WP13, WP ACED, WP DONUT and WP FUTURE are happy to announce a joint training course on Geochemical & Reactive Transport Modelling for Geological Disposal. This course will be organized from February 6th until February 10th 2023. More information will be made available in Autumn 2022. Please find a description of this training course below.
The geochemical evolution of the near field barrier system of a geological disposal is important in view of the safety and performance analyses of the repository as it will influence (i) the durability of the different materials, and (ii) speciation and mobility of radionuclides. Given the time scales involved (ten thousand to hundred thousand years), assessing the evolution can only be done with numerical models in which geochemistry is linked to transport, thus with reactive transport codes.
Geochemical models calculate geochemical state variables based on thermodynamic equilibrium and kinetic processes accounting for processes as aqueous speciation reactions, dissolution/precipitation based on saturation state, sorption based on mechanistic sorption models (exchange reactions, surface complexation) and possible kinetic processes (related e.g. to the dissolution of glass or clay minerals or the corrosion of steel canisters).Reactive transport codes typically couple these geochemical models to flow, heat transport and solute transport solvers. State-of-the-art reactive transport codes may couple this also the water flow or heat transport. Therefore, these codes are capable to simulate coupled thermal, hydraulic, chemical and biological (THCB) processes and possible feedback between the processes. They became a powerful tool for understanding and assessing these coupled processes and the consequences for containment.
Given the complexity of the system and the long-time scales, models typically have large computational times and many uncertainties associated with it. Recent developments in new couplings between different solvers, faster methods to solve equations including methods based on machine learning, and efficient algorithms for uncertainty analysis are crucial in the framework of the analysis of the long-term evolution, optimization and performance assessment of a radioactive waste repository.
In EURAD, the work packages ACED (Assessment of Chemical Evolution of ILW and HLW Disposal cells) and DONUT (Development and improvement of numerical methods and tools for modelling coupled processes) improve and implement codes and models for assessing the geochemical evolution in the near field of a repository. FUTURE develops further understanding in radionuclide migration. In view of that, this training will continue from the state-of-the-art and introduce the new developments acquire in these work packages.
The training aims at enlarging knowledge and expertise in geochemical and coupled reactive transport modelling in the framework of disposal of radioactive waste with the focus on geological disposal. The theoretical basis will be enhanced by lectures on (i) principles of geochemical and reactive transport modelling, (ii) their applications for processes and evolution of materials in a geological disposal (cementitious materials, glass, steel, clay, granite) (iii) speciation and migration of radionuclides, and (iv) advanced topics related to uncertainty and machine learning. Practical skills will be improved by computer sessions in which participants will use available software to implement and analyze models for calculating properties and evolution of materials and speciation of radionuclides.
Upon completion of the training course, the participants should be able to:
- Understand the principles of geochemical thermodynamic and kinetic modelling and reactive transport modelling
- Use these principles for application in the field of radioactive waste disposal
- Transform specific research questions related to geochemical properties or evolution into a conceptual model
- Implement simple conceptual models into numerical codes for geochemical and reactive transport modelling
- Identify advanced methods for sensitivity analysis, uncertainty analysis and integration of machine learning techniques
- Better understanding of some of the limitations of thermodynamic modelling
Training course description.
The one-week training is built around two pillars:
- Lectures by experts in the field on topics on
- Geochemical and reactive transport modelling, properties and evolution of materials in the engineered barrier system of a geological disposal, geochemical properties of the host matrix, and sorption of radionuclides.
- Advanced topics on sensitivity and uncertainty studies and surrogate learning (machine learning)
- Demonstration of state-of-the-art codes for implementing geochemical and reactive transport models.
- Practical sessions on applying numerical codes for (geo)chemical equilibrium speciation calculations and reactive transport simulations. The course will provide training sessions for some of the more widely used codes (PHREEQC, GEMS and ORCHESTRA, and their coupling with transport models) between which the participant can choose. In this course, the participants will know a number of different codes each with their own characteristics. The training sessions will be built around two examples that are used in the machine learning benchmark study of ACED DONUT:
- Ageing/leaching and/or carbonation of cementitious materials based on ordinary Portland cement,
- Uranium sorption on clay materials.