To elucidate the residual rate in the realm of glass alteration, the underlying mechanisms remain a subject of ongoing debate, as they potentially encompass multiple channels that collectively result in a decrease by several orders of magnitude of the initial alteration rate [1].
The Monte-Carlo (MC) method is an effective method for studying complex scientific problems. For studying the glass dissolution by the Monte-Carlo method, early attempts were made in the late 20th century by Aerstens and Van Iseghem followed by Devreux [2] and in the near past by Sebastian Kerisit and his coworkers [3]. The major limitations of the models developed was that it had no way to explain the residual rate and the evolution of gel maturation.
In my thesis, we investigate glass alteration using a new Monte-Carlo code described below to explore the role of the different elements. In parallel, classical force fields for molecular dynamics (MD) simulation are developed to simulate boron diffusion in water within a model alteration gel. Additionally, experiments are performed with glasses containing different Al2O3 quantities. The MD simulations and the experiments will help to determine the Monte Carlo parameters. At the end, we plan to obtain a better description of the role of the different elements on glass alteration in the so called residual rate regime.
This poster is dedicated to the Monte Carlo method. The algorithm has been developed by J.-M. Delaye and coworkers. It differs from the previous algorithm in the fact that the diffusion of water inside the solid is taken into account, using two intricated networks to represent the glass on one side, and the solution on the other side. The MC code is able to reproduce the formation of an alteration layer in glasses containing Si, B, Na and Al. Even if the current version of the code is quite slow, which imposes limitations on the number of calculations that can be executed, we have simulated the alteration of three distinct glass compositions each characterized by varying quantities of Al2O3. The Monte Carlo parameters have been fitted to reproduce at the best the experiments, in particular the quantity of Si and B released in solution. Our findings will demonstrate that two different types of passivating layers can form depending on the way Si is released in solution. When a large quantity of Si atoms is rapidly released in solution, it can lead to the formation of an external layer characterized by a notable enrichment of Si. Conversely, when the release of Si into the solution occurs in a more continuous manner, no layer enriched in Si is observed but the external part of the gel undergoes a progressive reticulation process. These two distinct mechanisms offers a plausible explanation for the alteration behavior change observed with an increase in Al2O3 content [4].

REFERENCES
[1] K. Damodaran, S. Gin, J.-V. De Montgolfier, C. Jegou, J.-M. Delaye, J. Non-Cryst. Solids, 598 (2022) 121938.
[2] F. Devreux, A. Ledieu, P. Barboux, Y. Minet, J. Non-Cryst. Solids, 343 (2004) 13-25.
[3] S. Kerisit, E.M. Pierce, Geochim. Cosmochim. Acta 75 (2011) 5296-5309.
[4] K. Damodaran K. “Insights into the mechanisms controlling the dissolution of alumino-borosilicate glass and development of a new Monte Carlo model ”, PhD, Université de Montpellier, 2022.