Pore Scale Multiphase Dynamics Redefining Cyclic Hydrogen Storage from Pores to Reservoir

Objective

Europe’s transition to a climate-neutral energy system by 2050 depends on safe and efficient underground hydrogen storage to balance intermittent renewable supply and secure energy independence. Depleted hydrocarbon fields and deep aquifers are the most practical options, yet deployment is limited by uncertainties in hydrogen-brine dynamics under cyclic injection and withdrawal. Existing approaches rely on static flow functions and empirical hysteresis rules, leaving injectivity, deliverability, and residual trapping poorly constrained. This fellowship will establish the first predictive, cycle-aware pore-to-reservoir framework for underground hydrogen storage by integrating advanced experiments, innovative descriptors, and physics-informed modelling. High-speed microfluidics and synchrotron micro-Computed Tomography will generate unprecedented multi-cycle hydrogen-brine datasets under realistic boundary conditions. These datasets will be analyzed using morphology- and topology-aware descriptors with machine learning methods to capture interfacial memory, stability, and flow regime transitions. Insights will underpin new hysteresis-energy coupling models and dynamic flow functions, enabling reservoir-scale simulations that account for path dependence and dissipation. The project addresses Horizon Europe MSCA priorities on climate action, clean energy transition, and innovation-driven growth. Scientifically, it will deliver open-access benchmark datasets, reproducible workflows, and transferable methods for porous media research. Technologically, it will provide cycle-aware modelling tools for direct integration into industrial simulators, reducing risk and improving operational efficiency. Societally, this project directly supports the European Green Deal, Hydrogen Strategy, and REPowerEU, enabling reliable renewable integration, reducing fossil fuel reliance, and strengthening Europe’s global leadership in hydrogen technologies.