Upscaling microstructured emulsification devices

Emulsions, which are dispersions of two immiscible liquids (e.g. oil and water), are part of our daily life through many products that we use such as milk, mayonnaise, salad dressings, ice cream, lotions, shampoos, medicines, wall paints, etc. Many quality attributes of these products such as stability, texture, colour, visual and sensorial perception are affected by droplet size and size distribution.

Conventional emulsification technologies such as high pressure homogenizers have poor control on droplet size distribution, they are energy intensive and not suited for fragile multiple emulsions. In the last decades, alternative emulsification concepts that employ microengineered structures have been developed. They can produce uniform droplets of a specific size using orders of magnitude less energy, and are suitable for multiple emulsions.

However, most of these techniques generate one droplet at a time and the productivity of a single droplet generation unit is very low. To reach significant throughput, many units need to be run in parallel, which is far from trivial especially for the production of droplets below 10 micrometres. In this regard, EDGE (Edge-based Droplet GEneration) devices are better suited for upscaling since they can generate multiple uniform droplets simultaneously from one droplet formation unit.

Unlike standard upscaling in industry, the characteristic dimension remains the same for microstructured (EDGE) devices, and issues related to upscaling were found to be linked to (sub-) micrometre scale (e.g. wettability and flow geometry). In EDGE emulsification, the contact surfaces need to be wetted well by the continuous phase, and in chapter 2 we show that the interactions of the liquids and surfactants with all available surfaces/interfaces influence wettability. In general, oils that have strong interaction with the surface can only be emulsified successfully in combination with surfactants that bind strongly to the surface. Also the pressure range in which droplets can be produced is greatly influenced by these interactions, e.g. proteins showed much wider pressure stability and an order of magnitude higher productivity, therewith also showing that EDGE emulsification is well suited for food-grade emulsions; that is as long as an appropriate combination of construction material and emulsion components is used.

Also the geometry of the EDGE devices can be used to increase productivity. Previous research indicated that higher resistance on the plateau can improve the pressure stability, which inspired us to redesign the droplet formation units and place regularly spaced micron-sized partitions on the main plateaus, as reported in chapter 3. The micro-plateaus were positioned such that the number of droplet formation points was increased compared to regular EDGE, and it was found that the additional flow resistance resulted in remarkably wide pressure range while supplying oil to all micro-plateaus that were equally active, thereby leading to two orders of magnitude higher droplet productivity. Interestingly, at high pressures a second wide range generating approximately three times larger uniform droplets was discovered.

In chapter 3, only one partitioned EDGE geometry was investigated for hexadecane and 0.5% SDS solution; therefore, in chapter 4 the underlying droplet formation mechanisms was investigated further by systematically varying the geometry of the micro-plateaus and the viscosity of the liquids. It was found that the micro-plateau geometry greatly influenced emulsification behaviour. The second regime, in which large droplets were formed, was only observed for narrow micro-plateaus, suggesting that a certain minimum flow resistance is needed for the second regime to occur. In the first regime, in which small droplets were formed, droplet size was dependent on the viscosity ratio of the liquids, in a similar way to that found for regular EDGE devices.

The partitioned EDGE devices were upscaled in chapter 5 with 75000 micro-plateaus. This first upscaled device, the so-called multi-EDGE, was used to produce monodisperse hexadecane droplets of ~10 micrometres at 0.3 m3 m-2 h-1 (80% micro-plateau activation). As expected, the differences in plateau geometry (due to technical limitations) compared to the devices reported in chapter 3 led to an order of magnitude lower productivity. Nonetheless, the initial results were promising, and provided clear leads to improve the productivity further. Last but not least, with the current multi-EDGE device enough product can be made to conduct rheology and stability tests for truly monodisperse emulsions.

In chapter 6, we studied a different microstructured device, the packed bed premix emulsification equipment, and showed that food-grade double emulsions (containing 5% v/v primary emulsion) can be refined at high throughputs, typically in the range of 100‑800 m3 m-2 h-1, while keeping their encapsulation yield above 90%. Droplet size reduction was similar to that found for single emulsions; the refined droplets were smaller than the pore sizes of the packed bed, and no marked fouling was observed under the conditions tested. Further, the process was robust and reproducible, making the technique a genuine option for double emulsion production.

In the last chapter, chapter 7, we compare microstructured emulsification techniques on various aspects, and explain how the findings of this thesis help mitigate the identified bottlenecks (e.g. wettability, parallelization, productivity) that prevent upscaling of the technology. Finally, we conclude with an outlook on upscaling and discuss the aspects related to possible applications of the technology in the future.