Julien Maes
Assistant Professor,
Environmental Fluid Dynamics,
Heriot-Watt, U. K.
Multiphysic multiscale modelling of Porous Media Applications
Recent advances in the domain of Computational Fluid Dynamics (CFD) have given us the ability to simulate complex processes directly on X-Ray computed micro-tomography images of natural or manufactured porous media, offering an unprecedented window into the physics of multiphase flow and reactive transport at the pore-scale from micro-porosity (~1-10 microns) to macro-porosity (~10-1000 microns) to cracks and fractures (>1mm). My research concerns the use of CFD at these various scales to complement experiments, bring insight into the physics of complex processes and investigate upscaling to the laboratory scale (~10 cm)
GeoChemFoam: Direct pore-scale and multiscale modelling of complex porous media applications
I lead the development of GeoChemFoam, IGE's in-house pore-scale CFD simulator. With my colleagues, I have used GeoChemFoam to perform some of the most advanced CFD investigation of porous media applications in recent years, including the first-ever simulation of multiphase reactive transport in the micro-CT image of a rock. GeoChemFoam is the only CFD simulator in the world capable of simulating the dissolution of gas bubbles in porous media at the microscale, paving the way for a completely new kind of simulations of CCS processes, where the CO2 is not injected along with the water, but directly comes from dissolution of trapped bubbles (Here). For more information, visit the dedicated GeoChemFoam page.
Modelling of reactive transport with mineral dissolution in micro-CT images
Our pore-scale reactive transport solver has improved! We can now perform simulations of mineral dissolution in micro-CT images with similar accuracy as with an interface tracking method but in the fraction of the computational time. Our novel method, which we called Improved Volume-Of-Solid (iVoS0, is based on the micro-continuum approach and is presented in our latest paper, accepted for publication in Frontier in Earth Science (Here). It is of course developed in GeoChemFoam.
Modelling of flow and heat transfer in micro-CT images
Are you looking for a simple, fast and accurate conjugate heat transfer solver? Check out GeoChemFoam! In our latest paper published in Springer Heat and Mass transfer, we used the micro-continuum approach to simulate steady-state flow and heat transfer (Here) in micro-CT images of porous material, and it's really easy and efficient!
Microfluidic flow experiments based on 3D printing
I am responsible for the development of a microfluidic laboratory based on 3D printing. This laboratory is co-funded by the Engineering and Physical Science Research Council (EPSRC) and Aramco service. We develop state-of-the-art techniques for fast, cheap and repeatable flow experiments in micromodels, 2D single-layer pore structures representative of porous media. We use the 3D printer formLabs form2 (www.formlabs.com) to manufacture many realisations of the same micromodel in a repeatable fashion and at low cost, in order to perform repeatable experiments, complemented by CFD simulations. In addition, we use Particle Image Velocimetry to reconstruct the velocity field to benchmark the validity of our 3D printed micromodel (Here). We have an ongoing collaboration with Dr Sophie Roman (www.sophieroman.com) at the Institut des Sciences de la Terre d'Orleans (ISTO), founded by the French and British councils (Alliance Hubert Curien).
Upscaling using multiscale simulations and machine-learning regression models
In our latest paper authored by Dr Hannah Menke, we have extracted key structural features from 3D images of the rock at the micrometre and nanometre scales and then used machine learning to understand how these features impact permeability. We then integrated this learning into a next generation flow model to predict permeability as reported in Nature.
​
​
To produce the 3D images in the machine learning database, we used state-of-the-art x-ray imaging at the micron and nanoscales at Zeiss Labs. The images were then correlated in space and the structural features were extracted and overlayed. Numerical flow models were run at three progressively increasing scales (micron, mm and cm) to solve for permeability using the open-source numerical framework OpenFOAM. Finally, a machine learning model was used to regress the permeability results against the features to create an upscaled description of permeability for the largest scale that encapsulated all structural information from the smaller scales. This new upscaled model successfully predicts permeability with improved accuracy at a fraction of the computational cost of traditional approaches.
Lockdown Digital Rock Physics with Toys
The coronavirus pandemic has shaken the world in 2020. Lockdowns and travel restrictions force us to rethink how we share our work with the community. I recently did a fun and helpful exercise of making a two-minute video explaining my research with whatever I can find in my house, and I chose to do it using my kids toys!
The Continuous Species Transfer (CST) method to model multiphase flow with interfacial transfer
I am one of the main developers of the CST method that allows for simulation of multiphase flow with interfacial transfer based on chemical disequilibrium. In particular, I have developed an extension of the CST method that takes into account local volume change at the interface. The method has been developed with the objective of simulating CO2 dissolution in underground reservoir brine, but can be applied to any multiphase process with interfacial transfer. The pictures show simulation of dissolution of a rising bubble of gas in liquid for two different types of bubble and for two different Henry constants. For more information, see my paper in Journal of Computational Physics: A unified single-field Volume-of-Fluid-based formulation for multi-component interfacial transfer with local volume changes
The method has been developed in our open-source openFoam based simulator (GeoChemFoam)
Direct pore-scale reactive transport modelling of dynamic wettability changes induced by surface complexation
By linking GeoChemFoam with PHREEQC, the US geological survey geochemical solver, we are capable of modelling multiphase reactive transport with aqueous reactions, surface complexation and dissolution. This technology can be used to investigate reactive subsurface processes at the pore-scale, such as dynamic wettability changes induced by surface complexation. This is at the core of low-salinity flooding, an enhanced oil recovery process. The images show the evolution of ionic strength I during simulation of tertiary low-salinity flooding in a 2D carbonate grainstone domain (1 mm x 0.5 mm). The top image shows the residual oil (in green) after secondary high-salinity (in dark blue) injection. We then inject low-salinity brine (light blue), which reacts with the carbonate surface, modifies the bounding force between the solid and the water/oil interface, changing the contact angle toward more water-wet systems. As a result, an additional network of pores is invaded. For more information, see my paper in Advances in Water Resources: Direct pore-scale reactive transport modelling of dynamic wettability changes induced by surface complexation.