Comparison of hydrogen gas-, atom- and ion-metal interactions

In the first part of the paper. we describe in more detail the dissociative chemisorption model, presented by Wang for the gas-metal interaction and then developed by Pick and Sonnenberg for the atom-metal interaction, in order to show the important factors which were not taken into account by previous authors. One of these factors is the initial sticking coefficient so which can be less than unity even for a clean surface and can strongly influence the plasma-driven permeation and inventory. Then, we extend the dissociative chemisorption model on the case of the interaction of hydrogen molecular gas and fast hydrogen atoms with a multi-layer metal, and we verify the validity of the model by comparison of experimental data with calculations. Although in recent theoretical papers [Richards, Pisarev and Ogorodnikova] it was shown that Baskes' model has an error, many authors apply Baskes' recombination coefficient for their calculations till now. Experimental measurements show that the plasma-driven permeation and inventory are a function of the sticking coefficient which is very sensitive to any surface contamination that is in contradiction with Baskes' model. Finally. on the basis of the dissociative chemisorption model, the assessment of the plasma-driven permeation through martensitic steel MANET covered by some plasma-facing metals (copper, beryllium and beryllium oxide) is presented. We show that the experimental data of hydrogen isotope permeation and inventory measured for the gas-metal interaction cannot be applied for the plasma-driven permeation and inventory. The analysis of the existing database of the sticking coefficient s updates those estimates for endothermic metals which can be applied for the fusion reactor. (C) 2000 Elsevier Science B.V. All rights reserved.