*1.1. General Information Regarding the Underground Storage of Hydrogen*

To contribute more effectively to the fight against climate change and the preservation of the environment, as well as reinforcing their energy independence, France published the Energy transition law for green growth in 2015 [1]. This law aims to increase the share of renewable energies to 23% of gross final energy consumption in 2020 and 32% in 2030, compared to 16% currently [2]. The development of these renewable energies will come up against the need to manage the fluctuating or intermittent nature of some of them. This will involve storing the energy produced in excess or not consumed so that this energy can be re-used later (directly as fuel or mixed with natural gas,

or indirectly by converting it into heat or electricity). The underground environment has many advantages with regard to its potential for high capacity storage in the short or medium-term [3]. France already has 100 operational underground reservoirs of which 78 are salt cavities: these are very large underground cavities, of the order of a million m<sup>3</sup> , formed by injecting freshwater into deep salt formations. Currently, the storage capacity of all of these salt cavities together totals around 14 million m<sup>3</sup> of liquid or liquefied hydrocarbons and 2 billion m<sup>3</sup> of natural gas [3]. Against a background of the gradual abandonment of fossil fuels, a number of research studies are looking into the possibility of storing hydrogen (H2) in such deep salt cavities in the future.

It is within this context that the ROSTOCK-H project (Risks and Opportunities of the Geological Storage of Hydrogen in Salt Caverns in France and Europe) has been financed by GEODENERGIES the French Scientific Interest Group. This project started in 2017 and will end in 2021. One of its objectives is to define monitoring methods for the detection of sub-surface hydrogen leakage, with the dual aim of (i) sizing a measurement scheme capable of detecting diffuse hydrogen leaks, and (ii) studying the diffusive process and the chemico-physical impacts of hydrogen in a shallow aquifer. The approach is centered on two experimental simulations separated in time on the same experimental site. Simulation 1 consists of analyzing the migration in groundwater of a plume of water saturated with neutral gas (helium) and containing various tracers. The objective is to test the operation protocol envisaged for the future injection of hydrogen and to optimize the associated monitoring systems. Simulation 2 consists of creating a plume of dissolved hydrogen in groundwater, according to the same protocol used with helium, to simulate a sudden and brief leak from a deep geological hydrogen storage site towards a shallow aquifer. The evolution of the plumes thus created in the saturated zone, and any potential outgassing to non-saturated zone and the surface will then be followed. All of these simulations will take place at the Catenoy (Northern France) experimental site, which has already been used in the context of similar experiments that studied the behavior of CO<sup>2</sup> for the purpose of Carbon Capture and Storage, or CCS [4,5].

The first injection simulation, which is the subject of this article, therefore involves helium and aims to size the entire leak simulation system and to adapt its protocol and monitoring for the simulation of a hydrogen leak which will subsequently be carried out on the same experimental site. Helium, the gas is chosen for this test, exhibits a physical behavior similar to that of hydrogen, in particular a very low solubility and a high diffusion coefficient in water. At the same time, it is a non-flammable gas, as opposed to the highly flammable hydrogen. This fact makes the organization of this pre-test less complicated from a safety point of view while respecting the similarities with the future hydrogen experiment.

The test site is located in the chalk layer within the Paris Basin. The protocol adopted consists of extracting water from the shallow aquifer, saturating it with gas (helium), and then reinjecting it into the aquifer with tracers to follow the propagation of the dissolved gas plume. This test aims to improve the experimental protocol to be used for the subsequent experiment involving injecting dissolved hydrogen into the aquifer (simulation no. 2).
