Abstract
Capillary microfluidic devices provide minimally-instrumented and user-friendly liquid handling for a range of biochemical applications. 3D printing is an attractive method for microfabricating capillary microfluidics because it requires little training, is cost-effective compared to conventional cleanroom fabrication, and can drastically decrease prototyping time. High-resolution (≤ 100 μm) microfluidics are typically 3D printed using digital light projection (DLP) stereolithography (SLA); however, the high capital cost (>$10,000) of high-resolution DLP-SLA printers limits their accessibility and widespread use. Recent advances in liquid crystal display (LCD) technology have provided inexpensive (<$500) masked stereolithography (mSLA) 3D printers with sufficient optical resolution for microfluidics. Despite their potential, there have only been a few demonstrations of microfluidic fabrication with mSLA printers, limited characterization of the effects of printing parameters like UV power and exposure time on print quality, and no direct comparisons with DLP-SLA printing. We compared a 40-μm pixel resolution DLP-SLA printer (∼$18,000) with a 34.4-μm pixel resolution (∼$380) mSLA printer. When printing with the same resin and optimized parameters for each printer, the DLP-SLA printer produced features with an average 4.8% difference between designed and measured XY dimensions and a coefficient of variation [CV] = 2.17 ± 2.53% (N=3). The 34.4 μm mSLA printer produced features that were 15.3 ± 11.4% larger than designed dimensions, although they were very precise [CV = 1.80 ± 1.11%, N =3]. Using a low-cost, water-wash resin optimized for the mSLA printer, microchannels printed with the mSLA differed from designed dimensions by only 4.20 ± 4.83%. To demonstrate that we could optimize print conditions for multiple mSLA printers, we printed microchannels with a 22 μm mSLA printer (∼$450) and obtained features that were on average 2.43 ± 5.88% different from designed dimensions. We 3D-printed capillary microfluidic circuits – composed of stop valves, trigger valves, retention burst valves, retention valves, and domino valves – for automating sequential delivery of five liquids. Taken together, our results show that mSLA printers are an inexpensive alternative for fabricating capillary microfluidics with minimal differences in fidelity and accuracy of features compared with a 20X more expensive DLP-SLA printer.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
