Under the co-supervision of:

• Jean-Michel Pereira (jean-michel.pereira@enpc.fr) and Phillip Braun ( (ENPC)
• Diego Manzanal (d.manzanal@upm.es) and Miguel Martín (UPM)

Our society is increasingly demanding the use of renewable energy sources such as solar and wind power and, at the same time, is searching for alternative energy carriers such as hydrogen to mitigate the intermittency of renewable sources. There is significant potential for renewable energy generation and hydrogen production by electrolysis in Europe and America. However, large differences between the variable production of renewable energy and the hydrogen demand worldwide are expected, which would need intermediate energy storage solutions. Therefore, large-scale energy storage with viable technologies will be one of the challenges of the coming decades in ensuring the availability of excess energy when required. The intermittent nature of renewable energies, as a result of daily and seasonal fluctuations, such as the strength of the wind or the variability of solar radiation, can be minimised through hydrogen generation by electrolysis and storage in depleted oil and gas fields (green hydrogen). The option is also to generate hydrogen from fossil fuels and store the captured carbon dioxide in geological reservoirs (blue hydrogen) to avoid atmospheric greenhouse gas emissions. Naturally, the development and widespread application of geological storage technology can take place only if the permanence of the stored CO2 or H2 is assured, which can only be done through a deep understanding of the reservoir and seal characteristics and the development of methodologies to monitor upward migration and to accurately estimate available storage volumes.

Underground hydrogen storage (UHS) in geological formations offers a promising solution for large-scale, temporary storage of hydrogen. However, the cyclic injection and withdrawal of hydrogen impose complex mechanical and dynamic stress paths on the reservoir and surrounding geological structures. These processes can lead to reactivation of discontinuities, such as faults and fractures, which could compromise the integrity of the storage site.

Furthermore, hydrogen can alter the geochemistry of the reservoir, affecting porosity and permeability. Additionally, hydrogen can reduce Fe³⁺ in minerals such as hematite, forming Fe²⁺ and potentially altering the properties of rocks. Some clays may also swell in response to hydrogen injection, further decreasing permeability.

This thesis aims to simulate these microstructural changes using a phase field approach. Such models have been successfully applied to various material processes, involving microstructural changes. The key advantage of the phase field method is that it enables the simulation of evolving arbitrary morphologies and complex microstructures without explicitly tracking interface positions.

A phase-field variational numerical tool will be developed to investigate the cyclic Chemo-Hydro-Mechanical (CHM) behaviour of reservoirs used for underground hydrogen storage, focusing on the reactivation of geological discontinuities. The variational framework will incorporate both the Finite Element Method (FEM) and, in cases involving very high distortions, the Material Point Method (MPM).

The outcomes of this research will improve the understanding of the interplay between cyclic loading and fault behaviour in geological reservoirs, providing valuable insights for the design and operation of safe and efficient hydrogen storage. Furthermore, the development of the variational-based numerical tools proposed in this thesis will help address broader geotechnical and geomechanical challenges in underground gas storage.