We have exploited a compact and facile microfluidic droplet creation device consisting of a poly(dimethylsiloxane) microfluidic chip possessing T-junction channel geometry, two inlet reservoirs, and one outlet reservoir, and a piezoelectric (PZT) diaphragm micropump with controller. Air was evacuated from the outlet reservoir using the PZT pump, reducing the pressure inside. The reduced pressure within the outlet reservoir pulled oil and aqueous solution preloaded in the inlet reservoirs into the microchannels, which then merged at the T-junction, successfully forming water-in-oil emulsion droplets at a rate of ∼1000 per second with minimal sample loss. We confirmed that the onset of droplet formation occurred immediately after turning on the pump (<1 s). Over repeated runs, droplet formation was highly reproducible, with droplet size purity (polydispersity, <4%) comparable to that achieved using other microfluidic droplet preparation techniques. We also demonstrated single-molecule PCR amplification in the created droplets, suggesting that the device could be used for effective droplet digital PCR platforms in most laboratories without requiring great expense, space, or time for acquiring technical skills.

We previously developed a technique that enabled automatic creation of monodisperse water-in-oil droplets with the use of an air-evacuated PDMS microfluidic device. Although the device generated droplets over a long-time period, the production rate was slow (∼10 droplets per second). In the current study, we aimed to improve this rate, using the same fluid pumping principle described in our previous work, by remodeling our device configuration. To achieve this aim, we developed a new device with a much larger PDMS surface area-to-volume ratio within the air-trapping void space (178 cm ), than that of our earlier device (5.0 cm ). This design approach was based on the idea that a larger PDMS surface area-to-volume ratio was likely to create a higher vacuum inside the void space, thereby contributing to faster liquid flow and an increased droplet generation rate. The new device consisting of five layers featuring a degassed PDMS slab as a detachable liquid-suction actuator, which was stacked on a lower microfluidic layer. In this device, the rate of droplet production increased during the time-course droplet formation and reached ca. 470 droplets per second immediately before completely consuming the loaded aqueous solution (20 μL).

We previously established an automatic droplet‐creation technique that only required air evacuation of a PDMS microfluidic device prior to use. Although the rate of droplet production with this technique was originally slow (∼10 droplets per second), this was greatly improved (∼470 droplets per second) in our recent study by remodeling the original device configuration. This improvement was realized by the addition of a degassed PDMS layer with a large surface area‐to‐volume ratio that served as a powerful vacuum generator. However, the incorporation of the additional PDMS layer (which was separate from the microfluidic PDMS layer itself) into the device required reversible bonding of five different layers. In the current study, we aimed to simplify the device architecture by reducing the number of constituent layers for enhancing usability of this microfluidic droplet generator while retaining its rapid production rate. The new device consisted of three layers. This comprised a degassed PDMS slab with microfluidic channels on one surface and tens of thousands of vacuum‐generating micropillars on the other surface, which was simply sandwiched by PMMA layers. Despite its simplified configuration, this new device created monodisperse droplets at an even faster rate (>1000 droplets per second).