How Hydrodynamic Flow Focusing Lipid Nanoparticles Microfluidic Works
Lipid nanoparticles (LNPs) are used to deliver genetic medicines such as mRNA vaccines. A critical step in LNP production is mixing lipids dissolved in ethanol with an aqueous buffer containing the nucleic acid cargo. This mixing must happen quickly and uniformly to form stable particles of the right size. Hydrodynamic flow focusing is one of the most direct and controllable methods to achieve this. In this approach, the ethanol-lipid stream is squeezed between two sheath flows of the aqueous buffer inside a microchannel. The focused stream thins to the micrometer or nanometer scale, and the rapid diffusion of ethanol into the water triggers the self-assembly of the lipids into nanoparticles.
The Physics of Flow Focusing
Hydrodynamic flow focusing relies on laminar flow, a condition common in microfluidic channels where fluids move in parallel layers without turbulent mixing. In a typical device, the central stream containing the lipids in ethanol enters a junction. Two side channels deliver the aqueous buffer at higher flow rates. These side streams pinch the central stream into a narrow thread.
The width of this focused thread is determined by the ratio of the flow rates. A higher sheath-to-core flow ratio produces a thinner thread. Because diffusion happens quickly over very short distances, the ethanol rapidly moves into the aqueous layers. This sudden change in solvent quality causes the lipid molecules to precipitate and self-assemble into nanoparticles. The process is governed by the Reynolds number, which remains low (typically less than 100) in these devices, ensuring that mixing occurs by diffusion rather than by chaotic advection Wikipedia — Hydrodynamic trapping.
Controlling Particle Size
The size of the resulting LNPs is a critical quality attribute. It affects biodistribution, cellular uptake, and the release of the therapeutic cargo. In a flow-focusing device, particle size is primarily controlled by three factors:
- Total flow rate: Faster flow rates increase the shear forces on the forming particles, which can reduce particle size.
- Flow rate ratio: A higher ratio of aqueous buffer to ethanol stream produces a thinner focused stream and faster mixing, leading to smaller and more uniform particles.
- Lipid composition and concentration: The specific types of lipids used (ionizable lipids, helper phospholipids, cholesterol, and PEG-lipids) and their total concentration in the ethanol stream directly influence the final particle size and stability.
A typical working range for total flow rate is 0.1 to 10 mL per minute, with flow rate ratios from 1:1 to 5:1 (aqueous to ethanol). The process is stable and reproducible, making it suitable for both research-scale production and early-stage clinical manufacturing National Center for Biotechnology Information.
Device Design and Fabrication
Microfluidic chips for hydrodynamic flow focusing are usually made from polydimethylsiloxane (PDMS) using soft lithography. The process starts with a silicon wafer patterned with SU-8 photoresist to create a raised mold. Liquid PDMS is poured over the mold, cured, and peeled off to form a channel layer. This layer is then bonded to a flat glass slide or another PDMS layer.
A simple flow-focusing device has a cross-slot geometry. The central channel for the lipid-ethanol stream is typically 50 to 200 micrometers wide. The two side channels for the aqueous buffer meet the central channel at a 90-degree angle. The outlet channel, where the LNPs form and exit, can be the same width or slightly wider.
Integration with Control Layers
For more precise control, a second PDMS layer containing pneumatic valves can be bonded on top of the fluidic layer. These valves act as on-off switches that regulate the flow of the lipid stream or the buffer. This two-layer design allows the user to start and stop nanoparticle formation with high temporal precision, which is useful for collecting multiple samples or for automating the process Lab on a Chip (Royal Society of Chemistry).
Comparison with Other Microfluidic Mixing Methods
Hydrodynamic flow focusing is one of several microfluidic approaches for LNP production. Two common alternatives are staggered herringbone mixers and T-junction mixers.
| Method | Mixing Mechanism | Typical Particle Size Range | Key Advantage |
|---|---|---|---|
| Hydrodynamic flow focusing | Diffusion across a thin focused stream | 30–100 nm | Simple geometry, easy to fabricate |
| Staggered herringbone mixer | Chaotic advection from grooves in the channel walls | 30–80 nm | Faster mixing at lower flow rates |
| T-junction mixer | Diffusion at the interface of two streams | 50–200 nm | Very simple design, but less uniform |
Hydrodynamic flow focusing offers a good balance between simplicity and control. While herringbone mixers can achieve slightly smaller and more uniform particles at very low flow rates, flow focusing devices are easier to design, fabricate, and scale up by simply increasing the number of parallel channels.
Scaling Up for Production
One limitation of single-channel hydrodynamic flow focusing devices is their low throughput. A single device running at a total flow rate of 1 mL per minute produces about 1.5 liters of LNP suspension per day. For preclinical studies requiring grams of material, this is acceptable. But for commercial manufacturing of mRNA vaccines, which may require hundreds of millions of doses, the output must be much higher.
The most common solution is parallelization. Multiple flow-focusing channels are arranged on a single chip, each fed by the same lipid and buffer inlets. A 10-channel device can operate at 10 mL per minute while maintaining the same particle size and polydispersity as a single channel. Companies like Precision NanoSystems and Dolomite Microfluidics sell commercial systems that use this principle, with chip designs containing 8, 16, or even 32 parallel channels.
Another approach is to increase the channel dimensions while keeping the flow-focusing principle intact. Wider channels allow higher flow rates, but the focused stream becomes thicker, which slows down mixing and can produce larger particles. To compensate, the flow rate ratio must be increased, which requires higher buffer volumes. This trade-off between throughput and particle quality is a central challenge in process development.
Common Pitfalls and Troubleshooting
Several issues can arise during hydrodynamic flow focusing of LNPs. One frequent problem is clogging at the mixing junction. This happens when the lipid-ethanol stream precipitates too quickly, forming large aggregates that block the channel. Reducing the lipid concentration in the ethanol stream or lowering the flow rate ratio can help. Some researchers add a small amount of a co-solvent like dimethyl sulfoxide to the lipid stream to improve solubility.
Another issue is air bubble formation. Ethanol has a lower surface tension than water, and the rapid mixing can trap tiny air bubbles in the outlet channel. These bubbles scatter light and interfere with inline particle size measurements. Degassing the buffer and lipid solutions under vacuum before use reduces this problem. Priming the chip with buffer before introducing the lipid stream also helps.
A third common problem is batch-to-batch variability in particle size. Even with the same flow rates and lipid composition, different chips or different days can produce slightly different results. This is often due to small variations in channel dimensions from the soft lithography process or to changes in ambient temperature. Running a calibration test with a standard formulation at the start of each production session helps establish a baseline.
Key Takeaways
- Hydrodynamic flow focusing uses laminar flow to squeeze a lipid-ethanol stream between two aqueous buffer streams, allowing rapid diffusion-driven mixing.
- Particle size is controlled by the total flow rate, the flow rate ratio, and the lipid composition.
- Devices are typically fabricated from PDMS using soft lithography and can include pneumatic valves for precise flow control.
- This method produces LNPs in the 30–100 nm range and is well-suited for research and early-stage clinical production.
