• Title of article

    Semi-stochastic generator (FraGMA) of 2D fractured media by mechanistic analogy – Application to reactive transport in a fractured package of vitrified nuclear waste

  • Author/Authors

    Crevoisier، نويسنده , , David and Bouyer، نويسنده , , Frédéric and Gin، نويسنده , , Stéphane، نويسنده ,

  • Issue Information
    روزنامه با شماره پیاپی سال 2011
  • Pages
    12
  • From page
    1387
  • To page
    1398
  • Abstract
    In a geological storage context, the durability of nuclear glass is first and foremost determined by its alteration rate and by the surface being leached. This surface is comprised of a network of three-dimensional cracks due to the stress relief generated by the cooling of glass packages. In order to improve the methods employed so far for quantifying the impact of a fissure network on the life time of glass packages in geological disposal, we have developed a software, referred to as FraGMA, which can be used to generate a realistic 2D geometry of the fissure network inside the glass based on mechanistic analogies. This type of tool is required for providing a mesh adapted to the calculation codes simulating water flow and glass alteration in such a medium. The main principles of this tool have been developed and its parameters have been defined based on the analysis of theoretical fracture processes and observations of real fissure networks. The 2D networks developed using FraGMA are compared with cross-sections of real networks. It is outlined that the general structure of the fracture maps is correctly reproduced, compared to the observations; consequently, it is relevant to apply the tool for further studies in order to generate more complex crack networks, and to study the channeling in such structures, as expected by the strong organized pattern of the experimental fracture maps.
  • Keywords
    FraGMA , Nuclear waste , Fractured medium , Fissure , Propagation , MeSH , Statistical distribution , Mechanistic analogy , Quadtree
  • Journal title
    Computational Materials Science
  • Serial Year
    2011
  • Journal title
    Computational Materials Science
  • Record number

    1688632