CE05 - Une énergie durable, propre, sûre et efficace

underground STOrage of reneWable ENergies in low permeability Geomaterials – STOWENG


Underground STOrage of reneWable ENergies in low permeability materials

Hydro-mechanical instabilities in low permeability geomaterials

Subsurface energy technologies will play a central role in the smooth transition towards a lower-carbon future. Storage of energy into underground reservoirs is one of the current solutions to solve energy production and consumption irregularities. The project investigates the effects of the long-term storage of hydrocarbons on the surrounding environment. In particular the interface between the reservoir rock and the sealing caprock, which is a potential area for critical hydro-mechanical behavior, generating possible instabilities and environmental risks, is investigated. <br />The objective of the project is to analyze instability trigger conditions, focusing on solid fracturing and fluid fingering, at the caprock interface. The case of low permeability geomaterials is therefore addressed, as they constitute the near field of the reservoir. <br />A multi disciplinary approach is adopted merging competences in microscale and laboratory scale experimental geomechanics with constitutive and numerical modeling, to identify the macro-scale characteristics of low permeability geomaterials in terms of hydraulic tightness and mechanical strength.<br />Advanced experimental activities and ad-hoc image analysis methods are developed based on the one hand on an original biaxial testing machine, endowed with multi-resolution cameras, specially designed to monitor fingering and strain localization through granular media, and on the other one on drainage and imbibition tests developed under Neutron and X-ray tomography to get measures of local fluid flow velocities and possibly localized strains.<br />The modeling activity is devoted to the formulation and the numerical implementation of a phase-field model to describe partial saturation, coupled with plasticity and damage; the model is initially implemented within a Matlab Finite Element code and finally within LAGAMINE (FE code developed at University of Liège), more adapted for solving real-scale engineering problems.

The project is based on a multi disciplinary approach merging experimental geomechanics, constitutive law and numerical modeling.
These mutually connected domains are explored to detect and to model fluid fingering and localized strains, in low permeability geomaterials, by means of advanced methods and techniques involving: (i) the design and the implementation of a novel experimental biaxial testing machine for studying coupled hydro-mechanical instabilities in granular materials; the biaxial machine is a unique equipment in which the sample is closed within two transparent sapphire windows which allow to take images of its current state by a system of multi-resolution optical cameras (between 5 and 50 MPixels) and multi-frequency acquisition (between 1 and 500 Hz in full view); (ii) the development of tests on natural samples imaged using both x-ray tomography and neutron tomography to identify fingering; this is possible in the NeXT line of the ILL of Grenoble where samples initially saturated with heavy-water (deuterium) will be subjected to drainage tests and dry samples will be subject to the injection of heavy-water; (iii) the development of new techniques based on Digital Image Correlation (DIC) to detect the propagation a fluid front through a partially saturated medium and to measure displacements and strains induced by the fluid flow; (iv) the enriched constitutive modeling of partially saturated granular media and the characterization via stability analysis of the triggering conditions for fingering and strain localization; (v) the numerical implementation of the model adopting a mixed finite element approach, initially within the Matlab environment, later into the LAGAMINE FE code (vi) the simulation of the response of low permeability geomaterials at the interface between underground aquifer reservoirs and the “sealing” caprock for different hydro-mechanical loading paths.

The partial results at M18 are generally in line with the program. As for the experimental activities on the biaxial testing machine, they were preceded by oedometric tests, carried out to fix the basics of the techniques to be used for tracing the front of the fluid and calculating strains through digital image correlation methods. Since November 2019 the validation of the biaxial testing machine started and the acquisition of the first high resolution images made it possible to refine the method of detection of the fluid front. The method of measuring displacements by eliminating the presence of the fluids on the images is currently under development.
Regarding the analysis of natural geomaterials, a preliminary experimental campaign was carried out in order to choose the material and the fluids, as well as to define the test protocols and the parameters to be used for imaging – based first of all on X-ray. The selected material is the Fontainebleau sandstone; water with sodium iodide is injected into the oil-saturated sample, which causes viscous (high injection speeds) or capillary (lower speeds) fingering. However, the rate at which fingers developed only made it possible to reconstruct a 3D image in the case of capillary fingering, while for viscous fingering we manage (for the moment) to acquire only a sequence of 2D images.
Regarding the numerical modeling activities, the enriched phase field model for partially saturated media has been implemented in a FE code programmed in Matlab. The implementation of the poromechanical model coupled with a damage model capable of describing crack nucleation and propagation in a partially saturated medium is in progress. The first numerical simulations show very encouraging results in terms of model's ability to capture the fingering phenomenon.

We are full of expectations for the second half of the project which should provide the first answers to the main questions we asked ourselves at the beginning of the project itself: is there any coupling between fingering and strain localization? Are we able to experimentally observe fingering induced strain localization due to hydraulic loading and vice-versa shear band induced fingering due to mechanical loading? Are we capable to numerically reproduce the experimental evidences?
Short term goals, which could be achieved before the end of the project, are consequently: (i) a full validation of the biaxial testing machine and the development of laboratory scale drainage and imbibition tests on prototype granular materials; the hydro-mechanical coupling will be characterized by means of DIC methods specifically developed to capture the current position of the fluid front as well as the corresponding induced strains; (ii) a new experimental campaign using NeXT facility at ILL Grenoble to capture the hydro-mechanical coupling in natural deformable geomaterials under drainage and imbibition; (iii) a complete validation of the numerical model, based on the experimental results, coupling fluid instability, localized plastic strains and damage.
Long term goals, which will probably go beyond the end the project, concern: (i) the validation of the results of the biaxial testing machine by means of a triaxial apparatus adapted to partial saturation, whose acquisition in ECN is actually in progress, (ii) the identification of suitable analog materials, alternative to the vermiculites initially considered in the proposal, adapted to identify, at the laboratory scale, chemo-mechanical couplings, characteristic of clayey geomaterials, by means of an experimental setup to be specifically designed; (iii) the realization of field scale simulations describing the interaction between the aquifer reservoir, the tight caprock and the surrounding environment.


1. G. Sciarra A phase field approach to fingering and fracturing in partially saturated porous media, Keynote Lecture EMI International Conference, Lyon 3-5 Juillet 2019
2. S. Ommi, G. Sciarra, P. Kotronis A Phase field approach of damage evolution in partially saturated porous media EMI International Conference, Lyon 3-5 Juillet 2019
3. G. Viggiani. P. Bésuelle, A. Papazoglu, A. Tengattini. Thermo-hydro-mechanical localized patterns in geomaterials: recent experimental observations, Keynote Lecture, IAS Workshop on Emerging Scales in Granular Media, Hong-Kong, 14-16 January 2020

Subsurface energy technologies will play a central role in the smooth transition towards a lower-carbon future. Storage of energy into underground reservoirs is one of the current solutions to solve energy production and consumption irregularities. Two types of underground storage are particularly important: the short-term storage of subsurface energies such as compressed air, hydrogen, and water, and the long-term storage/sequestration of hydrocarbons (liquid or liquefied) and CO2. In this project the attention focuses on this last topic.
Petrophysical and hydro-thermo-mechanical properties of underground reservoir materials are key properties to evaluate yield and storage potential; however they are locally different than the rest of the basin. The interface between the reservoir rock and the near field, in particular the caprock, is therefore a potential area for critical hydro-mechanical behavior, generating possible instabilities and high environmental risks. The most relevant issue, for all underground storages, concerns the environmental impact and the health and safety of people in the occurrence of the loss of sealing capacity of the reservoir,. Fracture triggering and propagation caused by temperature or pressure variations, particularly sudden changes, may compromise the storage confinement, cause gas/fluid migration through groundwater and induce micro-seismicity.
The main objective of this project is to investigate the trigger conditions of these critical states, focusing in particular on solid fracturing and fluid fingering, at the interface between the underground reservoir and the host formation, and attempting to establish proper causality relations between them. The particular case of low permeability clayey geomaterials is addressed, which typically constitute the near field (in particular the caprock) of the reservoir. A multi disciplinary approach will be adopted, which aims at merging competences in micro-scale and laboratory scale experimental geomechanics, constitutive laws and numerical modeling, to identify the main characteristics of low permeability geomaterials in terms of hydraulic tightness and resistance to applied stresses. The main interest, and the originality, of the project reside in covering the gap of knowledge and the lack of modeling tools of the micro-scale physical processes occurring within low permeability clayey geomaterials as well as in describing the effects of these phenomena on the macro-scale properties of the clayey soil.
In order to understand to what extent tightness of underground reservoirs is affected by the occurrence of the above mentioned critical states, prototype low permeability geomaterials, sand-clay mixtures, prepared in the laboratory, and natural clayey geomaterials will be tested imaged and modeled, under different hydro-mechanical loading conditions.
First of all the experimental activity will concentrate on the analysis, at the laboratory scale, of suitable analog materials: on the one hand, granular materials saturated by highly viscous fluids, which simulate low hydraulic conductivity materials; on the other hand macro-crystals of clay, which will be used to investigate chemo-mechanical couplings, but at a much larger scale than usual. Secondly, micro-scale experiments will be implemented to analyze the behavior of sand-clay mixtures and natural geomaterials, under biaxial and in-situ loading conditions.
The modeling activity will be devoted to the formulation and the numerical implementation of a phase-field model for describing the condition of partial saturation. The presence of more than one fluid phase within the porous network, and the possible occurrence of strain localization and fracture of the solid component will be considered.

Project coordination


The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.


Université de Liège / ArGenCo
3SR Sols, Solides, Structures, Risques
Università di Roma TorVergata / Dipartimento di Ingegneria Civile e Informatica

Help of the ANR 465,719 euros
Beginning and duration of the scientific project: October 2018 - 36 Months

Useful links

Explorez notre base de projets financés



ANR makes available its datasets on funded projects, click here to find more.

Sign up for the latest news:
Subscribe to our newsletter