3D-Printed Closed-Channel Spiral Inertial Microfluidic Device for Size-Based Particle Separation
Year: 2026
Type: article
Abstract: Spiral inertial microfluidic devices provide a simple, high-throughput approach for size-based particle separation; however, translating PDMS-optimized designs into monolithic, fully enclosed 3D-printed channels is often limited by printability and post-print channel clearing. In our previous PDMS study, a 400×120µm spiral achieved high separation performance after computational optimization and experimental validation. To translate this high-performing PDMS concept into a faster and more cost-effective manufacturing approach, the same separation principle is transferred to a fully 3D-printed, closed-channel spiral device, and the geometry is re-optimized around manufacturability constraints. Printing trials showed that enclosed channels at 400×120µm and 600×180µm could not be cleared reliably due to trapped resin and frequent blockage, most often near the inner-outlet region. In contrast, 800×240µm and 1200×360µm channels were printed and flushed successfully, and 800×240µm was selected as the smallest reproducibly functional cross-section. Particle-tracking simulations were then used to re-optimize spiral development length, showing that a 4-turn device provides limited collection for 12µm targets (10%), intermediate lengths (5-7 turns) improve collection to 50%, and an 8-turn spiral achieves complete large-particle collection (100%) across tested target sizes (12-24µm) while reducing small-particle crossover. Experimental validation of the 8-turn 800×240µm device at Q=6mL min-1 using fluorescent polystyrene particles (18µm target; 6µm background) yielded an average collection efficiency of 84% and an inner-outlet purity of 92%. Overall, these results demonstrate that spiral inertial separation can be retained in a monolithic 3D-printed format when the design is re-optimized around the smallest reliably clearable enclosed cross-section and sufficient spiral development length.
Source: Micromachines
DOI: 10.3390/mi17040435
Authors: Eda Ozyilmaz, Gamze Gediz Ilis
Institutions: Gebze Technical University, Northeastern University
Language: English
Subfield: Biomedical Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,534
OpenAlex ID: W7147216505