High-Throughput and Low-Cost Precision Fluid Handling for Direct Material Printers in Autonomous Research

Autonomous and machine learning driven research presents abundant opportunities to accelerate materials discovery at the rate needed for rapid climate action. Due to the high data demands of machine learning driven research, high-throughput synthesis methods are essential to successful implementation of accelerated materials research. However, high-throughput synthesis methods are often inaccessible due to their high complexity, commonly employing robotic systems with large space and technical skill demands at a price tag of over $500k. Low-cost, home-built, high-throughput material printing tools offer an accessible platform for high-throughput materials synthesis; these methods, however, can lack both precision and reproducibility.<br/><br/>We address the need for low-cost, high-precision, and high-throughput fluid handling devices for materials synthesis by designing and validating a home-built system capable of dispensing and mixing gradients of up to 10 fluids in the microliter to milliliter scale at speeds of up to 1000 compositions a minute. The system features a carefully curated wetted material path consisting only of chemically inert PTFE, nylon, and borosilicate glass. Three subsystems are used to ensure the functionality of the device: fluid dispensing, fluid mixing, and droplet generation. Fluid dispensing allows for known and measurable dispensing of fluids at ul to ml scale, fluid mixing allows for uniform mixing of all streamlines in low Reynolds number regimes, and droplet generation allows for dispensing uniform droplets. All components are built in-house and contribute to a total price of about $7000, making it an affordable tool when compared to existing fluid-handling systems. Fluorescent dyes of varying wavelengths are used to test and validate each subsystem of the fluid-handling device. Careful and thorough measurements and characterization of the fluid dispensing system are done to understand its operating conditions and associated errors. Hyperspectral images of the system output are taken to ensure proper fluid mixing and the composition of droplets with varying concentrations of dyes.