In industrial silicate glasses, chemical diffusion of elements plays an important role at all steps of the glass melting and transformation processes. Specifically, chemical diffusion controls the inter-diffusion between thin films and the glass substrate when heating tempered or curved glass above the glass transition. Although inter-diffusion is beneficial for a minority of applications, most of the time it alters the optical, electrical or mechanical properties of thin film stacks. Silica thin films (often doped with other elements) partly reduce alkali migration towards functional layers. Elemental migration has a complex dependency with temperature and glass composition, because of inter-species couplings, which are required for the displacement of ions within a strongly polymerized structure. No model has been proposed yet in order to predict coupled multicomponent diffusion in silicate melts, with the exception of a few narrow regions of simplified systems (with respect to complex phase diagrams of industrial interest). In addition, extrapolating beyond such narrow regions would require a deep fundamental understanding of diffusion-controlling mechanisms, which is still lacking.
The MAGI project aims to study and model multicomponent diffusion in silicates, in the Na2O-CaO–Al2O3–SiO2 (NCAS) system, optionally doped with Sn or Zn, for conditions relevant to flat glass melting and transformation. Our approach consists in studying diffusive couplings between species as well as the influence of glass structure on such couplings, in order to obtain predictive models of diffusion phenomena encountered in the glass industry, via the basic understanding of kinetic and thermodynamic microscopic mechanisms at the origin of diffusion. To this aim, five academic teams, together with the glass manufacturer Saint-Gobain, will use an unprecedented combination of experimental and numerical techniques. Concentration profiles along diffusion gradients will be measured using various micro-analysis techniques, with a spatial resolution ranging from nanometers to millimeters. From such measures, we will obtain the diffusion matrices of the compositions of interest. Diffusion matrices are the matrix extension of a scalar diffusivity, and are found in a multicomponent version of Fick's law taking into account coupling between species. For a local view of diffusion, local configurations and their dynamics will be obtained thanks to structural measurements, and molecular-dynamics simulations will give access to atomic trajectories and rearrangements.
The observed phenomena will be rationalized by identifying dominant exchange reactions between species, which can be obtained from the eigenvectors of diffusion matrices, or by processing trajectories in molecular dynamics. We will explore the persistence of these exchange reactions throughout the phase diagrams of interest (within the Na2O-CaO–Al2O3–SiO2 system, and with the addition of Zn or Sn as dopants), and at different temperatures (in the molten state, and for metastable liquids close to the glass transition). Finally, we will strive to give a thermodynamic interpretation of our results using Onsager's formalism for multi-diffusion, the latter approach being commonly used for metallic alloys, contrary to silicate melts. Thanks to these results, we will be able to predict diffusive exchanges between a glass substrate and silicate thin films, and to suggest optimal process parameters (such as temperature, compositions or film widths) for different applications. Diffusion laws obtained from our work will be implemented in a software modeling diffusive exchanges. The MAGI project will make an important breakthrough for the understanding and modeling of coupled diffusion in silicate liquids of industrial interest, for bulk materials as well as for interactions between a bulk substrate and silicate thin films.
Madame Ekaterina Burov (Surface du Verre et Interface)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
SGR SAINT GOBAIN RECHERCHE
CNRS-CEMHTI CNRS- UPR 3079 Conditions Extrêmes et Matériaux : Haute température et Irradiation
PHENIX PHysicochimie des Electrolytes et Nanosystèmes InterfaciauX
IRAP Institut de Recherche en Astrophysique et Planétologie
IMPMC Institut de Minéralogie, de Physique des Matériaux et de Cosmochimie
SVI Surface du Verre et Interface
Help of the ANR 635,412 euros
Beginning and duration of the scientific project: December 2017 - 42 Months