(K-408) Mass Manufacture of Hydrogel Beads using Step Emulsification

Step emulsification is a droplet generation method that uses the device geometry, rather than shear forces, to generate droplets due to a Laplace pressure difference in different regions of the device. With this technique, the size of droplets (or beads) produced depends more on material properties than the rate at which the dispersed phase flows (3). Step emulsification can be used to stably and reliably synthesize hydrogel beads en mass using parallel bead generation channels. These beads can then be used for a number of different applications, including sequencing preparation and 3D cell cultures(1) and tumor or organoid models(5).

As with most microfluidic devices, the most common way to manufacture a device mold is through photolithography. While this method is very popular due to its ability to manufacture dimensions at the microscale that can be transferred to the final PDMS microfluidic device, it requires specialized equipment and training. 3D printing has emerged as a potential alternative mold fabrication method, owing to its relatively lower cost and training barrier. In addition, 3D-printed molds can often be produced in less than 30 minutes, compared to hours in a specialized microfabrication facility. Despite these advantages, 3D-printed molds are still a less common mold fabrication technique, and their adoption has been hindered by concerns that include the resolution of features possible with this technique. The benefits of a 3D-printed mold for microfluidic devices are valuable and deserves deeper exploration as access to 3D-printing technology has grown in the last years.

SolidWorks was used to design the 3D-printed molds. The design consists of 10 parallel microfluidic channels, arranged radially with a central inlet and each of the channels opening into a large reservoir. Step emulsification requires a step height difference from these channels to the reservoir to cause an imbalance in Laplace pressure and ultimately break off the fluid as a bead. Each nozzle is designed to have a channel height of 60 μm and a nozzle opening of 390 μm. The ratio of nozzle opening to channel height is 6.5, fitting the 5.5 to 19 ratio that is needed for monodispersed beads to form(2). Neutralized collagen solution was selected as the hydrogel (dispersed) fluid while mineral oil was selected as the immiscible continuous fluid. 

A COMSOL simulation of bead formation in the device was run to predict the dimensions of the beads. Unlike other droplet or bead production methods, step emulsification has not been modeled extensively in literature. Using T-junction COMSOL models as inspiration, a two-phase multiphysics simulation was run for 60 minutes. The simulation included averaged recorded viscosity and density for both collagen solution and mineral oil. In order to replicate the continuous fluid (mineral oil) and the reservoir, a rectangular prism was drawn around the nozzle. 

A CADworks 3D (M50-405) 3D printer with CADworks 3D Microfluidic Photopolymer Resin was used to produce the microfluidic device mold. The PDMS cast was permanently bonded to a glass microscope slide via plasma bonding and then annealed using a hot plate.

An optical profilometer captured the dimensions of the channel height and the nozzle opening of the 3D printed mold to be 56 um and 334.9 um, respectively. These values are close to the designed dimensions of 60 um and 390 um, respectively. The critical design ratio (of nozzle opening: channel height) was 6.5 as designed and was 5.98 as printed. Importantly, this means that the printed mold’s aspect ratio falls within the range of 5.5 to 19 required for bead formation using step emulsification(2). 

COMSOL studies predicted a bead diameter of about 210 um, and the size of the beads produced from the device agreed well with this prediction, having an average diameter of 253.6 um + 44.6 um (Figure 2, right). The diameters of droplets generated from each of the 10 nozzles in the device were compared using ANOVA. The test produced an alpha value of 0.59665 which was greater than our defined critical value of 0.05. Thus, there is no statistically significant difference between bead sizes generated from different nozzles. 

In this work, 3D-Printed molds served as masters for producing PDMS microfluidic devices (Figure 2, left) that were used successfully to generate collagen microbeads in a parallel fashion, utilizing step emulsification for robust and steady bead generation. 3D-printed molds enable the fabrication of microfluidic devices for labs and facilities lacking access to traditional photolithography equipment and increase accessibility while decreasing the associated cost of adopting microfluidic technology for applications in droplet microfluidics including genetic screening and creation of 3D cell culture systems.