Shear-triggered coalescence

Control over emulsion stability and phase separation is important for industries such as oil and gas, where it is critical to remove all the oil content from oily wastewater before discharging it into the environment or, conversely, to remove small droplets of water from oil prior to upgrading. Prior work has qualitatively shown that it may be possible for crude oil emulsions to be highly stable at rest and then rapidly destabilized by shearing interactions between droplets due to a phenomenon that we coin shear-triggered coalescence in this paper. In this paper, we provide quantitative evidence of this phenomenon using a cantilevered-capillary force apparatus to precisely manipulate two liquid droplets within another immiscible liquid (i.e., mineral oil in water or water in mineral oil). We first show that droplets in surfactant solutions clearly do not exhibit shear-triggered coalescence because they have the same probability of coalescing when a head-on collision of the droplets is compared to a shearing collision. In contrast, we show that when droplets have spherical Janus microparticles adsorbed onto the interface, the droplets undergoing shearing collisions coalesce faster than those undergoing a head-on collision. Similarly, we show that when rod-shaped nanoparticles (cellulose nanocrystals) are adsorbed onto the interface, with an intermediate salt concentration in the suspending phase, that shear-triggered coalescence occurs in dramatic fashion. We offer mechanistic explanations for why shear-triggered coalescence is possible in these two cases, but is not observed in other cases such as with disklike microparticles, non-Janus spherical microparticles, or rodlike nanoparticles without electrostatic screening or with strong electrostatic screening.

Figure 1

(a) A schematic showing the geometric parameters used to define the collision of two emulsion droplets. (b) Two dilbit droplets in contact in the head-on configuration ($H=0$). (c) Two dilbit droplets in contact during an oscillatory shearing collision ($H=50\mu \mathrm{m}$). See video in Supplemental Material [51]. Note that the time-averaged droplet contact area in the shear configuration ($1025\mu {\mathrm{m}}^2$) is approximately the same as the contact area in the head-on configuration ($1060\mu {\mathrm{m}}^2$).

Figure 2

(a) An oil droplet is formed at the tip of the auxiliary capillary (upper capillary in images) with the chamber and capillaries filled with an aqueous solution. (b) A small amount of oil is sucked into the left capillary. (c) The auxiliary capillary is moved away and an oil droplet is then extruded at the tip of the capillary. The inner diameter of the left capillary is $50\mu \mathrm{m}$.

Figure 3

Cumulative distribution for the probability of coalescence for water droplets in light mineral oil containing 0.05 $\%\mathrm{v}$/v Span 80, light mineral oil droplets in 0.002 mM aqueous Triton X-165 solution, and light mineral oil droplets in 0.001 mM aqueous SDS solution. Filled and unfilled symbols denote head-on and shearing collisions, respectively.

Figure 4

Cumulative distribution for the probability of coalescence of dilbit droplets in deionized water or simulated process water (SPW). Filled and unfilled symbols denote head-on and shearing collisions, respectively.

Figure 5

(a) Upper: A schematic showing how spherical particles with a contact line pinned at 90º (e.g., the spherical Janus microparticles used in this study) might roll during a shearing collision to drag a thin layer of oil and facilitate bridging between the two droplets. (a) Lower: A schematic showing the two other particles used in this study that do not roll (disklike clay microparticles) or might roll without sufficient distortion of the interface to trigger coalescence (spherical silica microparticles). (b) Cumulative distribution for the probability of coalescence of mineral oil droplets in 0.001 wt% silica suspension, 0.0001 wt % Janus silica suspension, and 0.0007 wt % clay suspension. Filled and unfilled symbols denote head-on and shearing collisions, respectively.

Figure 6

(a) Cumulative distribution for the probability of coalescence for mineral oil droplets in 0.04 wt % cellulose nanocrystals in 0, 10, and 100 mM NaCl solution. Filled and unfilled symbols denote head-on and shearing collisions, respectively. (b) A schematic showing how cellulose nanocrystals might reorient and align during a shearing interaction in the presence of 10 mM NaCl.