A four-roll mill was used to experimentally investigate the coalescence of two equal-sized drops in general linear flows. The experimental system consisted of polybutadiene drops suspended in polydimethylsiloxane. Under the experimental conditions studied, the bulk-phase rheological properties of both fluids are Newtonian. We studied both head-on collisions for a purely extensional “hyperbolic” flow that always lead to coalescence, and collisions with a finite offset from the inflow axis for several different flow types produced in the four-roll mill. The experimental results have been compared with approximate theoretical predictions of coalescence, based on an asymptotic theory for small capillary number, where the drops are spherical apart from a small planar deformation at the frontal surfaces between the two drops. In head-on collisions, it was found experimentally that the product of the film drainage time and strain rate is independent of capillary number (Ca) and drop radius at very low Ca. This scaling behavior agrees qualitatively with theoretical predictions from the approximate small Ca theory. At higher values of Ca, this dimensionless drainage time varies as Ca3/2. This scaling is also consistent with theoretical predictions. For collisions with a nonzero offset from the head-on configuration, coalescence occurs only for capillary numbers below a critical value, Cac. Measurements were made of Cac as a function of the drop size and the flow type, for various values of the offset. The critical capillary number for coalescence was found to decrease with increasing offset, in qualitative agreement with predictions from the theoretical model. However, these Cac versus offset results do not agree quantitatively with the theoretical predictions. In the model the minimum film thickness occurs when the two-drop pair has rotated to the angle at which the external flow just begins to pull them apart. However, for configurations with small but nonzero offsets, it is found experimentally that coalescence occurs earlier in the collision process. Thus, the actual time available for film drainage is shorter, and the critical capillary number is smaller than what is predicted by the model. At the same time, for larger offsets, it is shown experimentally that the collision and initiation of film drainage is delayed relative to what is predicted, and thus there are offset values where the model predicts that coalescence is possible, whereas, in fact, no coalescence is observed.

No abstract

No abstract

No abstract

No abstract