(381i) Inertial Spreading and Imbibition on Porous Surfaces in Microgravity

To inform the development of numerical simulations of complex capillary phenomena, we observed the imbibition of relatively large water drops into a network of initially dry, hydrophilic, unconnected, parallel cylindrical capillaries on the International Space Station (ISS) (Fig. 1).

When a drop touches a porous medium, it spreads as if laid on a composite surface. The surface first behaves as a hydrophobic material, since liquid must penetrate pores filled with air. When contact is established, some of the liquid is drawn into pores through a capillarity that is resisted by viscous forces growing with length of the imbibed region. This process always begins with an inertial regime that is complicated by possible contact pinning.

Because time and distance must be shrunk to mitigate gravity-induced distortion, it is difficult to observe details of the rapid early phase of this process on Earth. In contrast, microgravity on the ISS allowed us to stage slower experiments involving relatively large water spheres of 12mm diameter. These drops spread on, -- and imbibed in -- nine kinds of porous capillary plates made of electroless gold-coated brass treated with Self-Assembled Monolayers of known advancing and receding contact angles. In these tests, astronaut Luca Parmitano slowly extruded water drops from a pipette until they touched the plate. Their diameter was large enough for high-speed GX1050C video cameras to visualize their curvature, spreading rate, dynamic contact angle, and imbibition rate (Fig. 1).

The data confirmed numerical results of Frank and Perré (Phys. Fluids 24, 2012) showing that the contact patch radius spreads as a power-law of time (Fig. 1D), even for strongly imbibing plates (Fig. 1E). The imbibition rate data also pointed to limitations of the ab initio spherical cap model of Louge and Sahoo (AIChE J 63, 2017).

The dynamic contact angle behavior was surprising. It did not conform to the models of Cox (J. Fluid Mech. 168, 1986) or Kistler (1993) on flat surfaces. On hydrophilic capillary plates, the angle transitioned from the dry Cassie-Baxter effective angle to its wet counterpart (also called non-hysteretic regimes IV and VI in the statistical mechanics of Louge Phys Rev. E 95, 2017), as the imbibing drop receded before disappearing altogether (Fig. 1C).

The serendipitous repetition of an experiment on a flat plate that was no longer dry, but instead covered with a thin water film from a previous test, confirmed the observation of Biance, et al (Phys. Rev. E 69, 2004) that the inertial phase of spreading is unaffected by presence of such film. However, a similar accidental repetition on a plate with 50% porosity revealed substantially different behavior as the drop spread and imbibed beyond the film.

Lastly, we monitored the time-history of curvature at the neck above the contact patch to test the conjecture of Biance, et al that it is inversely proportional to the square of the patch radius. We could only confirm this conjecture for 25% porosity and a dry Cassie-Baxter angle of 67°, and for a numerical simulation on a flat plate with similar angle. For these conditions, both experiment and simulation exhibited a contact patch power-law exponent near one half, as predicted by the model of Biance, et al. However, in other cases where we could determine curvature, the latter did not appear to uphold the conjecture.