surface structure is not formed shows the highest erosion rate. This significant erosion is due to the multistage sputtering through impurity. In RAFM steel, sponge-like nanostructure is developed and it grows with increasing helium fluence beyond 1 μm. In the sponge-like nanostructure, a composition change from the base material is observed in which the tungsten ratio increases while the iron ratio decreases showing differences in sputtering ratio depending on the atomic mass. (authors)

[en] In-vessel dusts will be an important issue in next generation fusion devices. They affect not only plasma operations but also safety limits. A new image processing technique for in-vessel dust research is developed in Tore Supra (TS). This paper describes the image processing technique for in-vessel dust study based on simple standard image processing techniques. The technique is intended to use visible standard RGB CCD cameras at a standard frame rate already installed in TS. Studying the traces of straight line-like dust trajectories appearing in the CCD images during the plasma shots, the temporal evolution and the spatial locations of dust creation events (DCEs) can be monitored. Building a database of DCEs, information on the short/long term behavior of in-vessel dusts are obtained. Analyzing the database, statistics on DCEs can be found. Finally, DCEs during 22 ohmic discharges in 2006 CIMES campaign in TS are illustrated as an example.

[en] Recent studies show that low grain sizes are favorable to improve ductility and machinability in tungsten, as well as the resistance to ablation and spallation if this material is to be used in thermonuclear fusion environment. However, current production routes are not suitable for the fabrication of large bulk nanostructured tungsten samples. The authors propose here a new methodology based on powder metallurgy, including the powder synthesis by the reduction of tungsten trioxide by magnesium using the Self-propagating High-temperature Synthesis process in the presence of a reaction moderator, and the densification procedure. Results show that full tungsten densification may be obtained by SPS at a temperature lower than 1800 degrees C and that the resulting morphology, keeping a partial nanostructure inherited from the synthesized powders, seems indeed favorable to the use of these materials in fusion environments. (authors)

[en] The behaviour of hydrogen isotopes in ITER monoblocs was studied using the code FESTIM (Finite Element Simulation of Tritium In Materials) which is introduced in this publication. FESTIM has been validated by reproducing experimental data and the Method of Manufactured Solutions was used for analytical verification. Following relevant plasma scenarios, both transient heat transfer and hydrogen isotopes (HIs) diffusion have been simulated in order to assess HIs retention in monoblocs. Relevant materials properties have been used. Each plasma cycle is composed of a current ramp up, a current plateau, a current ramp down and a resting phase before the following shot. 100 cycles are simulated. The total HIs inventory in the tokamak during resting phases reaches 1.8 x 10^-3 mg whereas during the implantation phases it keeps increasing as a power law of time. Particle flux on the cooling channel of the monobloc is also computed. The breakthrough time is estimated to be t = 1 x 10⁵ s which corresponds to 24 cycles. Relevance of 2D modelling has been demonstrated by comparing the total HIs inventory obtained by 2D and 1D simulations. Using 1D simulations, a relative error is observed compared to 2D simulations which can reach -25% during the resting phase. The error during implantation phases keeps increasing. (authors)