Wettability control on multiphase flow in patterned microfluidics

B. Zhao, CW MacMinn, and R Juanes. PNAS, 113(37):10251–10256, 2016. doi:10.1073/pnas.1603387113

When two fluids coexist in a porous medium, from water and air in a kitchen sponge to water and oil in a rock, the solid will prefer to be coated by one of the two fluids rather than the other. This preference is known as wettability. Wettability is important whenever two fluids interact with a solid surface, and it is particularly important in porous media because of the very large contact area between the fluids and the solid. It has been known for decades that wettability can have a profound impact on fluid flow in porous media, particularly when one fluid invades by displacing a different one, such as when rainwater infiltrates into soil by displacing air, when carbon dioxide is stored in deep saline aquifers by displacing brine, or when water is injected into petroleum reservoirs to extract oil.

When the solid prefers the defending fluid to the invading fluid, the displacement is known as drainage. The reverse, when the solid prefers the invading fluid, is known as imbibition. Here, we study these processes in detail by systematically varying the wetting properties of a microfluidic porous medium. The power of this technique is that we can modify the wettability while keeping everything else constant — the two fluids and the geometry of the porous medium are identical in all experiments. We perform experiments for six unique wettability conditions across the full range, from strong drainage (where the solid strongly prefers the defending fluid) to strong imbibtion (where the solid strongly prefers the invading fluid). For each case, we then perform experiments at three different flow rates, as measured by the Capillary number Ca. A low value of Ca corresponds to a "slow" flow, whereas a high value corresponds to a "fast" flow.

Our high-resolution images reveal that wettability exerts a fundamental control on multiphase flow. These results elucidate the inherently 3D pore-scale mechanisms and help to explain the striking macroscopic displacement patterns that emerge. The videos below highlight a few of our key results. In all cases, the colormap shows the gap-averaged saturation of the invading water (black indicates no water, red to white incidates an increasing amount of water).

Video 1. The evolution of the invasion pattern for strong drainage, weak imbibition, and strong imbibition at a relatively high value of the capillary number, Ca=0.29 ("fast" displacement). These videos correspond to the snapshots shown in Figs. 2a, 2j, 2m of the paper, respectively.

Video 2. The evolution of the invasion pattern for strong drainage, weak imbibition, and strong imbibition at an intermediate capillary number, Ca=0.029. These videos correspond to the snapshots shown in Fig. 2b, 2k, 2n of the paper, respectively.

Video 3. The evolution of the invasion pattern for strong drainage, weak imbibition, and strong imbibition at a relatively low value of the capillary number, Ca=0.0029 ("slow" displacement). These videos correspond to the snapshots shown in Fig. 2c, 2l, 2o of the paper, respectively.

Video 4. The unique pore-scale displacement processes in strong imbibition at different Ca (left to right is slow to fast: Ca=0.0029, Ca=0.029, and Ca=0.29).