Design automation of microfluidic single and double emulsion droplets with machine learning

Comprehensive droplet generation dataset

To generate a comprehensive dataset detailing the impacts of device designs, flow rates, and fluid properties on droplet diameters and generation rates, we curated and combined two previously generated SE and DE experimental datasets^32,37. This comprehensive dataset includes 46 different polydimethylsiloxane (PDMS) and polycarbonate device designs (43 SE and 3 DE generators with 49 flow-focusing geometries combined), 8 different dispersed fluids, and 6 different continuous fluids for generating aqueous-in-oil and oil-in-aqueous droplets of…