Emulsification in microdevices (microfluidic emulsification) involves micrometer-sized droplets and fast interface expansion rates. In addition, droplets are formed in less than milliseconds, and therefore traditional tensiometric techniques cannot be used to quantify the actual interfacial tension. In this paper, monodisperse droplets formed at flat microfluidic Y-junctions were used to quantify the apparent dynamic interfacial tension during (microfluidic) emulsification. Hexadecane droplets were formed in ethanol-water solutions with a range of static interfacial tensions to derive a calibration curve, which was subsequently used to access the dynamic interfacial tension of hexadecane droplets formed in surfactant solutions. For SDS and Synperonic PEF108, various continuous- and disperse-phase (hexadecane) flow rates were studied, and these conditions were linked to interfacial tension effects, which also allowed convective transport of surfactants to be investiagted. On the basis of these findings, various strategies for the formation of emulsion droplets can be followed and are discussed.

Y junctions with a large width-to-depth ratio were used for the emulsification of hexadecane in various ethanol-water mixtures with different static interfacial tension and viscosity. The resulting droplets were monodisperse. To describe droplet size a force-balance model was derived and was found to apply well. The model shows that the droplet size scales with the channel depth, and with the square root of the inverse capillary number (Ca-1/2) based on the continuous phase, the disperse-phase flow rate was found to be unimportant.

No abstract

No abstract

In a previous article, we studied the basics of emulsification in microfluidic Y-junctions, however, without considering the effect of viscosity of the disperse phase. As it is known from investigations on many different microstructures that viscosity and viscosity ratio are governing parameters for droplet size, we here investigate whether this is also the case for microfluidic Y-junctions and do so for a wide range of process conditions. The investigated Y-junctions have a width of 19.9 or 12.8 lm and a depth of 5.0 lm, and the formed monodisperse droplets (CV \ 1%) are between 3 and 20 lm. We varied the disperse-phase viscosity using different oils (1-105 mPa s), and continuous-phase viscosity using glycerol-water and ethanol-water mixtures (1.0-6.2 mPa s), which corresponds to disperse-to-continuous-phase viscosity ratios from 0.4 to 105.0. Through the variation of the liquids, also a range in interfacial tensions (12-55 mN m À1) is assessed. The disperse-phase flow rate is varied from 0.039 to 18.0 lL h À1 , the continuous-phase flow rate from 1.39 lL h À1 to 0.41 mL h À1, and this corresponds to flow rate ratios from 1.1 Â 10 À3 to 0.14, which is once again based on wide range of conditions. For all these conditions, in which droplets are formed in the dripping and jetting regime, the droplet size could be described with a model based on the existing force-balance model, but now extended to incorporate the cross-sectional area of the droplet and the resistance with the wall. Surprisingly enough, it was found that the droplet size is not influenced by the disperse-phase viscosity, or the viscosity ratio, but it is dominated by the resistance with the wall and the continuous-phase properties. Because of this, emulsification with Y-junctions is intrinsically simpler than any other shear-based method as droplet size is only determined by the continuous phase.