Experiments on the dynamics of buoyancy-driven flows in reactive porous media

Reactive flows in porous media are of great interest to a number of geological and industrial situations where fluid of one composition invades a formation containing a solid component with which it can react. Geological situations include dolomitisation of limestone following invasion of sea water into freshwater aquifers. Industrial situations include hydrocarbon extraction and geothermal energy production, where fluid is injected into a porous rock to drive out trapped hydrocarbons or to gain thermal energy. If the injected fluid is compositionally different from the rock, chemical reactions can cause scaling or precipitation to occur in the rock’s pore spaces, affecting the productivity of the system. Much work has been carried out investigating these flows under large pressure gradients, as is the case in the aforementioned industrial applications. However, the body of work which focusses on reactive flows under small driving pressures is smaller. In these situations of slow flow, the effects of buoyancy may become important. Understanding these flows is important in a number of new industrial applications, such as geological disposal of nuclear waste. In this thesis, a series of new experiments are presented investigating the phenomena associated with slow flows in reactive porous media. In these experiments, an invading fluid displaces a formation fluid from a porous medium. The medium contains a soluble component which can dissolve into the invading fluid, prompting a chemical reaction between the fluid and the medium, which takes the form of a reaction front which moves vertically through the medium. In the first two sets of experiments covered in Chapters 3 and 4, the formation fluid is displaced vertically downwards by a buoyant reactive fluid. This creates a rectilinear reaction front, which is stabilised by buoyancy, and which moves downwards through the formation. The reaction front can be described by a simple mass balance in the case of slow displacement, and by a balance between advection and diffusion in the case of faster displacement. Invading fluid moves through the reaction front and dissolves the component, causing its density and viscosity to increase. Depending on the relative properties of the reacted and formation fluids, this may then prompt a Rayleigh-Taylor or Saffman-Taylor-type instability at the fluid-fluid interface downstream of the reaction front. We investigate the critical condition for the formation of these instabilities, before proposing and testing a model for their nonlinear evolution. In the final set of experiments covered in Chapter 5, the buoyant reactive fluid is introduced from the base of the formation, causing it to form a buoyant plume with a central dissolved channel and a reactive head. The speed of the plume head is shown to be controlled by the rate of reaction, with the fate of the reacted fluid depending on its buoyancy relative to the formation fluid. Finally, Chapter 6 provides a commentary on these phenomena, and offers ideas on their relevance to the problem of geological disposal of nuclear waste.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

University of Cambridge

Year dc:date.issued

2025

Author dc:creator

• Clarke, Sam

Advisor dc:contributor.advisor

• Woods, Andrew

• nuclear waste
• rayleigh-taylor
• saffman-taylor
• reactive infiltration
• buoyancy instability

Licence

• https://www.repository.cam.ac.uk/bitstreams/ff85f3a6-9b4d-45a3-9aee-55f0a6d1fd3c/download
• https://creativecommons.org/licenses/by-nc/4.0/

Language dc:language

eng

DOI dc:identifier.doi

https://doi.org/10.17863/CAM.124398

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/394382