Automated Microfluidic Droplet Handling System for Precision Nanoliter-Scale Fluid Operations
This microfluidic platform uses software-controlled membrane displacement traps and real-time vision to manipulate nanoliter droplets and particles—generating, capturing, sorting, splitting, merging, and ejecting them for automated chemical and biological assays.
Background
In many areas of life science and chemical analysis, tiny droplets can serve as reaction vessels, enabling high-throughput experiments with minimal reagent use. Standard laboratory robots and micropipettes struggle to handle volumes below the microliter scale with consistency, leading to wasted materials and reduced experimental fidelity. As research pushes toward single-cell assays, genetic analyses, and combinatorial chemistry, there is a pressing demand for systems that can generate, move, and combine droplets at the nanoliter or picoliter scale. Such precision handling promises to accelerate discovery while cutting the cost and time required for many assays.
Despite advances in microscale fluidic devices, current approaches are often limited by rigid channel geometries and lack of real-time control, which restricts the range of operations that can be performed. Many designs rely on fixed valve networks and manual adjustments, making it difficult to adapt protocols or correct errors on the fly. Delays in detection or actuation can cause droplets to miss capture sites, leading to losses or unpredictable mixing. Furthermore, dead volume in connecting passages and low feedback rates contribute to poor reproducibility between runs. The absence of flexible software interfaces and integrated sensing also means users must juggle multiple tools, hindering automation and scaling.
Technology
This platform uses a small, two-layer microfluidic chip to handle nanoliter droplets with high precision. In the top layer, fluid channels feed into tiny circular wells, each underlaid by a thin, flexible membrane that can be bent by pressure to capture, split, merge or eject droplets on demand. A built-in camera and computer vision software track droplets in real time and trigger the membrane traps, while programmable pressure controls and a scripting interface coordinate multiple operations such as droplet generation, metering and on-chip mixing. By adjusting flow rates, valve timings and membrane actuation, droplet sizes from picoliters to nanoliters can be produced with better than 5% size consistency.
This approach stands out by combining automated vision-guided feedback with membrane-based trapping to deliver truly deterministic droplet handling. Unlike passive or random droplet formation methods, it actively compensates for timing delays and flow variations, enabling complex sequences—like stepwise pH ladder formation—in a single chip. Its software-driven control and compact design make it more flexible, scalable and precise than traditional robotic liquid handlers or fixed-pattern droplet devices, opening new possibilities for biochemical assays and single-cell studies.
Benefits
- Precise, programmable control of droplets and particles at nanoliter volumes
- Automated, high-throughput operation via integrated computer vision and FluidScript software
- Versatile droplet operations (generation, capture, splitting, merging, ejection) in a single platform
- Minimal reagent consumption and reduced dead volume for cost-effective workflows
- Real-time feedback and adjustment for improved reproducibility and accuracy
- Compatibility with standard PDMS/glass chips and common oil/surfactant systems
- Scalable design using pneumatic multiplexers for larger array implementations
- Broad applicability in biomedical assays (digital PCR, whole-genome amplification, single-cell analysis)
Applications
- Digital PCR diagnostics
- Single-cell RNA sequencing
- Whole-genome amplification
- High-throughput drug screening
- Automated nanoliter liquid handling
Additional info
A microfluidic platform integrates a two-layer PDMS chip with membrane displacement traps (MDTs) and hydraulic valves, a computer vision system, and Python-based control to perform on-demand nanoliter droplet operations. Real-time imaging guides MDT actuation via pressure-driven flows through an H-bridge topology, enabling reproducible droplet generation (80 pL–2 nL), capture, splitting, merging, ejection, and metering. FluidScript sequences automate assays like pH ladder formation for high-throughput biochemical workflows.
PUBLICATION
“Programmable Control of Nanoliter Droplet Arrays Using Membrane Displacement Traps” Jason Harriot, Michael Yeh, Mani Pabba, Don L. DeVoe
Advanced Materials Technologies Vol. 8 Issue 21: 15 August 2023
PATENT pending
US 19/052810 filed February 13, 2025