Significance: The simultaneous flow of multiple fluid phases through a porous solid occurs in many natural and industrial processes. Microscale physical mechanisms such as the relative affinity of the solid for the fluids (i.e., wettability), capillarity, and viscosity combine with pore geometry to produce a wide variety of macroscopic flow patterns. Pore-scale modeling is an essential tool to connect microscale mechanisms with macroscopic patterns, but quantitative comparisons between different models, and with experimental data, are lacking. Here, we perform an unprecedented comparison of state-of-the-art models from 14 leading groups with a recent experimental dataset. The results underscore the challenges of simulating multiphase flows through porous media, highlighting specific areas for further effort in what is already a flourishing field of research.

Fluid inertia which has conventionally been neglected in microfluidics has been gaining much attention for particle and cell manipulation because inertia-based methods inherently provide simple, passive, precise and high-throughput characteristics. Particularly, the inertial approach has been applied to blood separation for various biomedical research studies mainly using spiral microchannels. For higher throughput, parallelization is essential; however, it is difficult to realize using spiral channels because of their large two dimensional layouts. In this work, we present a novel inertial platform for continuous sheathless particle and blood cell separation in straight microchannels containing microstructures. Microstructures within straight channels exert secondary flows to manipulate particle positions similar to Dean flow in curved channels but with higher controllability. Through a balance between inertial lift force and microstructure-induced secondary flow, we deterministically position microspheres and cells based on their sizes to be separated downstream. Using our inertial platform, we successfully sorted microparticles and fractionized blood cells with high separation efficiencies, high purities and high throughputs. The inertial separation platform developed here can be operated to process diluted blood with a throughput of 10.8 mL min⁻¹via radially arrayed single channels with one inlet and two rings of outlets.

The momentum exchange method has been widely used in lattice Boltzmann simulations for particle-fluid interactions. Although proved accurate for still walls, it will result in inaccurate particle dynamics without corrections. In this work, we reveal the physical cause of this problem and find that the initial momentum of the net mass transfer through boundaries in the moving-boundary treatment is not counted in the conventional momentum exchange method. A corrected momentum exchange method is then proposed by taking into account the initial momentum of the net mass transfer at each time step. The method is easy to implement with negligible extra computation cost. Direct numerical simulations of a single elliptical particle sedimentation are carried out to evaluate the accuracy for our method as well as other lattice Boltzmann-based methods by comparisons with the results of the finite element method. A shear flow test shows that our method is Galilean invariant.

Inertial effects during the process of supercritical CO₂ displacing brine in porous media may not be negligible according to recent studies. Capturing the inertial effects of the physical CO₂‐brine system imposes a requirement on the grid resolution and viscosity to surface tension ratio for pore‐scale simulations, which some commonly used simulators may not be able to meet. To fulfill the parameter requirement, we combine the continuum‐surface‐force based color‐gradient lattice Boltzmann (LB) multiphase model and the geometrical wetting model and extend the model to 3D under the multiple‐relaxation‐time framework. We validate the model via simple benchmarks which show significant improvement over the traditional models. We then perform 3D drainage simulations in a heterogeneous micromodel where the simulation result agrees well with experimental data, while our previous work fails to reproduce certain displacement patterns in the experiment due to the use of a traditional LB model that cannot fulfill the parameter requirement. Finally, we perform high‐fidelity 3D drainage simulations to study the inertial effects in a Bentheimer sandstone sample. Our results show that stronger inertial effects generally help develop more CO₂ flow pathways for the same capillary number which results in higher CO₂ saturation, consistent with the micromodel results. The phenomena can be found in both low and high capillary number cases, indicating that the inertial effects are not dependent on the mean velocity. In addition, the change of the invasion patterns is not proportional to the change of inertial effects, thus exhibiting threshold behavior.

Multiphase flow is ubiquitous in subsurface energy applications and natural processes, such as oil recovery, CO₂ sequestration, and water flow in soils. Despite its importance, we still lack a thorough understanding of the coupling of multiphase flow and reaction of transported fluids with the confining media, including rock dissolution and mineral precipitation. Through the use of geomaterial microfluidic flow experiments and high‐performance computer simulations, we identify key pore‐scale mechanisms that control this coupling. We compare the reactivity of fractured limestone with CO₂‐saturated brine (single phase) and a mixture of supercritical (sc) CO₂ and CO₂‐saturated brine (multiphase). We find that the presence of scCO₂ bubbles significantly changes both the flow dynamics and the resulting reaction patterns from a single‐phase system, spatially homogenizing the rock dissolution. In addition, bubbles redirect oversaturated fluid into low‐velocity regions, thereby enhancing carbonate precipitation occurs.