Room temperature roll-to-roll additive manufacturing of polydimethylsiloxane-based centrifugal microfluidic device for on-site isolation of ribonucleic acid from whole blood

March 12, 2026October 17, 2023 by CadWorks3DAdmin

The rendered image of the room-temperature roll-to-roll additive manufacturing platform

Academic Article

Room temperature roll-to-roll additive manufacturing of polydimethylsiloxane-based centrifugal microfluidic device for on-site isolation of riboneucleic acid from whole blood

by Trung Hoang, Han Truong, Jiyeon Han, Saebom Lee, Jihyeong Lee, Sajjan Parajuli, Jinkee Lee and Gyoujin Cho

Abstract: Polymer-based lab-on-a-disc (LoaD) devices for isolating ribonucleic acid (RNA) from whole blood samples have gained considerable attention for accurate biomedical analysis and point-of-care diagnostics. However, the mass production of these devices remains challenging in manufacturing cost and sustainability, primarily due to the utilization of a laser cutter or router computer numerical control (CNC) machine for engraving and cutting plastics in the conventional prototyping process. Herein, we reported the first energy-efficient room-temperature printing-imprinting integrated roll-to-roll manufacturing platform for mass production of a polydimethylsiloxane (PDMS)-based LoaD to on-site isolate ribonucleic acid (RNA) from undiluted blood samples. We significantly reduced energy consumption and eliminated thermal expansion variations between the mold, substrate, and resists by accelerating the PDMS curing time to less than 10 min at room temperature without using heat or ultraviolet radiation. The additive manufacturing technology was applied to fabricate a multi-depth flexible polymer mold that integrated macro (2 mm) and micro-sized (500 μm) features, which overcomes the economic and environmental challenges of conventional molding techniques. Our integrated R2R platform was enabled to print adhesion-promoting films at the first printing unit and continuously in-line imprint with a high replication accuracy (99%) for high-volume manufacturing of a new centrifugal microfluidic chip with an enhancement of mixing performance by integrating an efficient mixing chamber and serpentine micromixer. This research paved the way for scalable green manufacturing of large-volume polymer-based microfluidic devices, often required in real-world sample-driven analytical systems for clinical bioanalysis.

Keywords: room-temperature PDMS; centrifugal microfluidic; RNA extraction; roll-to-roll nanoimprint lithography; sustainable manufacturing

We kindly thank the researchers at Sungkyunkwan University for this collaboration, and for sharing the results obtained with their CADworks3D system.

1. Introduction

In the last two decades, microfluidic systems have evolved rapidly for numerous chemical, biomedical, biological, and environmental applications [[1], [2], [3], [4]]. Among those devices, lab-on-a-disc (LoaD) platforms are gaining much attraction for biomedical applications due to the ability to integrate rapid sample preparation stages such as the isolation of nucleic acids from large-volume whole blood samples [5] with assays. The LoaD devices offer significant advantages over other microfluidic systems, especially a reliable, controllable, and compact pumping mechanism that enables efficient mixing of reagents [6], rapid response time, and enhanced assay sensitivity. In this platform, fluid flow is driven by centrifugal pumping which involves minimal instrumentation by using only a simple and compact motor to generate the force. This approach eliminates the need for external interconnects and syringe pumps, thereby preventing contamination of the sample by the surrounding environment [7]. By combining the benefits of both microfluidics and centrifugal forces in a single device, the centrifugal microfluidic technology has been identified as a standard tool for mainstream diagnostics especially point-of-care in vitro diagnostics (IVD), and achieved significant commercial success [8].

A typical LoaD consists of a multilayer platform made from thermoplastics, such as polycarbonate (PC), poly (methyl methacrylate) (PMMA), polystyrene (PS), cycloolefin polymer (COP), assembled by adhesive or through hot embossing and injection molding methods [[9], [10], [11]], which is cost-effective for high-volume manufacturing process. These devices can also be fabricated through a laser or router computer numerical control (CNC) machine for engraving and cutting plastics, in conjunction with the utilization of plotter machines to cut the adhesive film [[12], [13], [14]]. However, for CNC-manufactured microfluidic devices, the surface becomes extremely rough, leading to slow and inaccurate fluid flow and bonding inhibition as well as non-specific binding which can negatively impact the performance of microfluidic devices. Also, the utilization of these subtractive manufacturing systems has been constrained by the capability of the cutting tools, making the fabrication process of micro-scale features even more expensive, lengthy, and complex, limiting the fabrication throughput, and presenting challenges for mass production ability.

Among polymer-based materials, polydimethylsiloxane (PDMS) has been widely used to fabricate microfluidic devices via soft lithography [[15], [16], [17], [18], [19]]. The use of PDMS in manufacturing LoaD devices has been considered as an alternative strategy to the CNC-based one due to its capability to reduce production costs, increase flexibility, facilitate ease of fabrication, and permit rapid prototyping without the use of harmful etching chemicals. The precursors required for preparing PDMS, comprising prepolymers and curing agents, are notable for their cost-effectiveness and widespread commercial availability. The fabrication process of PDMS-based microfluidic devices can be executed without the utilization of specialized cleanroom facilities enabling rapid prototyping of devices at a lower cost than what is feasible using silicon technology [16]. Moreover, the surface properties of PDMS can be easily tuned, enabling the ability to bond with many materials like glass and PDMS itself [20]. The optical transparency and gas permeability of PDMS-based microfluidic devices are ideal for numerous biomedical applications, particularly in optical detection methods and cell culture. Therefore, large-scale manufacturing methods of PDMS-based microfluidic devices have recently gained much importance in various research areas of science and engineering to bring the usages of these devices to practical clinical applications.

Roll-to-roll (R2R) nanoimprint lithography is considered as an up-and-coming alternative to traditional manufacturing methodologies, owing to its ability to achieve high-throughput production and thus facilitate its application at an industrial scale [21]. PDMS was first demonstrated to be R2R processable by Ahn and Guo [22] (2008) for sub-micrometre test structures on a polyethylene terephthalate (PET) substrate. Later, Hiltunen proposed R2R fabrication of integrated PDMS-paper microfluidics for molecular diagnostics [19]. However, both techniques relied on thermal imprinting to cure PDMS resist by heating the imprinting roll to a high temperature with a large amount of energy. This requires roll modifications with complex and expensive auxiliary systems such as piping networks to circulate the heating liquids, the pre-heating and heating units, cooling system as well as temperature monitoring and control system, which increase waste heat and energy consumption, consequently causing greenhouse gas emissions. Additionally, the mismatch in thermal expansion coefficients between the mold and substrate results in the generation of lateral strain, which degrades the quality of the imprinted patterns and the lifetime of the mold [23]. Also, the molds used in these proposed methods were limited to a micrometre scale, making them impossible to apply for the fabrication of large-volume LoaD which requires millimetre-scale thickness.

To overcome those issues mentioned above, we developed the first room-temperature printing – imprinting integrated roll-to-roll (R2R) in-line manufacturing platform for mass production of a PDMS-based LoaD for on-site RNA isolation from whole blood samples with low cost, less energy consumption, and less by-products. In this work, multi-depth master stamps were first fabricated by using a 3D printer, overcoming the challenges of conventional CNC-based and photolithographic molding technology. The large area flexible polymer shim was fabricated by using a rubber sheet to stick PDMS molds which were replicated from 3D-printed stamps. For mass producing the large volume LoaD by R2R technology, the printing unit was integrated in-line with an imprinting platform for coating an adhesion promoter onto PET substrate to facilitate the demolding process. We incorporated well-defined compounds into commercial PDMS Sylgard 184 formulations to accelerate the curing time of PDMS at room temperature enabling the success of a low-temperature R2R imprinting process which helped to reduce heat waste and energy consumption. Finally, as a proof-of-concept study, novel LoaD devices with a high enhancement of mixing performance were sustainably manufactured by our green R2R platform. For the first time, the utility of these R2R-manufactured LoaD devices was demonstrated by isolating RNA from undiluted blood samples.

Apparatus Used

Master Mold for PDMS

Learn More

Curezone

Learn More

PR110
3D Printer

Legacy

2. Material and methods

2.1. Fabrication of negative 3D-printed mold

Our LoaD device comprises two layers: the designed patterns layer and a blank PDMS as a coverlid. The 3D model of the top part was designed by using SOLIDWORKS software, which was then exported as an STL file. We used the Utility Version 6.3.0t3 software to convert STL files into sliced files with appropriate layer heights. Printing options were optimized for high-resolution printing of mold features by using a 3D printer (PR110-385 from CADworks3D company) utilizing a 385 nm light source with a printing area of 110 × 62 × 120 mm and XY resolution of 40 μm. In this work, we printed all masters using the 3D printing protocol with the following parameters: a UV projection time per layer: 9.5 s and a z-axis resolution: 50 μm. It has been shown that 3D-printed templates interfere with the curing of PDMS due to uncured resin remaining on their surface [24]. Therefore, a post-print surface treatment technique is commonly necessary to overcome the aforementioned issues and prevent the adhesion of PDMS to the 3D-printed mold. A treatment method involving coating the template with a protective ink using airbrushing was reported by Comina et al. [25]. The group claimed this technique requires much effort for achieving optimal results. Ho et al. proposed another complicated surface treatment method for 3D printed templates, including heating, plasma treatment, and surface silanization [24]. However, cracks were formed in the template during the heating process. To avoid these time-consuming, nonreplicable, and ineffective surface treatment methods, we fabricated master molds directly via the DLP 3D printing method utilizing a novel resin formulated by Creative CADworks company (CCW Master Mold for PDMS devices), composed of methacrylated oligomers and monomers. After removing printed molds from the picker, they were subjected to a thorough rinsing procedure utilizing isopropanol. Subsequently, an air nozzle was employed to eliminate residual resin from edges and within extremely fine features. Finally, we postcured the molds by exposing each part to UV light at a wavelength of 405 ± 5 nm within a curing chamber, and then, the resulting molds were employed in our imprinting works.

2.2. Fabrication of positive PDMS mold

To fabricate positive PDMS mold, we prepared a mixture including PDMS base and curing agent in the ratio (W/W) of 10:1 (Sylgard 184 from Dow Corning, MI, USA). The mixture was mixed and defoamed with a centrifugal mixer (Thinky Mixer ARE-310) for 3 and 2 min, respectively, and then poured onto the 3D-printed molds without undergoing any surface treatment process. Later, we cured it in the oven for 4 h at 60 °C. Subsequently, the cured PDMS was carefully detached from the molds and cleaned thoroughly with isopropyl alcohol (IPA) and ethanol at least three times, followed by air drying between each wash. To be used as the mold for the R2R imprinting process, we applied a conformal coating of parylene C which serves as an anti-adhesion layer for demoulding these positive PDMS replicas [26]. Finally, a flexo plate with 5 mm thickness was cut with precise squares, used as the substrate for inserting PDMS positive molds, and wrapped to form a sleeve for preparing the imprinting cylinder.

2.3. PDMS formulations for R2R imprinting process

We formulated a new recipe for fast curable PDMS at room temperature (named Room-Temp PDMS in this paper) by adding compounds (the Ashby–Karstedt catalyst and tetrakis (dimethylsiloxy) silane) into commercial PDMS Sylgard 184 formulations [27]. SYLGARD™ 184 Silicone Elastomer and curing agent were supplied by Dow. Ashby–Karstedt catalyst (platinumcyclovinylmethyl-siloxane complex; 2 % Pt (0) in cyclomethylvinylsiloxanes) and Tetrakis (dimethylsiloxy) silane (TDS) were supplied by Gelest. We used a ratio of 10:1 (w/w) for the elastomer and curing agent to make the standard Sylgard 184 framework. In this study, these compounds were always added following the optimized recipe in exact order: (1) TDS 2 wt%; (2) Elastomer base 88.95 wt%; (3) Curing Agent 8.9 wt%; and (4) Ashby–Karstedt catalyst 0.15 wt%. After all the chemicals had been added, we placed the mixture in Reactor-Ready included circulator Huber Ministat 230, both were supplied by Radleys company (Fig. S8) for continuous mixing at 500 rpm and cooling at −15 °C to prolong the lifetime of PDMS for the imprinting process. After mixing for 1 h, we turned off the mixer and set the reactor at a vacuum of 0.2 psi for degassing in 1 h meanwhile maintaining cooling during the whole process. For comparison, we also prepared the standard PDMS Sylgard 184 as control samples (named Sylgard 184 in this paper) and the reference samples (named Vinyl-terminated PDMS in this paper) by optimizing the recipe of a fast, thermal-curable liquid resist [28] based on a modified PDMS. The formulation for this reference sample consists of three components: vinyl-terminated PDMS was supplied by Gelest, poly (dimethylsiloxane-co-methylhydrosiloxane) trimethylsilyl terminated which acts as a crosslinker and platinum (0)-1,3- divinyl-1,1,3,3-tetramethyldisiloxane complex solution as a catalyst, both were supplied by Sigma-Aldrich. In our optimized recipe, we mixed a ratio of 5:1 (w/w) for the vinyl-terminated PDMS and crosslinker, then added Pt catalyst into the mixer at a concentration of 470 ppm. This formulation effectively accelerated the curing time of this PDMS-based resist at room temperature, making it possible to be used for the R2R imprinting process.

2.4. Roll-to-roll imprinting process and chip assembly

The replication by the R2R imprinting process was conducted at room temperature through an imprinting unit as illustrated in Fig. S1. First, we treated a roll of 150 mm wide PET substrate with an adhesion promoter (Wacker Primer G790) by gravure printing method with a speed of 3 mm/s for five layers then dried in line at room temperature. The coated PET substrate remained as transparent as the original one, as shown in Fig. S2a. The premixed PDMS was deposited to the gap between the imprinting roll and pressure roll during the R2R imprinting process. We operated the process at room temperature with optimized pressure by testing 1 to 10 kgf/cm² under an imprinting speed of 0.3 mm/s. After coming out from the imprinting unit, imprinted chips on the PET substrate were collected at the rewind roller and were applied to a cutting process for making single devices. The coverlids were simply fabricated by pouring our room-temperature PDMS into a 3D-printed square mold with a dimension of 55 × 55 mm, resulting in unstructured PDMS layers with a uniform thickness of 1 mm. The inlets and outlets for each chamber were manually punched. After that, a 3 mm diameter circular disc magnet which was supplied by First4Magnets, was preloaded into the mixing chamber of each device, and lids were then bonded onto the imprinted layer with a plasma treatment method.

2.5. Replication accuracy and material characterization

Replication accuracy. Dimensional analysis of the 3D printed mold, PDMS mold, and the R2R imprinted LoaD was performed with an industrial microscope Olympus BX53M.

Optical properties measurement. Autofluorescence measurements of three types of PDMS and glass were conducted by using a Nikon Eclipse Ti2 microscope with 4 different excitation wavelengths. The optical transmittance was measured on all three PDMS types: Sylgard 184, Modified Vinyl-terminated PDMS, and Modified Sylgard 184 using a complete setup of a 508 PV™ UV–visible–NIR Spectrophotometer integrated with a bright field microscope Olympus BX53M. For both measurements, we prepared 3 samples for each PDMS by pouring uncured PDMS into the circle Petri dish with a diameter of 55 mm with a uniform thickness of 2 mm and curing at 80 °C. All the samples were cleaned by sonicating in ethanol for 30 min and then dried with a nitrogen gun before analysis. To check the discoloration phenomenon of these materials, we prepared 5 samples for each type of PDMS in the same way as mentioned above, but all were cured at room temperature. After fully curing all samples, we aged them thermally in 5 isothermal ovens respectively at 20 °C, 40 °C, 60 °C, 80 °C, and 100 °C for 1 h before analyzing.

Contact angle measurements. We measured the static contact angle for each type of PDMS by using a Drop Shape Analyzer (DSA100) from Krüss. Measurements were conducted using an automated drop dispenser and deposition device to dispense a 2 μl drop of deionized (DI) water on the material surface. The static contact angle was calculated using computer software.

Mechanical characterization techniques. Tensile strength tests were conducted using the Instron Electropuls E3000 testing system. All samples for three types of PDMS were prepared following ASTM D-412 standards which were cut into dumbbell shapes, referred to as dogbone, using a type-D specimen die. The samples were put into the uniaxial grips and dragged at a speed of 3.84 mm/s until they reached a breakpoint [27]. We performed the tests on five trials for each sample then the engineering stress and strain were reported as averages and corresponding standard deviations.

2.6. Nucleic acid design and reagents

Complete genomes of SARS-CoV-2 Wuhan wildtypes (accession MN908947.3) were retrieved from NCBI (https://www.ncbi.-nlm.nih.gov/). Forward Primer -ACAGGTACGTTAATAGTTAATAGCGT and Reverse Primer -ATATTGCAGCAGTACGCACACA were purchased from Cosmogenetech Inc., (Seoul, Korea). The experiment used SARS-CoV-2 mRNA spiked in human whole blood (Innovative Research,USA) for further analysis. RNA purification reagents were purchased from MagListo 5 M viral DNA/RNA extraction (Bioneer). The buffer solutions (proteinase K, virus binding buffer (VB)), washing buffer (VWM1, RWA2, and WE buffer), elution buffer (ER buffer), magnetic nanobead (MNPs) were preloaded into the chip with the volume: 20 μl proteinase K, 200 μl VB, 400 μl absolute ethanol, 50 μl magnetic nanobead, 500 μl washing buffer, 100 μl ER buffer. 200 μl of blood was added at the beginning of the operation. Polymerase chain reaction (PCR) was performed using the following protocol: reverse transcription (42 °C for 5 min, 95 °C for 10 s), PCR test (40 cycles of 95 °C for 5 s, 60 °C for 30 s), melting curve analysis (95 °C for 0 s, 65 °C for 15 s, 95 °C for 0 s) with 20 μl of final PCR volume (One Step TB Green® PrimeScript™ RT-PCR Kit II (Perfect Real Time) (Takara Bio, Tokyo, Japan).

In our LoaD devices, we utilized ferro-wax valves which were made by mixing paraffin wax (327,204, Sigma-Aldrich, Inc., St. Louis, MO, USA) with Ferrofluid (EFH1, 60 cc, Ferrotec, Santa Clara, CA, USA) in the ratio of 1:1 b y mass and stirring the mixture at 65 °C on a hotplate for 12h.

3. Results

We developed the first sustainable R2R additive manufacturing platform for fabricating PDMS-based centrifugal microfluidic devices at room temperature without using heat and light sources that minimize the emission of greenhouse gas and hazardous by-products (Fig. 1). In this work, we used the continuous in-line printing and imprinting units of our R2R system, illustrated in Fig. 1a. By utilizing additive manufacturing, we introduced sustainable fabrication techniques for a deep-depth flexible shim that addressed the issues of traditional molding technologies (Fig. 1b). As illustrated in Fig. 1c, LoaD devices were successfully manufactured at ambient condition by using the new fast-room temperature-curing PDMS based on Sylgard 184 to eliminate the need for heat and light sources which are often used in conventional R2R imprinting process. The process of on-chip isolation of RNA from whole blood using our R2R imprinted LoaD was illustrated in Fig. 1d. Before the imprinting process, the PET substrate was uniformly treated by R2R gravure printing unit with an adhesion promoter that effectively assisted the demoulding of structures with a large thickness (∼2 mm) (Fig. 1e and f). Finally, we successfully demonstrated the high throughput and sustainable manufacturing platform for producing the LoaD devices (Fig. 1g).

Figure 1. Overview of the room-temperature roll-to-roll imprinted PDMS-based centrifugal microfluidic devices. (a) Roll-to-roll additive manufacturing platform; (b) Enlarged image of R2R imprint unit; (c) Rewinder unit for collecting imprinted LoaD devices; (d) Illustration of on-chip isolation of RNA from whole blood using our R2R imprinted LoaD device; (e) and (f) R2R gravure coating unit; (g) Mass production of PDMS-based centrifugal microfluidic devices on PET substrate.

3.1. Design and 3D printing of the centrifugal microfluidic devices

Before employing the R2R manufacturing system to mass-produce the LoaD devices, the 3D printed LoaD was first tested to prove the working concept. The prototype of LoaD was designed and fabricated using a 3D printer with a diameter of 55 mm, a channel depth and width of 500 μm, and all the chamber depths of 2 mm which enabled the device to process human blood volume of 150 μl (Fig. 2a). The LoaD was designed to have twelve liquid storage chambers connected by microchannels with ten 3D-printed screw valves, which consist of a head and a rod with a square hole of the channel size. The head has dimensions of 3 mm in diameter and 2 mm in height, while the rod has dimensions of 2 mm in diameter and 3 mm in height. The valve is opened by aligning the hole with the microchannel. To close the valve and stop fluid flow, it is rotated so that the hole is perpendicular to the microchannel [29]. In addition, a groove parallel to the microfluidic channel on the top head of the valve allows easy operation by using a screwdriver. Finally, the whole device was fabricated by the 3D printing method (Fig. 2b) to test the device’s performance. The structure of the valve and operating mechanism are shown in Fig. 2c. We conducted experiments on different geometries of the plasma separation chamber, such as a simple square chamber, a tilted square chamber [30], and a square chamber with tilted structures [31] to evaluate the separation efficiency (Fig. S3). Since the plasma separation efficiency was almost the same, we decided to choose the simple square chamber for easy fabrication. The mixing performance of our device was enhanced by integrating an S-shaped microchannel as a serpentine micromixer [32] and an optimized mixing chamber [33]. The whole process of RNA isolation from whole blood on this device consists of four steps: lysis step, binding step, washing step, and elution step. In Fig. 2d, we qualified the effectiveness of mixing performance and logical design of the 3D printed LoaD by using food dye color solutions. The solution flow direction is indicated by the red dashed line area to the light blue dashed line area. First, during the lysis step, when valves 1 and 2 were opened then the chip was centrifuged to transfer the solution (orange dye color) in three chambers (blood, proteinase k, and binding buffer chambers as named in Fig. 2a) to the mixing chamber. Second, in the second binding step, valve 2 was closed while valves 3 and 4 were opened sequentially to transfer the solution in the ethanol chamber and MNPs chamber respectively by centrifuging the device. After opening valve 5, the solution in the mixing chamber was transferred to the waste chamber. Third, in the washing step, valves 6,7, and 8 were opened serially to move the solution in VWM1, RWA2, and WE buffer chambers to the mixing chamber respectively, meanwhile, valve 5 was opened and closed alternatively to release the solution in the mixing chamber to the waste chamber. Finally, the elution step was performed by closing valve 5 and opening valve 9 to transfer the solution from the ER buffer chamber to the mixing chamber then valve 10 was opened to move the solution into the elution chamber to complete the sample preparation process. The eluted solution can be used for the diagnosis by extracting it from the elution chamber. Since our design has been proven to have efficient mixing and transferring performance, it can be further fabricated by the R2R additive manufacturing platform. After several flow tests on the 3D printed LoaD, the leakage at screw valves was observed because the micro-gap between the valve and the holding hole is inevitable. To solve this issue, we employed laser-actuated ferro-wax microvalves [34] for our R2R imprinted LoaD. Briefly, the working mechanism of this photonic valve is relied on the phase transition of the ferro-wax, actuated by using only a single laser diode instead of many microfabricated heaters and magnets. The valve can be opened by melting the ferro-wax plug in the channel with laser irradiation for a few seconds, allowing the melting wax to flow into two assistant chambers. The response time for the actuation of these photonic valves was accelerated due to the effective heating of iron oxide nanoparticles embedded in the paraffin wax matrix by the laser beam. The ferro-wax can be solidified rapidly at room temperature when we stop the laser illumination, enabling us to make a plug in the channel again as a closed valve. This simplifies the control of multiple microvalves. We demonstrated the operation of the ferro-wax microvalves in Fig. S4.

Figure 2. Design and 3D printing of the centrifugal microfluidic devices. (a) 3D model and detailed function of lab-on-a-disc (LoaD) device; (b) 3D printed LoaD device; (c) 3D design of screw valves; (d) Demonstration of device operation by food dyes, “v.1-10” stand for valve 1–10 and fluid flow sequences were indicated by dashed lines and yellow arrow, while red color circle represent closed valve and yellow one represents for opened valve.

3.2. The multi-depth macro-to-micro flexible polymer shim

Due to the dramatic increase in complexity, more microfluidic devices require 3D structures, like multi-depth and layer channels. Moreover, microfluidic chips that combine micron-sized structures with large-volume liquid storage chambers are often required in real-world sample-driven analytical systems for clinical bioanalysis. The conventional way of using photolithography for fabricating these structures is time-consuming and labour-intensive, requiring a precise alignment process and extremely difficult to generate macro-sized features. By utilizing additive manufacturing technology, we developed a rapid and low-cost method for fabricating a multi-depth flexible polymer mold that overcomes the difficulties of traditional molding techniques, especially in integrating macro and micro-sized features. The whole process of fabricating this polymer shim is shown in Fig. 3a. We employed a commercially available resin from the Creative CADwork for the direct 3D printing of master molds that effectively addressed the current issues of time-consuming, nonreplicable, and ineffective surface treatment methods. A commercial flexoplate with low cost, flexible, and uniform thickness was used as the substrate for carrying the patterned molds. This method enabled to rapid manufacture of a large area flexible mold at the lab without using an industrial-scale high-resolution 3D printer. The multi-depth mold, which has a total thickness of 4 mm and consists of 2 mm in chamber depth with a channel depth of 500 μm was well fabricated with the dimension shown in Fig. 3d.

Before starting the R2R imprinting, we applied a conformal coating of parylene C served as surface anti-adhesion (Fig. 1a). This coating material is not only environment-friendly but also extremely effective to prolong the lifetime of the mold without any adhesion to the PDMS resist during the demolding process. To demonstrate that the mold treated with a single coating of parylene C can maintain its anti-adhesive property for a long lifecycle regardless of the number of replica molding cycles, we replicated 10 copies named from C1 to C10 from the master mold named M10. As shown in Fig. 3e, replicas remained high fidelity to the M10 without damaging the master mold. Finally, the flexible properties of this high-thickness polymer mold were demonstrated by wrapping on the imprinting roll with a conformal contact shown in Fig. 3c. This flexible polymer mold was found to be durable because of its capability to withstand high nip pressure (2MPa) for many imprinting cycles.

3.3. The fast-room temperature-curing PDMS

We effectively accelerated the curing time of our PDMS at room temperature by modifying Sylgard 184 formulation with Ashby-Karstedt catalyst and tetrakis (dimethylsoloxy) silane (TDS). As shown in Fig. 4a, the curing time at room temperature of standard Sylgard 184, our PDMS, and vinyl-terminated PDMS are 2 days (2880 min), 10 min, and 12 min, respectively. The addition of 0.1–0.3 wt% Ashby-Karstedt catalyst accelerated the curing time of Sylgard 184 at room temperature and improved its mechanical properties which were demonstrated by Murphy et al. [35]. Additionally, the incorporation of TDS can reduce the curing time significantly [27]. Aiming to reduce the heat waste and energy consumption of conventional R2R hot embossing methods, we optimized the concentration of Pt and TDS into Sylgard 184 formulation to make it possible for the R2R imprinting process at room temperature without using UV and thermal curing systems, which helps to reduce heat waste and energy consumption. In this work, we also modified vinyl-terminated PDMS to cure it rapidly at room temperature as a reference to compare with our PDMS.

Figure 4. Characterization of the fast-room temperature-curing PDMS. (a) Curing time of three different types of PDMS at room temperature; (b) Autofluorescence of all PDMS types and glass at four different excitations: 405 nm, 488 nm, 594 nm, and 647 nm; (c) Transmission spectra of PDMS samples; (d) Discoloration of all three PDMS formulations after 1 h of thermally accelerated aging; (e) Mechanical properties of Sylgard 184 and Room-temp PDMS.

The optical properties of three types of PDMS in this study were measured (Fig. 4b, c, and d). The autofluorescence intensities of three materials and glass substrate were measured by exciting light with four different wavelengths 405 nm, 488 nm, 594 nm, and 647 nm corresponding the excitation wavelength of DAPI, FITC, TRITC and Cy5. In Fig. 4b, the fluorescence intensity of 3 types of PDMS are almost the same in every excitation wavelength, while the glass showed a little decrease in fluorescence intensity. Overall, this confirmed the feasibility of using our room-temperature cured PDMS for biomedical devices at a good quality as the commercial PDMS Sylgard 184 and glass, which were used reference samples. The data about the fluorescence intensity is shown in Table S1. The transmittance of light through the microchannel, also referred to as optical transmittance, is a critical issue for a lab-on-chip (LOC) application since numerous analytical protocols employ visualization equipment operating within the visible wavelength range. For our devices, transmittance plays an important role in the efficiency of the laser-actuated ferro-wax valve. To verify the results, the optical transmittance was measured on all three PDMS samples after the thermal treatment, as shown in Fig. 4c. The transmittance of our room-temperature cured PDMS was lower than that of Sylgard 184 b y an amount of ∼3 %, and both samples exhibited an optical transmittance above 90 % for visible light. On the other hand, the vinyl-terminated PDMS showed poor transparency from 22 % to around 60 % for visible light, which causes difficulty for biomedical applications. This can be explained by the clear-to-yellow discoloration phenomenon of silicones caused by the interaction of platinum-complex [[36], [37], [38], [39]]. We observed the discoloration phenomena on all three types of PDMS by aging them for 1 h in a wide range of temperatures from 20 °C to 100 °C. As illustrated in Fig. 4d, the vinyl-terminated sample produced obvious color changes as temperatures increased. The significant discoloration of this sample can be explained due to the large concentration of platinum-complex by the addition of the Asbhy-Karstedt catalyst and the catalytic reaction of Pt was accelerated as temperature increased. The Sylgard 184 samples remained transparent because no additional platinum was added. Interestingly, even though platinum-complex was added in the formulation of our room-temperature cured PDMS, it remained almost transparent as Sylgard 184 for two reasons. First, the concentration of Pt added was small compared to the vinyl-terminated PDMS to be both cured rapidly at room temperature so that the discoloration level was significantly different between those samples. Second, the addition of TDS prevented discoloration in our PDMS [27].

The mechanical properties of our room-temperature cured PDMS and Sylgard 184 were measured by tensile testing on the dogbone specimens, as shown in Fig. 4e, confirming the quantitatively significant distinction between those two samples. The tensile test results revealed that the room-temp PDMS became harder and less flexible due to the addition of platinum-complex catalyst [35]. The maximum stress and strain of Sylgard 184 are 6.25 ± 0,83 MPa and 101.8 ± 7.02 %, while those of our room-temperature cured PDMS are 5.89 ± 0.98 MPa and 87.51 ± 9.64 %, respectively. We failed to measure the tensile strength of vinyl-terminated PDMS samples due to their extremely low modulus so they were broken during the gripping process before the measurement, as shown in Fig. S6. The water contact angle measurement results shown in Fig. S5 revealed that our room-temp PDMS has the same hydrophobicity property as Sylgard 184 (112.4°) in the meantime the vinyl-terminated PDMS showed a reduction to 105°, which caused adhesion problems during the imprinting process.

3.4. Roll-to-roll replication accuracy

Dimension analysis using the industrial microscope of imprinting tools and PDMS replica (Fig. 5) has shown that the master mold structures were transferred with high accuracy. The results (Fig. 5a) demonstrated that steep sidewalls could be fabricated by our R2R manufacturing platform with only slight bevelling. The cross-section images revealed that the multi-depth of the devices was successfully replicated with an accuracy of 99%. As shown in Fig. 5a, the imprinted chamber depth is 1.99 ± 0.011 mm, and the channel depth is 501.58 ± 1.36 μm while the CAD design were 2.00 mm and 500 μm, respectively. Since parameters such as roll temperature, applied nip pressure, and web transfer speed mainly influenced the quality of imprinted patterns, we optimized those parameters as shown in Table S1. In our developed platform, the operating temperature is low as room temperature, which is not only environmentally friendly but also addresses the common issue in the embossing process resulting from thermal expansion variations between the mold, substrate, and resists. The influence of nip pressure can be seen obviously in Fig. 5a. When the nip pressure increased from 1 kgf/cm2 to 10 kgf/cm2, the deformation of imprinted structures was generated and reduced the thickness of imprinted substrate resulting in failure products. Because the large volume of the devices required a large amount of dispensed PDMS, the imprinting speed was set at 0.3 mm/s to ensure the PDMS filled into the mold patterns with high fidelity and without generating bubbles. Therefore, the curing time of PDMS and printing parameters should be adjusted depending on the structures of the designs so that small devices could be manufactured more efficiently. The best condition for imprinting the LoaD is 1 kgf/cm2 and 0.3 mm/s as nip pressure and imprinting speed, respectively (Table S2). The dimension of three critical positions in the LoaD design (valve, inlet hole of each chamber, and S-shaped channel) of 3D printed mold, PDMS mold, and R2R imprinted LoaD were investigated as shown in Fig. 5b. The lowest variation of structural dimension between final products and computer aid design was in the range of ±2.7 μm, confirming the high replication accuracy of our R2R manufacturing method.

Figure 5. Roll-to-roll (R2R) replication accuracy. (a) Cross-sectional images with different magnifications from R2R imprinted samples under different operating nip pressure. Dashed areas on the left side images present the regions shown on the right; (b) Replication accuracy measuring at three positions: waste channel, S-shaped channel, and inlet hole on CAD design, 3D printed mold, PDMS mold, and R2R replicated LoaD with five samples per each.

3.5. RNA extraction from whole blood on the R2R additive manufactured LoaD

To evaluate RNA extraction on our device, we implemented a design featuring ten preloaded liquid storage chambers that are separated by photonic valves. The complete protocol for extracting RNA from whole blood can be executed utilizing our LoaD (Fig. S7), comprising plasma separation, sample lysis, magnetic binding, washing, and elution which were designed based on prior literature [31]. The magnetic nanobeads were previously loaded into the mixing compartment and coupled with a small magnet (d = 3 mm) for binding to the intended RNA. All procedures were executed using a centrifuge machine manufactured by Optolane (Fig. S9).

As demonstrated in Fig. 6a, the whole chip process was conducted by using food dye for visualization of leakage testing and real blood samples for RNA extraction which shown in left and right images of each step, correspondingly. A variety of food dyes were pre-loaded onto the LoaD to illustrate each step and validate the functional capabilities of the LoaD in relation to leakage, separation, and mixing criteria (step 1) as following the same process in Fig. 2e. Under the centrifugal force of 2000 rpm per 1 min, no leakage was observed between compartments, and the buffer solutions were efficiently conveyed to the mixing compartment without any backflow to the primary channel. We proceeded with the whole operations (step 1–10) for evaluating all the compartments and obtained similar outcomes. This indicated that the whole functions of our R2R-manufactured devices were successfully tested. As a result, the LoaD was subsequently utilized for testing whole blood for the extraction and validation of RNA. The reagents and procedures are summarized in Table S3. 150 μl of whole blood (step 1) were loaded onto the LoaD and centrifuged at 2000 rpm in 3 min for successfully separating red blood cells (RBCs) and plasma (step 2). The simulation of this separation of plasma from whole blood was reported in our previous work [33]. The blood chamber was optimized on different designs for the easy fabrication and enhancement of the sedimentation rate of RBCs (Fig. S3). Furthermore, the connection channel to the mixing chamber is positioned at a higher elevation in the blood storage chamber to prevent the adhesion of RBCs on the connection channel, allowing the plasma to freely flow into the mixing chamber (step 3). After the plasma separation step, photonic valves 1 and 2 were activated by illuminating them with a laser (808 nm, 500 mW). The iron oxide nanoparticles embedded in the paraffin wax matrix (called ferro-wax) were heated by a laser beam, resulting in the melting of the ferro-wax and moving it from chamber 1 to chambers 2 and 3, thereby opening/closing the connection channel (as shown in Fig. 6b,c and Fig. S4). The chip was then centrifuged at 1000 rpm for 30 s which enabled the transfer of plasma, proteinase K, and VB buffer into the mixing chamber for lysing cell compartments (step 3). Subsequently, photonic valve 3 was opened to allow for the centrifugation-assisted transfer of ethanol to the mixing chamber which precipitated the DNA/RNA released from plasma (step 4) then rotating the devices at a mixing mode for 30 s and incubated at 60 °C for 10 min (step 5). After that, valve 4 was opened to transfer the magnetic nanobeads to the mixing chamber by spinning the chip at 1000 rpm for 15 s (step 6). The DNA/RNA is then attached to the MNPs and magnet in the mixing chamber by mixing and incubating at room temperature for 60 s. The magnet was placed in the mixing chamber from the beginning to reduce the number of steps as shown in Fig. 6d. For further improvement, we can preload magnetic nanobeads in the mixing chamber to avoid issues in improper movement of MNPs to the mixing chamber caused by resistance from PDMS walls. Following the binding process, photonic valve 5 was opened to transfer the aqueous part to the waste chamber. Next, valve 5 should be closed again to prevent waste solution from flowing back into the mixing chamber. Subsequently, photonic valve 6 was opened to release VWM1 and washout any impurities remaining on the mixing chamber and on the DNA/RNA (step 7). After washing, the VWM1 buffer was removed to the waste chamber by reopening valve 5 through centrifugation. This washing process was repeated twice with RWA2, and WE buffer to thoroughly washout all impurities (step 8 and 9). Finally, photonic valve 9 was opened to release ER buffer into the mixing chamber which detached the purified DNA/RNA from the magnet. The eluted DNA/RNA was then transferred to the eluent chamber through valve 10 (step 10). The eluent was then extracted for further analysis by RT-qPCR. The entire purification process of 150 μl blood on the chip could be completed within 30 min.

Figure 6. Room-temp PDMS-based LoaD operation. (a) Food dye visualization for testing leakage issues and photonic valves operation and snapshot images of the device for the whole process of RNA extraction from the whole blood, which were described by left and right images of each step correspondingly. The solution moves from the yellow dashed line area to the red dashed line area; (b) and (c) Photonic valve in close (left) and open (right) state. In the close state, ferro-wax was stored in chamber 1 and the connection channel, while assistant chambers 2 and 3 contained no wax (white dye color). In the open state, ferro-wax in the connection channel was melted by a laser and then moved to chambers 2 and 3 to open the channel enabling the transfer of liquid; (d) Magnification of magnet (d = 3 mm) in mixing chamber; (e) Gel electrophoresis of PCR result of on-chip extraction sample. Lane 1 represents the 50 bp ladder, lane 2 displays GAPDH gene in human, and lane 3 shows the amplification plot of SARS-CoV-2 (103 copies/μl) spiked in whole blood, and lane 4 is the SARS-CoV-2 (103 copies/μl) in whole blood and proceeded with conventional extraction method using the same kit as a positive control.

As a demonstration of the feasibility of our method, we performed RT-PCR to further confirm the extraction process by utilizing the LoaD. To validate the on-chip extraction and purification process, we included GAPDH gene primers as an internal control (IC) for house-keeping genes in human. The amplicons were visualized through gel electrophoresis after running a benchtop PCR. In Fig. 6e, lane 1 represents the 50 bp ladder, lane 2 clearly displays a strong band for the GAPDH gene in human, indicating that the plasma was successfully lysed and purified by our device. Next, lane 3 shows the amplification plot of SARS-CoV-2 (103 copies/μl) spiked in whole blood, and lane 4 is the SARS-CoV-2 (103 copies/μl) in whole blood and proceeded with conventional extraction method using same kit as a positive control for comparison with on-chip spiked samples. The intensity of COVID-19 in lane 3 exhibited adequate amplification efficiency compared to the positive sample in lane 4. However, it still demonstrated successful amplification on gel electrophoresis, indicating the extraction and purification of RNA from human whole blood. The weak amplitude may be caused by losing the spiked RNA during the centrifugation of whole blood in the plasma separation chamber. In summary, as a proof-of-concept test, we have successfully employed the R2R additive manufacturing platform to develop a whole blood extraction by utilizing the room-temperature cured PDMS chip and amplified both spiked SARS-CoV-2 and housekeeping gene (GAPDH) using our LoaD.

4. Discussion

The scope of this study was limited in terms of production rate compared to other techniques such as injection molding due to the lab-scale facilities. However, it is certainly possible to scale up this manufacturing process by increasing roll size with optimized mold design as well as reducing the PDMS curing time by adjusting the chemical composition. Firstly, we demonstrated that the production rate could be significantly enhanced approximately ∼7 times compared to the current one by simply optimizing the mold space with the current imprinting roll size (Fig. S10). Therefore, a larger imprinting roll can be employed to enable high throughput industrial-scale manufacturing process. Secondly, the curing time of PDMS is limited by 10 min due to the lack of an efficient dispensing unit that can continuously perform the mixing, degassing, and dispensing the proper amount of uncured PDMS in-line with the imprinting process. By developing this unit, a further study could assess the faster curing time at room temperature. As a result, the cooling condition (−15 °C) can be eliminated by simply adjusting the concentration of catalysts. Furthermore, a PDMS-tape bonding method can be employed for a rapid, simple, inexpensive, and energy efficient laminating method [40] which enable a greater degree of high throughput and sustainability for our proposed manufacturing process.

5. Conclusions

In conclusion, we demonstrated the printing-imprinting integrated R2R continuous in-line additive manufacturing platform, called as green R2R platform, for producing the PDMS-based LoaD with lower energy consumption and less by-products. To realize the green R2R platform, we addressed two main technological hurdles: multi-depth mold fabrication and the fast-room temperature-curing PDMS precursor, enabling a rapid imprinting process. Thus, we developed a rapid, cost-effective fabrication method of a multi-depth flexible polymer shim using 3D-printing technology, which overcomes the challenges in traditional molding techniques especially for integrating macro- and micro-sized features. In addition, we unveiled a novel PDMS formulation by utilizing Ashby–Karstedt catalyst that not only could cure quickly at room temperature, but also could gain better mechanical performance than Sylgard 184 standard. Finally, the resulting PDMS-based LoaD could be expandable for on-site RNA/DNA isolation from the large to a small sample size of whole blood (<150 μl). Our novel fabrication method operated at room temperature which eliminated energy consumption for UV light and heat source will pave the way for addressing the challenges in sustainable high-throughput manufacturing of PDMS-based microfluidic devices which have been highly demanded in the era of Coronavirus (COVID-19) pandemics.

Supplementary Materials

Download

References

X. Wei, W. Zhou, S.T. Sanjay, J. Zhang, Q. Jin, F. Xu, D.C. Dominguez, X. Li
Multiplexed instrument-free bar-chart SpinChip integrated with nanoparticle-mediated magnetic aptasensors for visual quantitative detection of multiple pathogens
Anal. Chem., 90 (2018), 10.1021/acs.analchem.8b02055
M. Dou, S.T. Sanjay, D.C. Dominguez, S. Zhan, X. Li
A paper/polymer hybrid CD-like microfluidic SpinChip integrated with DNA-functionalized graphene oxide nanosensors for multiplex qLAMP detection
Chem. Commun., 53 (2017), 10.1039/c7cc03246c
M. Dou, S.T. Sanjay, M. Benhabib, F. Xu, X.J. Li
Low-cost bioanalysis on paper-based and its hybrid microfluidic platforms
Talanta, 145 (2015), 10.1016/j.talanta.2015.04.068
H. Tavakoli, W. Zhou, L. Ma, S. Perez, A. Ibarra, F. Xu, S. Zhan, X.J. Li
Recent advances in microfluidic platforms for single-cell analysis in cancer biology, diagnosis and therapy
TrAC, Trends Anal. Chem., 117 (2019), 10.1016/j.trac.2019.05.010
F. Hu, J. Li, N. Peng, Z. Li, Z. Zhang, S. Zhao, M. Duan, H. Tian, L. Li, P. Zhang
Rapid isolation of cfDNA from large-volume whole blood on a centrifugal microfluidic chip based on immiscible phase filtration
Analyst, 144 (2019), pp. 4162-4174, 10.1039/C9AN00493A
Z. Noroozi, H. Kido, R. Peytavi, R. Nakajima-Sasaki, A. Jasinskas, M. Micic, P.L. Felgner, M.J. Madou
A multiplexed immunoassay system based upon reciprocating centrifugal microfluidics
Rev. Sci. Instrum., 82 (2011), 10.1063/1.3597578
R. Gorkin, J. Park, J. Siegrist, M. Amasia, B.S. Lee, J.M. Park, J. Kim, H. Kim, M. Madou, Y.K. Cho
Centrifugal microfluidics for biomedical applications
Lab Chip, 10 (2010), 10.1039/b924109d
I.J. Michael, T.H. Kim, V. Sunkara, Y.K. Cho
Challenges and opportunities of centrifugal microfluidics for extreme point-of-care testing
Micromachines, 7 (2016), 10.3390/mi7020032
T.H. Kim, V. Sunkara, J. Park, C.J. Kim, H.K. Woo, Y.K. Cho
A lab-on-a-disc with reversible and thermally stable diaphragm valves
Lab Chip, 16 (2016), 10.1039/c6lc00629a
M.M. Aeinehvand, F. Ibrahim, S.W. Harun, A. Kazemzadeh, H.A. Rothan, R. Yusof, M. Madou
Reversible thermo-pneumatic valves on centrifugal microfluidic platforms
Lab Chip, 15 (2015), 10.1039/c5lc00634a
S.M.A. Mortazavi, P. Tirandazi, M. Normandie, M.S. Saidi
Efficient batch-mode mixing and flow patterns in a microfluidic centrifugal platform: a numerical and experimental study
Microsyst. Technol., 23 (2017), 10.1007/s00542-016-3109-7
B.L. Thompson, Y. Ouyang, G.R.M. Duarte, E. Carrilho, S.T. Krauss, J.P. Landers
Inexpensive, rapid prototyping of microfluidic devices using overhead transparencies and a laser print, cut and laminate fabrication method
Nat. Protoc., 10 (2015), 10.1038/nprot.2015.051
S. Hosseini, M.M. Aeinehvand, S.M. Uddin, A. Benzina, H.A. Rothan, R. Yusof, L.H. Koole, M.J. Madou, I. Djordjevic, F. Ibrahim
Microsphere integrated microfluidic disk: synergy of two techniques for rapid and ultrasensitive dengue detection
Sci. Rep., 5 (2015), 10.1038/srep16485
E. Duffy, R. Padovani, X. He, R. Gorkin, E. Vereshchagina, J. Ducrée, E. Nesterenko, P.N. Nesterenko, D. Brabazon, B. Paull, M. Vázquez
New strategies for stationary phase integration within centrifugal microfluidic platforms for applications in sample preparation and pre-concentration
Anal. Methods, 9 (2017), 10.1039/c7ay00127d
Y. Xia, G.M. Whitesides
Soft lithography
Annu. Rev. Mater. Sci., 28 (1998), pp. 153-184, 10.1146/annurev.matsci.28.1.153
G.M. Whitesides
The origins and the future of microfluidics
Nature, 442 (2006), pp. 368-373, 10.1038/nature05058
L. Gervais, N. De Rooij, E. Delamarche
Microfluidic chips for point-of-care immunodiagnostics
Adv. Mater., 23 (2011), 10.1002/adma.201100464
C.D. Chin, V. Linder, S.K. Sia
Commercialization of microfluidic point-of-care diagnostic devices
Lab Chip, 12 (2012), 10.1039/c2lc21204h
J. Hiltunen, C. Liedert, M. Hiltunen, O.H. Huttunen, J. Hiitola-Keinänen, S. Aikio, M. Harjanne, M. Kurkinen, L. Hakalahti, L.P. Lee
Roll-to-roll fabrication of integrated PDMS-paper microfluidics for nucleic acid amplification
Lab Chip, 18 (2018), 10.1039/c8lc00269j
E.K. Sackmann, A.L. Fulton, D.J. Beebe
The present and future role of microfluidics in biomedical research
Nature, 507 (2014), 10.1038/nature13118
N. Kooy, K. Mohamed, L.T. Pin, O.S. Guan
A review of roll-to-roll nanoimprint lithography | nanoscale research letters | full text
Nanoscale Res. Lett., 9 (2014)
S.H. Ahn, L.J. Guo
High-speed roll-to-roll nanoimprint lithography on flexible plastic substrates
Adv. Mater. (2008), p. 20, 10.1002/adma.200702650
Y. Hirai, S. Yoshida, N. Takagi
Defect analysis in thermal nanoimprint lithography
J. Vac. Sci. Technol. B: Microelectronics and Nanometer Structures, 21 (2003), 10.1116/1.1629289
H.N. Chan, Y. Chen, Y. Shu, Y. Chen, Q. Tian, H. Wu
Direct, one-step molding of 3D-printed structures for convenient fabrication of truly 3D PDMS microfluidic chips
Microfluid. Nanofluidics, 19 (2015), 10.1007/s10404-014-1542-4
G. Comina, A. Suska, D. Filippini
PDMS lab-on-a-chip fabrication using 3D printed templates
Lab Chip, 14 (2014), 10.1039/c3lc50956g
Y. Chen, W. Pei, R. Tang, S. Chen, H. Chen
Conformal coating of parylene for surface anti-adhesion in polydimethylsiloxane (PDMS) double casting technique
Sens Actuators A Phys, 189 (2013), 10.1016/j.sna.2012.09.024
Z. Brounstein, J. Zhao, D. Geller, N. Gupta, A. Labouriau
Long-term thermal aging of modified Sylgard 184 formulations
Polymers, 13 (2021), 10.3390/polym13183125
C. Pina-Hernandez, J.S. Kim, L.J. Guo, P.F. Fu
High-throughput and etch-selective nanoimprinting and stamping based on fast-thermal-curing poly(dimethylsiloxane)s
Adv. Mater., 19 (2007), 10.1002/adma.200601905
M.T. Guler, P. Beyazkilic, C. Elbuken
A versatile plug microvalve for microfluidic applications
Sens Actuators A Phys, 265 (2017), pp. 224-230, 10.1016/j.sna.2017.09.001
T.H. Kim, H. Hwang, R. Gorkin, M. Madou, Y.K. Cho
Geometry effects on blood separation rate on a rotating disc
Sensor. Actuator. B Chem., 178 (2013), 10.1016/j.snb.2013.01.011
C.J. Kim, J. Park, V. Sunkara, T.H. Kim, Y. Lee, K. Lee, M.H. Kim, Y.K. Cho
Fully automated, on-site isolation of cfDNA from whole blood for cancer therapy monitoring
Lab Chip, 18 (2018), 10.1039/c8lc00165k
J. Clark, M. Kaufman, P.S. Fodor
Mixing enhancement in serpentine micromixers with a non-rectangular cross-section
Micromachines, 9 (2018), 10.3390/mi9030107
J. Lee, S. Lee, M. Lee, R. Prakash, H. Kim, G. Cho, J. Lee
Enhancing mixing performance in a rotating disk mixing chamber: a quantitative investigation of the effect of euler and coriolis forces
Micromachines, 13 (2022), 10.3390/mi13081218
J.M. Park, Y.K. Cho, B.S. Lee, J.G. Lee, C. Ko
Multifunctional microvalves control by optical illumination on nanoheaters and its application in centrifugal microfluidic devices
Lab Chip, 7 (2007), 10.1039/b616112j
E.C. Murphy, J.H. Dumont, C.H. Park, G. Kestell, K.S. Lee, A. Labouriau
Tailoring properties and processing of Sylgard 184: curing time, adhesion, and water affinity
J. Appl. Polym. Sci., 137 (2020), 10.1002/app.48530
A.K. Antosik, P. Bednarczyk, Z. Czech
Aging of silicone pressure-sensitive adhesives
Polym. Bull., 75 (2018), 10.1007/s00289-017-2083-2
K.R. McIntosh, J.N. Cotsell, J. Cumpston, A.W. Norris, N.E. Powell, B.M. Ketola
The Effect of Accelerated Aging Tests on the Optical Properties of Silicone and EVA Encapsulants
(2009), 10.4229/24thEUPVSEC2009-4AV.3.54
E.D. Cifter, M. Ozdemir-Karatas, A. Cinarli, E. Sancakli, A. Balik, G. Evlioglu
In vitro study of effects of aging and processing conditions on colour change in maxillofacial silicone elastomers
BMC Oral Health, 19 (2019), 10.1186/s12903-019-0798-1
J.J. Gary, C.T. Smith
Pigments and their application in maxillofacial elastomers: a literature review
J. Prosthet. Dent, 80 (1998), 10.1016/S0022-3913(98)70111-8
C.S. Thompson, A.R. Abate
Adhesive-based bonding technique for PDMS microfluidic devices
Lab Chip, 13 (2013), pp. 632-635, 10.1039/C2LC40978J

Skin-interfaced microfluidic systems with spatially engineered 3D fluidics for sweat capture and analysis

March 12, 2026May 3, 2023 by CadWorks3DAdmin

A 3D printed epifluidic device called a "sweatainer" used for sweat capture and analysis

Skin-interfaced microfluidic systems with spatially engineered 3D fluidics for sweat capture and analysis

by Chung-Han Wu, Howin Jian Hing Ma, Paul Baessler, Roxanne Kate Balanay and Tyler Ray

Abstract: Skin-interfaced wearable systems with integrated microfluidic structures and sensing capabilities offer powerful platforms for monitoring the signals arising from natural physiological processes. This paper introduces a set of strategies, processing approaches, and microfluidic designs that harness recent advances in additive manufacturing [three-dimensional (3D) printing] to establish a unique class of epidermal microfluidic (“epifluidic”) devices. A 3D printed epifluidic platform, called a “sweatainer,” demonstrates the potential of a true 3D design space for microfluidics through the fabrication of fluidic components with previously inaccessible complex architectures. These concepts support integration of colorimetric assays to facilitate in situ biomarker analysis operating in a mode analogous to traditional epifluidic systems. The sweatainer system enables a new mode of sweat collection, termed multidraw, which facilitates the collection of multiple, independent sweat samples for either on-body or external analysis. Field studies of the sweatainer system demonstrate the practical potential of these concepts.

We kindly thank the researchers at University of Hawai'i at Mānoa for this collaboration, and for sharing the results obtained with their CADworks3D system.

Introduction

Eccrine sweat is an attractive class of biofluid suitable for the noninvasive monitoring of body chemistry. Sweat contains a rich composition of biomarkers relevant to physiological health status including electrolytes (1), metabolites (2–4), hormones (5, 6), proteins (7), and exogenous agents (8). Studies demonstrate the intermittent or continuous assessment of these, and other sweat biomarkers offer time dynamic insight into the metabolic processes of the body relevant to applications ranging from athletic performance (9–11) to medical diagnostics (2, 12–14).

Recent advances in soft microfluidics, sensing technologies, and electronics establish the foundations for a unique class of skin-like epidermal microfluidic (“epifluidic”) systems. Adapting concepts from traditional lab-on-chip technologies, these wearable microfluidic platforms comprise sophisticated networks of channels, valves, and reservoirs embedded in elastomeric substrates (15–20). The thin, flexible device construct facilitates a conformal, fluid-tight skin interface by virtue of skin-compatible adhesives to collect sweat directly from sweat glands. The integration of colorimetric, fluorometric, and electrochemical measurement techniques enable such platforms to measure sweat constituents in situ across a wide array of applications and environments (21).

Traditional approaches for sweat collection use absorbent pads (22) or microbore tubes (23) pressed against the epidermis by virtue of bands or straps to capture sweat as it emerges from the skin. Requiring trained personnel, special handling, and costly laboratory equipment, such methods are incompatible with real-time sweat analysis and prone to sample contamination or loss (24). Epifluidic devices eliminate external sample contamination by virtue of the intrinsic encapsulation of the microfluidic network and conformal skin interface. Such systems are vulnerable to surface contamination from exogenous agents present on the epidermis, such as cosmetics or natural oils, without careful preparation of the skin surface before device attachment. Furthermore, the dependence on an adhesive interface for skin attachment limits these devices to single-use applications. Upon removal, the risk of contamination, potential sample loss, and active sweat response of previously covered glands pose substantial challenges to reapplication and continued sweat collection.—Traditional approaches for sweat collection use absorbent pads (22) or microbore tubes (23) pressed against the epidermis by virtue of bands or straps to capture sweat as it emerges from the skin. Requiring trained personnel, special handling, and costly laboratory equipment, such methods are incompatible with real-time sweat analysis and prone to sample contamination or loss (24). Epifluidic devices eliminate external sample contamination by virtue of the intrinsic encapsulation of the microfluidic network and conformal skin interface. Such systems are vulnerable to surface contamination from exogenous agents present on the epidermis, such as cosmetics or natural oils, without careful preparation of the skin surface before device attachment. Furthermore, the dependence on an adhesive interface for skin attachment limits these devices to single-use applications. Upon removal, the risk of contamination, potential sample loss, and active sweat response of previously covered glands pose substantial challenges to reapplication and continued sweat collection.

The typical epifluidic fabrication pathway uses soft lithography techniques (25) to produce devices with microfluidic components and complex geometries. A common, well-established process for fabricating lab-on-chip microfluidic devices (26), soft lithography, requires high-precision molds to form discrete, patterned layers of an elastomeric material [e.g., poly(dimethylsiloxane) (PDMS)] that when bonded together yield a sealed device. Traditionally, producing molds with sufficient feature resolution (>20 μm) requires expensive, time-consuming processing methods [micromachining (27) and micromilling (28)] and access to specialized environments (cleanroom). Such requirements result in elongated device design cycles, inequitable access to equipment necessary for innovation, and additional challenges for commercial deployment due to incompatibilities with large-scale manufacturing.

Additive manufacturing (AM), or three-dimensional (3D) printing, represents an attractive alternative to conventional planar (2D) fabrication methods. AM offers powerful capabilities for producing structurally complex objects with true 3D architectures through a rapidly expanding library of printing methods. In general, these methods create solid objects in a sequential, layer-by-layer manner directly from a digital computer-aided design (CAD) file. In the context of microfluidics, the use of 3D printing is well established (29) for the rapid, cost-effective fabrication of high-resolution templates for soft lithography. In particular, vat photopolymerization techniques [e.g., resin-based printing, stereolithography, digital light processing (DLP), and continuous liquid interface polymerization] (30) enable rapid production of microscale features (>100 μm) over large areas (>600 mm²) with high precision (31). Innovations in printer hardware, software processing, and materials chemistry further extend these 3D printing capabilities to enable the direct production of enclosed microfluidic channels for lab-on-chip applications. Although manufacturers advertise printers with high resolution (xy resolution: >50 μm and z-resolution: >5 μm), in practice, the obtainable channel dimensions and device complexity are typically limited to millifluidic features (>250 μm) (29). Printer specifications represent only one key constraint to printing devices with micron-scale internal fluidic features (<100 μm). Successful fabrication requires optimization of other critical factors including printing technology (e.g., vat photopolymerization versus extrusion), feature design and spatial location, and printer-dependent parameters. AM process optimization, particularly for vat photopolymerization, demands careful attention to the chemistry of printed materials (30, 32). Resin formulations must simultaneously satisfy application specific requirements, such as biocompatibility or optical clarity, while preserving printability. Recent reports (32, 33) leverage specialized DLP-based printers and customized resins to fabricate devices containing microfluidic components with <50-μm dimensions.

Apparatus Used

Clear Microfluidic Resin

Learn More

Curezone

Learn More

PR110
3D Printer

Legacy

In general, wearable system designs must address the inherent mismatch between the mechanical properties of skin and rigid, planar device components. The most advanced platforms fabricated by conventional (non-AM) methods exploit sophisticated strategies, combining complex device geometries and soft (low modulus) materials to establish a seamless, nonirritating epidermal interface. Recent advances in soft materials chemistry support 3D printing approaches to fabricating wearable devices for applications spanning biophysical (34), biochemical (35, 36), and environmental (37) monitoring. However, such capabilities remain limited for the 3D fabrication of epifluidic devices as a result of the high Young’s moduli of the primary material chemistries (i.e., methacrylate-based resins) (38) suitable for printing high-resolution microfluidics. Current efforts to fabricate skin-interfaced 3D printed microfluidics use alternative printing methods [e.g., fused deposition modeling (34) and direct ink writing (39)] that support fabrication with low modulus materials at the expense of printer resolution (>200 μm). In the context of epifluidics, the ideal fabrication scheme would use resin-based printing to fabricate devices with feature sizes comparable to conventional methods with biologically compliant form factors. Such an approach would transform the fluidic design space with truly 3D device architectures while enabling a rapid, iterative design process, facilitating individual-specific device customization, and reducing the cost for low-volume production.

This paper introduces a set of strategies, processing approaches, and microfluidic designs that support such fabrication capabilities using a commercial DLP 3D printer in a simple manner of operation. A modular 3D printed epifluidic platform, termed a “sweatainer,” demonstrates several unique aspects of an AM approach to fabricating epifluidic systems. This platform, to our knowledge, represents the first 3D printed epifluidic platform with true microfluidic dimensions. Specifically, the results highlight the potential of a true 3D design space for microfluidics through the fabrication of fluidic components (channels and valves) with previously inaccessible complex architectures. Printer optimization strategies and systematic experiments enable realization of micron-scale feature sizes (<100 μm) and enhancement of optical transparency of 3D printed channels. In combination, these concepts support integration of colorimetric assays to facilitate in situ biomarker analysis operating in a mode analogous to traditional epifluidic systems. Drawing inspiration from the vacutainer blood collection tube, the sweatainer system introduces a novel mode of sweat collection, termed “multidraw.” This method overcomes the inherent limitations of single-use devices by enabling the collection of multiple, independent pristine sweat samples during a single collection period. Field studies of the sweatainer system demonstrate the practical potential of these concepts.

Results

Sweatainer system design

Figure 1A shows a schematic illustration of the two primary modules of the sweatainer system: (i) the sweatainer device and (ii) an epidermal port interface. The sweatainer consists of a 3D printed microfluidic network of enclosed channels and unsealed reservoirs, a reservoir capping layer of PDMS (thickness: 200 μm), and a gasket formed from ultrathin biomedical adhesive (3M 1524; thickness: 60 μm). The bonded 3D printed photocurable resin structure and PDMS capping layer, as presented in Materials and Methods, define a closed microfluidic structure. Introduction of either dye or colorimetric assay before bonding enables sweat visualization or chloride concentration analysis, respectively. The cross-sectional width and thickness of the filleted serpentine channels presented here are 1200 and 1000 μm, respectively. The width and height of the rectangular-shaped internal microfluidic channels are 600 and 400 μm, respectively. The filamentary design of the rigid 3D printed structure (Young’s modulus: ~975 MPa) follows from the well-established principles of stretchable electronics (40) to impart sufficient stretchability to form a mechanically robust conformal interface. The gasket establishes a temporary, fluid-tight seal with the epidermal port interface permitting facile sweatainer application and removal via reversible adhesion to the PDMS surface.

Figure 1. Schematic illustrations and optical images of the 3D printed epidermal microfluidic devices for the collection and analysis of sweat. (A) An exploded render highlights key components of the sweatainer system and epidermal interface (port). PDMS, poly(dimethylsiloxane). (B) A photograph of the sweatainer mounted on the ventral forearm of an individual before the onset of sweat collection. (C) The construct of the sweatainer eliminates uncontrolled fluid transport under mechanical loading (e.g., finger pressure and device removal). (D) Illustration of the sweatainer highlighting key device aspects including the inlet, capillary burst valves (CBVs; blue and red dashed area), collection reservoir, and ventilation holes to eliminate backpressure. (E) Renders of three-dimensional (3D) CBV designs enabled by 3D printing with diverging angles of 90° (top) and 135° (bottom). (F) 3D printing enables fabrication of device geometries in a true 3D space as shown by the computer-aided design (CAD) render (top) and photograph of actual device (bottom). Location of sweat appears in blue. (G) Photographic sequence highlighting the complete filling of a sweat collection reservoir.

The epidermal port interface comprises a thin film of pigmented PDMS (white, thickness: 400 μm) and a medical-grade adhesive layer (3M 1524) with laser-patterned openings. The adhesive layer facilitates a biocompatible, fluid-tight interface with the epidermis in which the patterned opening defines the sweat collection region (~180 mm²). An aligned access point on the backside of the sweatainer allows sweat to enter the system directly from the skin with flow driven by the natural pressures created by the sweat glands. The sweatainer design can support collection of 50.8 μl of sweat (10.8 μl per reservoir and 18.4 μl of channel network). A fully assembled representative system appears in Fig. 1B, where it is shown worn on the ventral forearm. Figure 1C demonstrates the insensitivity of the sweatainer to mechanical deformation through the absence of uncontrolled fluid flow during physical handling (finger pressure). The schematic illustration in Fig. 1D shows the microfluidic network within the 3D printed sweatainer. Sweat enters the device by the central inlet and flows through a microfluidic channel leading to a series of capillary burst valves (CBVs) and corresponding reservoirs. The CBV at the ingress of each reservoir permits fluid flow only after exceeding a set pressure, thereby enabling time-sequential sweat collection (20). Integrated ventilation holes (width: 100 μm and height: 200 μm) on the reservoir eliminate the backpressure that would evolve from trapped air and impede flow. The high-barrier properties of the photocurable resin support a low sweat evaporation rate with minimal mass loss over a 24-hour period (fig. S1 and table S1).

A key feature of this system is the use of AM to enable fully 3D, monolithic microfluidic designs comprising sophisticated nonplanar internal channel structures, spatially graded geometries, and 3D CBVs. Representative examples of 3D CBVs and the spatially graded, nonplanar geometries enabled by this fabrication method appear in Fig. 1 (E and F, respectively). By comparison, soft lithography fabrication methods restrict the design space of traditional lab-on-chip and epifluidic devices to planar (2D) channel configurations. Although lamination of multiple channel layers can yield elaborate 3D microfluidic networks, each component layer is inherently a planar geometry. As detailed in the sections that follow, the 3D fabrication expands the design space for CBVs with finer control over resultant burst pressure in comparison to planar CBVs. In a similar manner, spatially graded geometries improve sweat collection efficiency by permitting a continuous transition between the microfluidic channel and reservoir (Fig. 1F). This engineered interface, in combination with ventilation holes, ensures a uniform fluid front during reservoir filling (Fig. 1G, blue dye for visualization), thereby eliminating trapped air bubbles that result from a rapid expansion.

Design and DLP printing considerations for optimized fabrication of 3D printed epifluidic devices

Successful fabrication of a fully enclosed microfluidic channel with feature sizes at the xy plane resolution limit of current DLP printers (~30 to 50 μm) depends on several related factors including: design aspects (e.g., channel vertical position), print process parameters [e.g., layer height, layer cure time (LCT), and print speed], and printer hardware (e.g., projector light power and wavelength). Optimization of user-adjustable factors results in a robust print process suitable for producing microfluidic devices with sufficient optical clarity, dimensional fidelity, and mechanical performance for use in epifluidic applications.

As expected, epifluidic device performance is dependent on the dimensional accuracy of a fabrication process. If not quantified, then unintended deviation from designed dimensions can adversely affect component performance (i.e., CBV burst pressure) or measurement accuracy (i.e., sweat volume, sweat rate). Fabrication of test structures (Fig. 2A) comprising a sequence of square channels (width and height range: 100 to 900 μm, 100-μm increments; length: 5 mm) embedded in a square base (width and height: 1 mm) facilitate determination of the minimum printable channel dimensions and sidewall thickness (minimum of 50 μm). The asymmetric vertical position of the channels establishes a uniform capping layer (100 μm) across all dimensions tested. Because the DLP printer fabricates the structure in an inverse manner (Fig. 2A, base prints first), the channel position minimizes photopolymerizing resin trapped in the channel during the printing process.

Figure 2. Optimized design strategy for fabricating 3D printed epifluidic devices with prescribed channel geometries. (A) Photograph of 3D printed test channels [100 to 900 μm, square; 2-s layer cure time (LCT)]. (B) Plot of variation of printed channel height from designed dimensions as a function of LCT. (C) Plot of variation of printed channel width from designed dimensions as a function of LCT. (D) Plot highlighting the printable region of the digital light processing (DLP) printer used in this work for various channel dimensions relevant to epifluidic devices.

Experimental studies reveal the similarly strong influence of LCT on print success and device quality. The LCT defines the energy dose used to cross-link the photopolymer given in time (seconds). The projector wavelength is hardware defined (385 nm for this work), and varying the power is not typically user accessible. Systematic studies of four LCT settings—selected starting from the minimum (0.54 s) to maximum values (2.0 s; 0.6-s interval) beyond which channels could not be fabricated successfully—establish a relationship among print performance (i.e., channel printed successfully), dimensional accuracy, and optical clarity. Measurement results from optical microscope images, shown in Fig. 2B for channel height and Fig. 2C for channel width, highlight the relationship between LCT and printed channel dimensions. The proportional relationship between increasing LCT and light propagation into the z dimension (thickness) of the masked regions (i.e., channels) results in smaller than designed channel heights. By comparison, the dimensional accuracy for a given channel width depends primarily on the size of the DLP printer pixels (xy plane resolution) rather than LCT. The observed positive channel width variation with decreasing LCT indicates incomplete photopolymerization. Subsequent postprocessing removal of uncured resin yields channels with dimensions greater than designed. In combination, these results establish the printable region for an epifluidic design as a function of LCT. As shown in Fig. 2 (B and C), successful fabrication of a 100-μm square channel requires a short LCT (i.e., 0.54 and 0.8 s), whereas a longer LCT results in photopolymerization of the otherwise unreacted resin. Conversely, for large dimensions (>700 μm square channels), a short LCT produces channels too fragile to survive printing and postprocessing due to incomplete photopolymerization. These results establish an LCT of 0.8 s as the optimal setting for balancing printability with dimensional accuracy for the printed epifluidic devices described in subsequent sections.

Additional systematic experiments establish the DLP-printable design space for epifluidic-relevant dimensions (100 to 600 μm). Evaluation of print success as a function of channel dimensions (width and height) for an enclosed microfluidic channel (length: 30 mm) identifies the printable region (Fig. 2D). An encapsulated microfluidic channel capable of supporting unrestricted fluid flow, in contrast to a sealed or partially restricted channel, defines a successful print. Intuitively, print failure rate increases as the enclosed channel dimensions approach the printer xy plane resolution limit (~32-μm x 32-µm square pixel). Results show a channel dimension of 100 μm (either width or height) corresponds to the lower limit for a successful printed device.

Print process optimization to support colorimetric analysis in 3D printed epifluidic systems

The optical transparency of a 3D printed microfluidic device depends on several factors including material selection, printer hardware (e.g., build plate and vat surface material), postprocessing, and surface roughness. In contrast to the typical surface roughness feature size necessary for optical transparency (<10 nm) (41), DLP printers produce parts with microscale surface roughness, resulting in a semi-translucent appearance (32).

As mentioned previously, the digital micromirror device (DMD) pixel size governs the xy plane resolution of a DLP printer. Minute gaps between individual DMD elements locally reduce reflected light intensity, yielding a surface roughness with features corresponding to DMD pixel size and layer height. While specialized printing methods (grayscale) (42) or printer hardware (oscillating lenses) (43) offer sophisticated strategies to reduce aliasing and improve surface roughness, the fundamental approach to eliminating this defect mode is enhancing the uniformity of projected light to ensure complete photopolymerization. Figure 3A illustrates that increasing the exposure dose by lengthening the LCT eliminates the observed grid pattern defects (from DMD element gaps) and improves optical transparency. Ultraviolet-visible (UV-Vis) spectroscopy experiments examine the transmission properties of 3D printed microcuvettes in comparison to a commercial plastic cuvette (Fig. 3B). While results show substantial modulation of light transmission with increasing LCT, ranging from ~20 (LCT: 0.54 s) to ~60% (LCT: 2 s), the reference commercial plastic cuvette offers higher light transmission (~80%). Intuitively, there is no observed wavelength dependence for light transmission within the Vis spectrum (400 to 1100 nm) beyond the anticipated strong absorbance within the UV region (<400 nm, necessary for photopolymerization) for the 3D printed samples. As a consequence of the presence of both the UV absorber and photoinitiator in the resin, green parts (i.e., before curing) have a light yellow hue. As presented in Materials and Methods, completion of the postprocessing sequence eliminates part coloring (fig. S2).

Figure 3. Optimized design strategy for enabling colorimetric analysis in 3D-printed epifluidic systems. (A) Optical micrographs of the surface of parts printed with different LCT settings. (B) Plot of light transmission of commercial and resin-printed cuvettes measured with ultraviolet-visible (UV-Vis) spectrometer. (C) Photographs of epifluidic reservoirs fabricated using static (0.54 s, 2-s LCT) and adaptive (AP1 and AP2) printing processes illustrating differences in optical transparency. (D) Calibration curves as a function of LCT highlighting improvement in optical transparency (and thus colorimetric performance) with increasing LCT. A.U., arbitrary units.

In addition to LCT, layer height affects both overall device quality (e.g., vertical resolution, optical clarity, and channel roughness) and print time, which corresponds to device yield. Conventional approaches to vat photopolymerization use constant values for a given print run (i.e., fixed layer height and LCT). At present, only one manufacturer [Formlabs (44)] supports an adaptive layer height process to increase print speed by adjusting layer height as a function of model detail (i.e., small layers for fine features and thick layers for coarse features). Adaptive printing is an attractive process for obtaining expanded design flexibility for 3D printed epifluidic systems. Although not supported by default, a combination of custom software and manual geometric code programming in this work enables definition of both layer height and LCT as a function of model dimensions. The representative example shown in fig. S3 illustrates the capabilities of this adaptive printing process to fabricate a cube (all dimensions: 2 mm) using four layer heights (5, 10, 30, and 50 μm) in an arbitrary order. By comparison to a constant LCT and layer height setting printing process, this approach enables successful, time-efficient fabrication of epifluidic systems with complex geometries and superior device quality.

Colorimetric assays facilitate passive, battery-free in situ quantitative measurement of sweat biomarkers. A chemical reagent reacts with a target species to generate an optical signal proportional to analyte concentration (45). Accurate colorimetric analysis requires channels with uniform height (i.e., path length), a high degree of optical transparency, and integrated color reference markers to support reliable image processing under variable ambient lighting conditions (46). The layer-by-layer control over LCT and layer height parameters enabled by an adaptive printing process is critical for fabricating microfluidic devices with the requisite surface finish and optical transparency to support colorimetric analysis. Figure 3C illustrates the influence of an adaptive LCT print process on the optical transparency of microfluidic channels. The optical clarity for two representative sweatainer reservoirs manufactured using a layer-constant LCT (0.54 and 2 s) increases with longer LCT (Fig 3C). While beneficial for reducing nonuniform illumination, the increased UV dose results in undesirable curing of resin in enclosed features (channels and CBVs). By comparison, an adaptive printing process (AP 1) using an LCT of 0.54 s for the reservoir surface and an LCT of 2 s for subsequent layers facilitates fabrication of a sweatainer with a translucent imaging plane, a transparent device, and preservation of internal channel features. An inverse adaptive printing process (AP 2; base LCT: 2 s and subsequent layer LCT: 0.54 s) results in an optically transparent imaging plane and a translucent device.

Systematic benchtop experiments evaluate the suitability of devices fabricated by adaptive printing for colorimetric analysis. The colorimetric assay silver chloranilate produces a dark violet color response proportional to chloride concentration. Imaging the device with a smartphone camera enables color extraction and subsequent quantification of color response. The inclusion of a color balance chart facilitates color calibration for each image. As in previous reports (47, 48), converting images from native red, green, blue (RGB) color space to CIELAB color space—which expresses color as lightness (L), amount of green to red (a^*), and amount of yellow to blue (b^*)—ensures device-independent color sampling. Conversion of the a* and b* components to chroma (C^*) by the relation

yields a calibration curve with chloride concentration by a power-law relation (fig. S4). Figure 3D shows calibration charts created from 3D printed sweatainers with different LCT parameters and reference colorimetric assay solutions. This plot reveals that the improvement in optical clarity with increasing LCT provides a corresponding enhancement in the range of detectable color measurements. As these findings indicate, an adaptive printing process is essential for fabricating epifluidic devices with an optical transparency sufficient to support colorimetric analysis.

3D CBV designs for sequential sweat analysis

CBVs are a key component for the sequential analysis of sweat biomarkers in many epifluidic platforms. The time dynamic variations in sweat rate arising from physical (e.g., sweat gland density), physiological (e.g., exertion and emotion), and external factors (e.g., temperature and pH) result in corresponding changes in analyte concentration. As previously described, CBVs prevent flow for fluid pressure conditions below a designed threshold [bursting pressure (BP)]; when the fluid pressure exceeds the BP, the CBV immediately bursts. Operating without use of actuation or moving components, CBV BP is governed by valve geometry.

The Young-Laplace equation describes the BP for a CBV (rectangular channel) as (49)

where σ is the fluid surface tension, θ_A is the critical advancing contact angle for the channel (material dependent, θ_A = 120° for PDMS) (50), θ*_I is the minimum of either θ_A + β or 180°, β is the channel diverging angle, and b and h are the diverging channel width and height, respectively. As the second term of Eq. 2 is constant for a planar (2D) CBV, channel width and diverging angle govern the BP for a given CBV. In practice, epifluidic device designs use geometric restrictions (i.e., modifications to channel width) to control valve BP.

The 3D printing concept for epifluidic devices presented here expands CBV capabilities by enabling a full 3D CBV design. As a consequence, Eq. 2 can be written as

for a 3D CBV, where θ*J is the minimum of either θ_A + γ or 180° and γ is the channel diverging angle (z axis). It follows that for a microfluidic channel with fixed dimensions, the CBV BP becomes a function of the channel diverging angles (β and γ). Computational predictions of four representative CBV designs, presented as a schematic in Fig. 4A with parameters specified in Table 1, illustrate this relationship. Figure 4B shows the theoretical BP versus channel size (square channel) for the four CBV designs with σ = 0.072 N/m (surface tension of water) and θ_A = 120° (PDMS) for the 2D CBV (type 1) and θ_A = 60° for the 3D CBVs (resin, types 2 to 4). It is shown that BP is inversely proportional to channel size. As expected, the analytical model reveals that for a given channel size BP increases for 3D CBV designs (resin) in comparison to a 2D CBV (PDMS). Within the subset of 3D CBV designs, the channel diverging angles (β and γ) dictate the valve BP (BP_Type4 > BP_Type3 > BP_Type2).

Figure 4. 3D CBV designs for sequential sweat analysis. (A) Schematic renders highlighting four design types of CBVs used in this work. Areas highlighted in blue indicate differences between CBV designs. (B) Plot of the theoretical maximum bursting pressures (BPs) calculated from the Young-Laplace equation as a function of channel size for a square geometry. (C) Sequence of photographs illustrating the performance of different CBV designs (labels 1 to 8). Use of backside illumination for the overview photograph facilitates visualization of valves and channels. (D) A sequence of photographs shows a 3D printed H channel with one central inlet and four CBVs (color indicates CBV design and fixed channel geometry) filling sequentially, highlighting the fluid control enabled by a true 3D CBV. (E) Plot of the theoretical BP as a function of diverging angle β for a channel with a fixed geometry (width: 600 μm and height: 400 μm). The CBV designs are identical to (B). (F) A sequence of photographs highlighting performance of the 3D printed sweatainer design used in human participant testing.

Benchtop experiments yield measurements of CBV BPs by means of a positive pressure displacement pump apparatus that perfuses water (dyed blue for visualization) into the microfluidic network at defined pressures. Figure 4C shows a representative test of the sequential filling performance of a network of 2D and 3D CBV-gated reservoirs, labeled chronologically in order of increasing BP. Table 2 and fig. S5 detail the CBV design parameters, theoretical CBV BPs, and effective theoretical BPs, which consider the theoretical CBV BP and fluidic resistance of the microfluidic channel network. Imperfections resulting from the 3D printing process result in experimental BP values below theoretical limits.

The 3D design space provides attractive capabilities for fine-scale control over CBV performance to enable compact fluid control features within epifluidic devices. Varying the diverging angle design parameters (β and γ) for a 3D CBV results in substantial differences in BP for valves with similar dimensions and form factors. Systematic experiments performed in similar manner as described previously verify the correlation between diverging angle and BP for the 3D CBV architectures illustrated in Fig. 4A with identical channel dimensions. Figure 4D shows a representative test of 3D CBV performance via a 3D printed microfluidic device with channels arrayed in an H configuration (channel dimensions: 600-μm width and 400-μm height). As Fig. 4E highlights, BP increases with β for a resin-based 3D CBV in contrast to a PDMS-based 2D CBV baseline reference. Material properties limit the valve design space on account of the BP dependence on contact angle. For hydrophobic materials such as PDMS (i.e., contact angle >120°), β values greater than 60° reduce to 180°, resulting in BP value dependent only on channel width (b). By comparison, the expanded design range, in which β governs valve BP, results from the smaller contact angle of hydrophilic materials (i.e., resin). The experimental results support these trends predicted by the analytical model with the variation between measured and predicted values attributed to geometric imperfections inherent to the fabrication process (i.e., slight rounding of corners) (51). A similar trend occurs for valve designs in which γ varies with respect to a fixed β.

Additional studies demonstrate 3D CBV performance in a device architecture relevant to practical use. Robust operation requires CBV designs with BP within the physiologically relevant range of sweat secretory pressure (0.5 to 2 kPa) (51). Tests of the sweatainer design shown in Fig. 4F proceed in the same manner whereby water enters the device through a central inlet. Reservoirs fill sequentially in the order indicated as the CBVs at entrance of reservoirs no. 2 and no. 3 prevent fluid flow until reservoir no. 1 fills completely. Variation of CBV diverging angle defines the BP for CBV no. 1 (blue, 0.66 kPa) and CBV no. 2 (red, 0.86 kPa). These results validate the design of the sweatainer for use in on-body testing.

Field studies of the sweatainer

A pilot study comprising healthy adult volunteers (N = 8) exercising on a stationary bike explores the on-body performance of the sweatainer system. Following the protocol detailed in Materials and Methods, the sweatainer intimately couples to the ventral forearm of a participant by means of the epidermal port (PDMS/skin-safe adhesive). Participants cycled at moderate intensity for a period of 50 min under controlled environmental conditions [22°C, 59% relative humidity (RH)]. Upon entering the device from the skin, sweat proceeds to sequentially fill the microfluidic reservoirs. The addition of chloride-free dye at the sweatainer inlet aids in visualization. Periodic imaging with a smartphone camera during exercise facilitates monitoring fill performance. The sweatainer typically fills within 40 min from the initiation of exercise; after filling, the device is exchanged mid-exercise with a new sweatainer in a seamless manner. Figure 5A highlights this event sequence with sweatainers distinguished by distinct visualization dyes (device no. 1: blue and device no. 2: orange). The simplicity of the exchange facilitates a rapid replacement time (<30 s), thereby minimizing potential interruption to the sweat collection process. In all tests, the adhesive gasket maintains a robust, water-tight interface between the sweatainer and epidermal port evidenced by the absence of observed leaks. The 3D printed sweatainer resists mechanical deformation during detachment, thereby eliminating unconstrained fluid flow. In combination, these features support the multidraw collection of pristine sweat samples and reduce the risk of sample contamination during collection process.

Figure 5. Sweatainer field studies. (A) Sequence of photographs highlighting operation of the sweatainer system. A sweatainer (device no. 1) collects sweat during an active exercise period, which, upon filling, is rapidly exchanged (<30 s) for a second sweatainer (device no. 2) facilitating multidraw sweat collection. (B) Photograph of sweatainer position during exercise trials (blue box). (C) A magnified view of the same sweatainer devices shown in (B) before the onset of sweating. The sweatainer shown on left is for collection (control) with the device on the right for colorimetric analysis. (D and E) respectively show the collection and colorimetric sweatainers at the conclusion of the exercise period. (F) Plot showing the concentration of sweat chloride from the collection (chloridometer) and colorimetric sweatainers for three independent exercise trials for a single participant (stationary cycling, 50 min, constant power). (G) Plot of showing sweat chloride concentration from two different colorimetric sweatainers worn sequentially (i.e., replaced during trial) during a predefined exercise period (stationary cycling, 50 min, constant power). The total sweat volume lost by a given individual during this exercise period corresponds to the total number of filled sweatainer chambers. Scale bars, 5 mm.

A second set of exercise tests focuses on the in situ measurement of the concentration of sweat chloride by colorimetric analysis. A sweatainer configured with an integrated colorimetric assay (replacing visualization dye) enables measurement of chloride concentration in collected sweat during exercise. Figure 5B shows the sweatainer mounted on the ventral forearm of a volunteer participant. Simultaneous deployment of a collection sweatainer (orange dye alone) in close spatial proximity of a colorimetric sweatainer (Fig. 5C) facilitates comparison of colorimetric chloride measurements with gold standard clinical methods for chloride analysis (chloridometer). The collection sweatainer operates in a similar mode to microbore tubes (i.e., Macroduct) traditionally used in clinical settings for collecting sweat for chemical analysis. Representative photographs of the colorimetric and collection sweatainers at the conclusion of an exercise period appear in Fig. 5 (D and E, respectively). As shown in Fig. 5F, for a representative participant, chloride concentrations measured using colorimetric sweatainers correlate well for given individual (Fig. 5F reports three independent exercise trials), within experimental uncertainties, to values determined using coulometry and are within the normal physiological range (48). The chronological sampling capabilities of the sweatainer enable monitoring of the time dynamic variation of sweat biomarkers. Figure 5G demonstrates the multidraw sweatainer operation for three participants (field study data for remaining five participants shown in table S2) during a fixed exercise period (stationary cycling, 50 min, constant power). In both sets of trials, the observed increase in sweat chloride concentration during exercise is consistent with results from previous studies (52). Here, an inverse relationship exists between the sweat duct efficiency in reabsorbing chloride and rate of sweat loss, resulting in a corresponding increase in sweat chloride. Factors such as fitness level, training status, and heat acclimation affect this relationship for a given individual. These findings demonstrate the sweatainer system as a viable platform for colorimetric-based biomarker analysis with reported values comparable to established clinical methods.

Discussion

The sweatainer system reported here introduces an AM approach to fabricating epidermal microfluidic devices to collect and analyze sweat. AM enables true 3D design of microfluidic channels and fluid control components, such as valves, with architectures typically inaccessible to planar (2D) fabrication methods. The detailed characterization and optimization of print parameters provides a pathway to fabricate microfluidic devices with enhanced optical transparency and feature sizes below 100 μm. Field studies using stationary cycling provide a practical demonstration of key concepts of the sweatainer platform including multidraw sample collection scheme and in situ colorimetric analysis of chloride concentration. Future studies will seek to investigate the generalizability of the sweatainer platform beyond clinical applications to sweat collection during more vigorous and dynamic physical activities through the development of optimized designs capable of supporting a broader spectrum of physical exertion.

The sweatainer platform represents a pivotal advancement in the collection and analysis of sweat samples. Inspired by the versatility of the vacutainer for blood collection, the sweatainer allows for the acquisition of multiple, independent aliquots of sweat from a single collection period. This collection mode enables an array of possibilities for sweat-based studies, including remote and at-home diagnostics, biobanking for future clinical research, and the integration of sweat analysis into existing clinical chemistry methods. Moreover, the utilization of AM for fabricating the sweatainer allows for customized geometries and streamlined integration into clinical workflows, further enhancing the potential of the platform for facilitating the quantification of ultralow concentration sweat biomarkers. The realization of multidraw sweat collection, enabled by the sophisticated sample collection strategies and customizable designs reported here, represents a major step forward in the field of sweat-based analysis.

Materials and Methods

Fabrication of 3D printed epifluidic devices

Each epifluidic device design (3D) was created using CAD software (AutoCAD 2019, Autodesk, CA, USA). Subsequent export to a stereolithography file format (.stl) yielded a file suitable for direct use by the DLP resin printer (Prime 110, 385 nm, MiiCraft, Taiwan and Creative CADworks, ON, Canada). The included printer control software (Utility, version 6.3.0.t3) provided direct control over print parameters for each file including layer height (5 to 50 μm), dose, and lamp power. High-fidelity printing was achieved by application of a removable Kapton polyimide tape over the surface of the polished aluminum build plate. The applied tape was free of bubbles/wrinkles to ensure a smooth build surface free of defects.

Devices were printed using transparent resin (MiiCraft BV-007A, Creative CADworks, ON, Canada) and a 10-μm layer height (six devices per build plate, ~20 min total print time). Gentle removal of printed parts from the build plate, soaking in 1% detergent solution (Alconox-1232-1, Alconox, NY, USA) under sonication (CPX2800, Fisher Scientific, PA, USA) for 10 min, drying of device using clean dry air (CDA), postprint UV cure for 10 min (CureZone, MiiCraft, Taiwan), and postcure bake at 70°C for 30 min (Model 40E Lab Oven, Quincy Lab Inc., IL, USA) yielded a 3D printed epifluidic device suitable for direct use or integration with PDMS.

A three-step process (fig. S6) facilitated printing fully enclosed 3D printed devices. Printing epifluidic devices with open reservoirs (step 1) and postprint removal of uncured liquid resin by CDA (step 2) enabled enclosure of the devices with a thin capping layer (30 μm) by means of a second print process (step 3). The printed device remains fixed to the build plate during the printing process to ensure feature alignment. Following the previously described postprocessing steps yielded a fully enclosed epifluidic device.

Apparatus Used

Clear Microfluidic Resin

Learn More

Curezone

Learn More

PR110
3D Printer

Legacy

Fabrication of ultrathin capping layer for microfluidic channels in hybrid devices

Pouring liquid PDMS (10:1 base:curing agent; Sylgard 184, Dow Inc., MI, USA) with white pigment (3% w/w; Ignite White, Smooth-On Inc., PA, USA) onto a sacrificial mylar film (2 mil thickness), spin coating for 30 s [400 revolutions per minute (rpm) for reservoir capping layer and 200 rpm for epidermal interface layer], and curing in an oven (70°C, 2 hours) formed films with thicknesses of 200 and 400 μm, respectively. A CO₂ laser cutter (30 W Epilog Mini 24, Epilog Laser, CO, USA) patterned the PDMS films into the final geometries used in the epifluidic devices. A medical-grade adhesive (1524, 3M Inc., MN, USA) is patterned in the same manner and bonded to the PDMS interfacial layer, established the epidermal interface for the device.

Hybrid 3D printed epifluidic devices use bonded PDMS capping layers to enclose 3D printed microfluidic reservoirs. Modification of a previously reported method (53) facilitated a strong bond between PDMS and the printed device. Specifically, rinsing with isopropyl alcohol (2-propanol, A416, Fisher Scientific, MA, USA), soaking in deionized (DI) water (Direct-Q 3 UV Water Purification System, MilliporeSigma, MO, USA) for 30 min, corona treating with air plasma (BD-20, Electro-Technic, IL, USA) for 30 s followed by immediate immersion in a 12% v/v solution of (3-aminopropyl)triethoxysilane (APTES; 440140, MilliporeSigma, MO, USA) for at least 20 min, rinsing in DI water, and drying with CDA prepared the oven-baked 3D printed device for bonding to PDMS. Pipetting colorimetric reagents or flow visualization dye (Soft Gel Paste, AmeriColor Corp., CA, USA) into predetermined regions occurred before sealing of the 3D printed device. After a 30-s corona treatment, laminating the PDMS capping layer to the APTES-modified printed surface sealed the epifluidic device. Heat treating the assembled device on a hotplate (70°C) under applied weight (3 kg) for 30 min formed a permanent bond. Removal of the sacrificial mylar layer, release from excess PDMS via laser cutting, and opening the central sweat ingress points using a 1.5-mm diameter circular punch (reusable biopsy punch, World Precision Instruments) yielded a final hybrid epifluidic device.

Measurement of evaporation rate for 3D printed microfluidic networks

3D printed microfluidic devices (N = 7) with theoretical capacity (~101 ml) facilitated the measurement of the rate of evaporation. Sealing of the inlet and outlet of a device with parafilm after filling with DI water (dyed blue for visualization) formed the complete device for testing. Measurement of the initial sealed device mass (inclusive of water, film, and printed microfluidic system) using a microbalance (Sartorius Quintix 224-1S, Germany) enabled recording of mass loss at 2 and 24 hours. Devices were maintained at room temperature in a controlled laboratory environment reflective of anticipated use environment (22°C, 55% RH). An optical camera (Canon 90D, Canon EF 100 mm f/2.8L USM lens) facilitated observation of visual changes to fluid levels at each measurement interval.

Characterization of 3D CBVs

A digital microscope (VHX-7100, Keyence Corp., Japan) produced micrographs of the devices. An optical camera (Canon 90D, Canon EF 100 mm f/2.8L USM lens) provided video capture capabilities (30 frames per second) for device analysis. Measurement of the CBV burst pressure consisted of a “fill test” in which water (dyed blue for visualization) entered a device until flow stopped the CBV. A modular, calibrated pressure displacement flow system (Flow EZ, Fluigent, France) controlled the fluid pressure and permitted near-instantaneous stepwise increase in pressure (0.1-mbar interval, 10-s dwell time). Video observation identified the pressure threshold for fluid to burst the valve.

Measurements of transmission properties of 3D printed devices

A UV-Vis spectrophotometer (UV-1900i, Shimadzu, Japan) enabled quantification of the optical transmission properties of the printed devices (300 to 1000 nm, 0.5-nm interval). A commercial plastic cuvette (path length: 10 mm; Shimadzu) served as a reference (control). Four sets of 3D printed cuvettes (N = 3 per set) using a different LCT setting (0.54, 0.8, 1.4, and 2 s) enable quantification of the relationship between LCT and optical transmission (dimensions: height, 50 mm; width, 8 mm; path length, 1 mm; and volume, ~21 μl).

Colorimetric assay for chloride

The chloride colorimetric assay solution resulted from thoroughly vortexing 50 mg of silver chloranilate (MP Biomedicals, CA, USA) in 200 μl of a solution of 2% (w/v) polyhydroxyethylmethacrylate (529265, MilliporeSigma, MO, USA) in methanol (A412, Fisher Scientific, MA, USA) to yield a homogenous suspension. Spotting 2 μl of this solution via laboratory pipette onto the 3D printed device near the central sweat ingress point, followed by drying in an oven for 30 min before encapsulation, prepared the epifluidic device for colorimetric chloride measurements.

Standard color development and color reference marker preparation

Mixing sodium chloride (S271, Fisher Scientific, MA, USA) in DI water produced standard test solutions (0, 10, 20, 30, 50, 75, 90, 110, 130, and 150 mM). Clinical-grade chloridometer measurements (ChloroChek, ELITech Group Inc.) yielded validated test solution concentrations. Digital imaging and analysis of sample reservoirs (N = 7) containing one standard solution reacted with the silver chloranilate assay under uniform illumination formed a set of reference images. The sample reservoirs were of the same depth as the epifluidic channels to ensure accurate color representation.

Digital image analysis for the evaluation of sweat chloride concentrations

A smartphone camera (iPhone 11 Pro Max, Apple, CA, USA) captured images during on-body field tests. A color calibration card (ColorChecker Classic, X-Rite, MI, USA) in the frame of each image facilitated accurate color extraction under various illumination conditions. An open-source photography software package (Darktable 3.0.0, Darktable.org) served as the platform for calibrating images using the color reference card. Analysis of calibrated images using MATLAB (R2019b, MathWorks Inc., MA, USA) enabled cropping multiple regions of interest (N = 3) from images and extraction of CIELAB color values (L, a*, and b*) for chroma analysis. Mapping of chroma values from colorimetric samples of known reference chloride solutions yielded colorimetric calibration charts with a power-law relationship. This calibration chart supported quantification of the sweat chloride concentration in on-body field testing.

Human participant sweat analysis

The purpose of this pilot study was to evaluate the performance of the 3D printed epifluidic device and use in collecting and analyzing sweat. Testing involved healthy young adults (N = 8, six male and two female) as volunteers during normal physical activity (stationary cycling) with no additional human participant risk. The study was International Review Board (IRB) approved through the University of Hawaiʻi (IRB no. 2018-1440). Informed consent was obtained after explanation of the nature and possible consequences of study participation.

Cleaning of the forearm of each individual with an alcohol wipe prepared the skin for robust adhesion to the device. The exercise regime comprised stationary cycling under approximately constant working load for 50 min in a controlled laboratory environment (22°C, 55% RH).

Evaluation of the colorimetric sweatainer performance required individual participants (N = 3) to wear two separate sweatainers, one colorimetric and one collection (as a control), located in close proximity on the same arm. Before device removal, a photograph of the colorimetric sweatainer was recorded at the conclusion of the collection period for image processing and chloride analysis. Extraction of sweat from the individual reservoirs of the collection sweatainer at the conclusion of the exercise period facilitated chloride measurements using a ChloroChek Chloridometer.

Evaluation of sequential generation of aliquots of sweat required periodic monitoring the filling of the epifluidic device (N = 8). Once all reservoirs filled, as determined by visual observation, the initial device (attached at the start of the exercise period) was removed from the interfacial layer and replaced with a new device while simultaneously continuing to exercise.

Applied tutorial for the design and fabrication of biomicrofluidic devices by resin 3D printing

March 12, 2026April 15, 2022 by CadWorks3DAdmin

Applied tutorial for the design and fabrication of biomicrofluidic devices by resin 3D printing

by Hannah Musgrove, Megan Catterto and Rebecca Pompano

Abstract: Stereolithographic (SL) 3D printing, especially digital light processing (DLP) printing, is a promising rapid fabrication method for bio-microfluidic applications such as clinical tests, lab-on-a-chip devices, and sensor integrated devices. The benefits of 3D printing lead many to believe this fabrication method will accelerate the use of microfluidics, but there are a number of potential obstacles to overcome for bioanalytical labs to fully utilize this technology. For commercially available printing materials, this includes challenges in producing prints with the print resolution and mechanical stability required for a particular design, along with cytotoxic components within many SL resins and low optical compatibility for imaging experiments. Potential solutions to these problems are scattered throughout the literature and rarely available in head-to-head comparisons. Therefore, we present here a concise guide to the principles of resin 3D printing most relevant for fabrication of bioanalytical microfluidic devices. Intended to quickly orient labs that are new to 3D printing, the tutorial includes the results of selected systematic tests to inform resin selection, strategies for design optimization, and improvement of biocompatibility of resin 3D printed biomicrofluidic devices.

Keywords: digital light processing, stereolithography, SLA, photopolymerizable resins, microfluidic fabrication, cell culture

We kindly thank the researchers at University of Virginia for this collaboration, and for sharing the results obtained with their system.

1. Introduction

In microfluidic device development, a recurring theme is to complete bioanalytical assays at a fraction of the time and cost required for macroscale methods. This aspiration makes rapid and accessible fabrication of microfluidic devices a key goal. Historically, microfluidic fabrication relied heavily on soft lithography methods such as casting polydimethylsiloxane (PDMS) on hard micropatterned masters.^1,2 Soft lithography is primarily a 2D fabrication method, in which multi-layer devices are generated by tedious or challenging manual alignment and bonding. This process is sensitive to any dust that falls onto the layers during aligning and bonding, especially if conducted outside a cleanroom environment. Thus, in order to produce complex 3-dimentional devices monolithically and without a clean room, many groups have turned to 3D printing as a simplified workflow for rapid fabrication in the laboratory.3–6

Since the early 2010s, stereolithographic (SL) 3D printing has emerged as a promising technique for fabricating microfluidic devices.^5,7,8 Briefly, this technique works by curing photopolymerizable resins with a UV or visible light source in sequential layers that build on top of one another. Stereolithographic apparatus (SLA) printers were some of the first SL printers and utilize a UV laser guided by mirrors, curing resin point-by-point in a scanning manner in the x and y directions. Direct light processing (DLP) printers were developed later and utilize UV projectors that allow an entire layer to be cured simultaneously from a direct light path. Because of their direct light path, DLP printers tend to have slightly better resolution compared to the similar SLA printers.^5,9 The development of higher resolution DLP printing, along with other SL printers, has increased the use of 3D printing in fields of dentistry, audiology, medicine, and microfabrication.

Compared to traditional fabrication methods, 3D printing reduces cost and fabrication time while increasing product customization.^5,10 Having the ability to readily customize microfluidics also expediates the “fabrication-application-results” process for devices.^1,11 Indeed, many bioanalytical 3D printed devices have been well documented in reviews.^12–16 Applications have ranged widely from bioreactors and probes made for direct contact with a range of cell lines,^17–20 to approaches that utilized 3D-printed molds to cast microfluidic devices in other materials.^21–23

Like all fabrication methods, 3D printing requires compromises between desirable features, e.g. resolution and cytocompatibility, and the past 5 years have seen an explosion of papers seeking to address key limitations. The ability to print leak-free devices, small internal microchannels (<1000 μm), and biocompatible, optically-clear devices depends on several factors that will be discussed further below, including aspects that affect mechanical integrity and the rate and resolution of photopolymerization (Fig.1).^24–26 For commercial resins, strategies to diminish resin cytotoxicity have started to emerge recently,^27–30 along with solutions addressing print resolution,^31–36 imaging compatibility,^33,37–39 and surface modification and bonding.^40,41 Sifting through the plethora of literature for best practices can be challenging for researchers who are new to this rapidly growing field.

Figure 1. Schematic of a DLP 3D printer, highlighting the design and mechanical factors as well photopolymerization parameters which influence print resolution. Digital light processing 3D printers project UV or violet light through optically clear sheets (usually Teflon) into a vat of photopolymerizable resin (pink). In locations where the light is projected, the resin crosslinks to form a solid structure. Exposure and crosslinking are performed layer by layer on the base plate, which lifts up as each concurrent layer is formed. Production of a clean print is dependent on instrumental, environmental, chemical, and design elements that impact either the print surface (base plate), mechanics (print orientation, design specifics), or chemical reaction (resin composition, light source, exposure settings, temperature, or humidity).

Here we present an applied, data-supported guide to the key factors that should be considered for design and fabrication of SL 3D printed biomicrofluidic devices, written especially for groups that are new to this growing field. Types of instrumentation and resin formulations have been reviewed and characterized in depth recently, and will not be presented in detail here.^8,16,42–44 This tutorial follows the order of a typical workflow by first considering resin selection and then demonstrating how to increase print resolution by using simple changes in feature design and printer settings. Following printing, cytotoxicity of materials is addressed, particularly the extent and effectiveness of post-treatment strategies for applications involving contact with primary cells. Finally, preparation and considerations for use with fluorescent microscopy is outlined with data displaying autofluorescence and optical clarity of materials. We include the results of systematic, head-to-head comparisons of printing and post-treatment conditions designed to optimize the integrity of a printed piece, the resolution of interior channels, and the biocompatibility of the part. It is our hope that this work will streamline the adoption of 3D printed devices by more specialized biomedical research fields, bioanalytical laboratories, and others new to 3D printing.

2. Materials and Methods

2.1. 3D Design and Printing

All printed pieces used for this work were designed either in Autodesk Fusion 360 2020 or Autodesk Inventor 2018 and exported as an .stl file. DWG files of prints shown in figures can be found in the supplementary information. Files were opened in the MiiCraft Utility Version 6.3.0t3 software, where pieces would be converted into sliced files with appropriate layer heights. The files were converted to the correct file format to include settings optimized for either the CADworks3D M50−405 printer (MiiCraft, CADworks3D, Canada), which had a 405 nm light source, or for the CadWorks 3D Printer P110Y, which had a 385 nm source. All prints were printed in one of three resins: FormLabs BioMed Clear V1 (FormLabs, USA), FormLabs Clear V4 resin (FormLabs, USA) or MiiCraft BV007a Clear resin (CADworks3D, Canada). These resins were chosen as representative of those formulated for biocompatibility (FL BioMed), standard clear printing (FL Clear), or high-resolution microfluidics (MC BV007a).

Apparatus Used

Clear Microfluidic Resin

Learn More

PR110
3D Printer

Legacy

After printing, all materials were rinsed with isopropyl alcohol using the FormWash from FormLabs, following the wash suggestions from the FormWash online guide.⁴⁵ Alternatively, materials were placed in a container with IPA and placed on a rocker for longer periods, extending times by 5 minutes for low viscosity materials and by 1 hour for more viscous resins. Once residual resin was rinsed from the prints, the pieces were dried with compressed air and post-cured with additional UV dosages using the FormCure (FormLabs, USA). Specific print and post-print settings for each resin and printer can be found in Table S1.

2.2. Viability testing with primary murine lymphocytes

2.2.1 Primary Cell Preparation

Animal work was approved by the Institutional Animal Care and Use Committee at the University of Virginia under protocol #4042 and was conducted in compliance with guidelines from the University of Virginia Animal Care and Use Committee and the Office of Laboratory Animal Welfare at the National Institutes of Health (United States). Following isoflurane anesthesia and cervical dislocation, spleens were harvested from female and male C57BL/6 mice between 8-12 weeks old. The spleens were collected into complete media consisting of RPMI (Lonza, Walkersville, MD, USA) supplemented with 10% FBS (VWR, Seradigm USDA approved, Radnor, PA, USA), 1× l-glutamine (Gibco Life Technologies, Gaithersburg, MD, USA), 50 U/mL Pen/Strep (Gibco, MD, USA), 50 μM beta-mercaptoethanol (Gibco, MD, USA), 1 mM sodium pyruvate (Hyclone, Logan, UT, USA), 1× non-essential amino acids (Hyclone, UT, USA), and 20 mM HEPES (VWR, PA, USA).

To produce a splenocyte suspension, harvested spleens were crushed through a 70-μm Nylon mesh filter (Thermo Fisher, Pittsburgh, PA, USA) into 10 mL of complete media. The cells were then centrifuged for 5 minutes at 400 xg. The pellet was resuspended into 2 mL of ACK lysis buffer, which consisted of 4.15 g NH₄CL (Sigma-Aldrich, St. Louis, MO, USA), 0.5 g KHCO₄ (Sigma, MO, USA), 18.7 g Na₂EDTA (Sigma, MO, USA) into 0.5 L MiliQ H₂O (Millipore Sigma, Burlington, MA, USA). The cells were lysed for 1 minute before being quenched by bringing up the solution to 10 mL with complete media and immediately centrifuging again. The pellet was resuspended into 10 mL of complete media, and density determined by trypan blue exclusion. To prepare for cell culture, the suspensions were diluted with complete media to a concentration of 1×10⁶ cells/mL of media.

2.2.2 Print preparation for biocompatibility studies

Disks with a diameter of 15 mm and a height of 1 mm, designed to fit snugly against the base a 24-well plate, were printed in all three representative resins (BioMed, Clear, and BV007a) following the print settings outlined in Table S1. These pieces were divided into “non-treated,” with no further post-treatment, or “treated”. The latter prints were post-treated by soaking in sterile 1x phosphate-buffered saline without calcium and magnesium (PBS, Prod. No. 17-516F, Lonza, USA) for 24 hours at 37°C for BV007a or at 50°C (BioMed and Clear resins) to mitigate cytotoxicity, with similar treatments shown to be effective in previous works.⁴⁶

To compare post-treatment strategies for the BioMed resin, disks were printed as above. The post-treatments included a 24-hr PBS soak at room temperature within a biosafety cabinet; 24-hr incubation in a 37°C cell culture incubator while dry or soaked in PBS; or autoclaving for 30 minutes at 120°C gravity cycle. In all cases, both treated and untreated pieces were rinsed again with IPA, dried, and UV sanitized for an additional 10 minutes before use.

2.2.3 Analysis of cell viability

Aliquots of suspended splenocytes (1 mL, 10⁶ cells/mL) were added to two 24-well plates containing samples of either treated resins or non-treated resins as previously outlined in section 2.2.2. Wells that did not contain any resins were reserved for plate controls. The cell cultures were incubated for 4 hours at 37°C with 5% CO₂. Following the culture period, the viability of the splenocytes was assessed by flow cytometry using a previously established protocol.⁴⁷ Concisely, 500 μL of the cultured samples were stained using Calcein AM (eBioscience, San Diego, CA, USA) at 67 nM in 1x PBS for 20 min at 37°C. The stained samples were centrifuged at 400 xg for 5 min and resuspended in flow buffer (1 x PBS with 2% FBS), after which 4 μL of 1 mg/mL 7-AAD (AAT Bioquest, Sunnyvale, CA, USA) was added. Calcein-AM single stains were prepared using live cells, and 7-AAD single stains were prepared using cells pre-treated for 20 min with 70% ethanol added in a 1:1 v/v ratio to the culture. Additional controls included unstained cells and an ethanol-treated double-stained control. All samples and controls were run on a Guava 4-color cytometer (6-2L) and analyzed with Guava^® InCyte™ Software. Live cells were defined as being high in Calcein-AM and low in 7-AAD signal, while dead cells were defined as the inverse.

2.2.4 Analysis of the viability after direct and indirect contact with treated resin

Indirect contact was defined as cell culture in media that had been conditioned by incubation with printed resin, whereas direct contact was defined as cell incubation in physical contact with the printed resins. To test indirect viability, treated BioMed disks were prepared for cell culture as noted in section 2.2.2. Following treatment, the disks were added to a 24-well plate and incubated in complete media for 24 hours at 37□C. After incubation, 1 mL of suspended splenocytes at 10⁶ cells per mL were spun down and brought back up in 500 μL of resin-conditioned media. All samples were then cultured for 45 minutes, 4 hours, and 24 hours. Viability was analyzed as in section 2.2.3 to determine the percent of live cells present for each sample.

Direct viability was tested in a similar manner. Treated BioMed disks were added to a 24-well plate, and 1 mL aliquots of suspended splenocytes at 10⁶ cells per mL in fresh complete media were added to sample and control wells. Viability was analyzed after 45 minutes, 4 hours, and 24 hours.

2.3. Characterization of material properties of printed pieces

2.3.1 Autoclave compatibility and heat tolerance

To test heat stability of printed pieces, small pieces with square channels (as used for print resolution tests, .DWG files are included in the supplement) were 3D printed in each of the three resin types using settings from Table S1 and autoclaved at 120°C for a 30-minute gravity cycle. Following autoclaving, the pieces were visually evaluated for cracks, delamination, or other alterations to the original design. Similar tests were conducted by leaving the prints overnight in ovens at 37°C, 70°C, and 120°C and then visually assessing the prints for discrepancies after 1, 3, and 7 days.

2.3.2 Autofluorescence

Disks with a diameter of 15 mm and a height of 1 mm were printed in each representative resin. A square piece of PDMS was used as a control. All images were collected on a Zeiss AxioZoom macroscope (Carl Zeiss Microscopy, Germany) with Zeiss filter cube channels including Cy5 (Ex 640/30, Em 690/50, Zeiss filter set #64), Rhodamine (Ex 550/25, Em 605/70, #43), EGFP (Ex 470/40, Em 525/50, #38), and DAPI (Ex ~320-385 nm, Em 445/50, #49). A 500 ms exposure time was used for all images. Following imaging, analysis was performed using Image J v1.530 (imagej.nih.gov). On each image, three 1 × 1 in² regions were analyzed for mean gray value in each channel. Background regions were also measured from the borders in each image (outside of the printed parts) and subtracted from each sample measurement individually. The mean gray intensity was calculated for each resin piece and the PDMS control; higher mean gray intensities represented higher autofluorescence of the pieces.

2.3.3 Optical clarity

Disks with a diameter of 15 mm and a height of 5 mm were printed in the FormLabs Clear resin following the print settings listed in Table S1. Several post-processing methods were compared to determine which had the greatest improvement on optical clarity of printed devices. These included printing on glass, a nail polish coating, a resin coating, sanding, and buffing the pieces. A non-processed piece was used as a control, and a glass slide (0.17 mm thick) was used as a benchmark for optimal material clarity.

To prepare the printed-on-glass piece, a print was set up as previously described by Folch, et al.³⁸ First, small drops of resin were applied to the baseplate using a transfer pipette. A large cover glass with dimensions of 1.42” × 2.36” and a thickness of 0.13-0.17 mm (Ted Pella, Inc. USA) was attached to the baseplate by lightly pressing the slide over the resin then using a UV flashlight to quickly cure the resin between the slide and the baseplate. In the printer software, the initial layer height (“gap height”) was increased by the thickness of the slide (1.7 mm) prior to printing to account for the change in the z-position of the first layer. After printing, the piece was removed from the glass slide with a razor blade and post-cured typically. The glass slide and adherent resin drops were also easily removed with a razor blade.

For acrylate coating, a baseplate-printed piece was coated with generic clear nail polish from a convenience store. The top was coated using the polish applicator, allowed to dry for ~15 minutes, and the process was repeated on the bottom of the piece. Similarly, a pipette tip was used to apply a thin layer of FormLabs Clear resin to a separate piece on both the top and bottom, with both sides being UV cured for 10 minutes.

For the sanding method, 3M WetorDry Micron Graded Polishing Paper (ZONA, USA) was used. The piece was sanded on both sides starting at a 30 μm grit paper and followed by 15 μm, 9 μm, 3μm, and finally 1 μm grit. Moderate pressure was used to press the piece into the polishing paper in a circular motion to smooth the surface of the piece. Similarly, a generic 4-sided nail buffer (similar products, Walmart, USA) was used to evaluate the impact buffing could have on the printed piece.

Following post-processing, all pieces were imaged to determine optical clarity. All images were collected in the brightfield under transmitted light on a Zeiss AxioZoom macroscope (Carl Zeiss Microscopy). The intensity of light that passed through each piece was measured using Image J for three 1 × 1 inch² sections on each image. The average intensity and standard deviation were recorded for n = 3 regions per sample, with the background subtracted from each measurement individually. The relative transmittance, T, of each sample was calculated according to Equation 1,

where I is defined as the average mean gray intensity of the sample, and I₀ is defined as the mean gray intensity of a glass slide. Error was propagated using Equation 2,

where δ is the standard deviation.

3. Results and Discussion

3.1. Selecting a resin based on materials properties

Design of a successful 3D printed bioanalytical tool begins with selection of a suitable resin, a process that currently requires compromises. Ideally, the resins used to 3D print bioanalytical microfluidic devices would be compatible with all cell types, be able to produce milli- and microfluidic sized internal features without mechanical defects and meet imaging requirements of having low background fluorescence and high optical clarity when needed. There has yet to be a commercial resin, however, that integrates all of these ideal properties. Custom resin formulations may offer improved performance, at least for laboratories prepared to produce them consistently and tweak them for the intended instrument and application.^31,38,48 Nevertheless, focusing on enhancing one feature (e.g. cytocompatibility) usually results in compromising on another (e.g. print resolution). Because of this, it is useful to understand the key components of resins before choosing a material to work with.

We have found that many polymeric resins suitable for microfluidics fall into one of three categories based on the use case for which they were designed: biocompatible, optically clear, or high microscale resolution. Though specific resin components differ, resins in the same category often have similar material properties. Therefore, for this work, one test resin from each of these categories was chosen as a case study, and these are listed in Table 1 along with important intrinsic properties. FormLabs BioMed Clear V1 (FormLabs, USA) is a representative “biocompatible” resin with a USP Class VI biocompatibility rating, where it is approved for contact with live mucosal membranes and skin tissue for >30 days.⁴⁹ The FormLabs Clear V4 resin (FormLabs, USA) is representative of a material that offers increased optical clarity. MiiCraft BV007a Clear resin (CADworks3D, Canada) is a low-viscosity resin designed for high print resolution specifically for microfluidic devices. Researchers using one of the dozens of other available resins may use these properties to identify the extent to which it falls into one of these categories and thus predict performance of the resin for the intended use. When protected from ambient light, all resins tested were stable in open vats at room temperature without noticeable variation in volume or viscosity for at least two months.

Resins for SL printing are comprised of a photocrosslinkable polymer base, photoinitiators to initiate crosslinking, and (optionally) additives such as photoabsorbers, dyes, and plasticizers.^54–56 Common photoinitiators such as Irgacure compounds or lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP) are activated by UV (365 or 385 nm) or violet (405 nm) light.⁵⁵ While commercial photocrosslinkable resins for SL printing usually keep their exact compositions as a trade secret, the fundamental chemistry can be found in MSDS documentation. All three test resins in this work contained monomers and oligomers of an acrylate or methacrylate polymer base.^50,51,53 Many SL resins for microfluidics are fairly similar in their basic composition and can be used across different printers, especially with printers that allow for exposure setting adjustments.

Resins may be formulated with additives to achieve a large increase in print resolution or other desired properties. For example, addition of photoabsorbers reduces the effect of scattered light between layers and dramatically improves z-resolution.³² Addition of plasticizers lowers the viscosity of the pre-cured resin, which improves print resolution of hollow internal features by facilitating drainage of uncured resin from the feature during printing and cleaning steps. MiiCraft’s microfluidic BV007a resin, with the highest percent of additives, has a manufacturer-reported viscosity about 10-fold lower than that of the other listed resins (Table 1), and superior resolution for microscale channels. In highly viscous materials like the BioMed resin, undrained resin is easily retained inside the internal features, where it may be crosslinked by light that has been transmitted or scattered in excess, especially as subsequent layers of resin are cured above the hollow features.

On the other hand, photoinitiators and plasticizers, as well as acrylate monomers, can increase cytotoxicity with various cell types due to factors including oxidative stress, enzymatic inhibition, and lipophilic reactions with cell membranes.^56–59 Deliberately leaching these toxic components from the printed materials after post-curing can increase cytocompatibility.^29,30 However, we have found that in some cases it also decreases material stability, causing cracking or a decrease in material strength and flexibility as the plasticizers are removed. This problem was particularly prevalent with BV007a prints, which peeled apart when leached in PBS for longer than 48 hours, presumably due removal of plasticizers that were essential to the structural stability of the print. For this reason, selecting a more biocompatible material (e.g. with fewer leachable toxic additives) to begin with may better address this problem when working with sensitive cells.⁴⁸

As heat stability is an important factor for chips that will be subject to autoclave sanitation or extended cell culture, we tested the heat stability of each resin (Table 1). The FL BioMed material was autoclavable and also stable overnight at 120°C, allowing for thorough sterilization should any prints need to be reused or prepped for use with live biomaterials. In contrast, MC BV007a withstood mild sanitation procedures, e.g. alcohol rinses or UV sanitation, but high heat (>50°C) delaminated the material over time, although it was stable for 7-day incubation at 37°C when dry, or for 48-hrs in PBS. The FL Clear resin was found to withstand conditions heating conditions up to 70°C for 7 days dry or in solution and also mild sanitation procedures (i.e. UV curing, solvent rinses).

3.2. Physiochemical and environmental factors that influence print resolution

In addition to resin composition, the functionality and printability of a piece are also influenced by other factors that affect photopolymerization and thus the resolution and the mechanical integrity of the part (Fig.1).

3.2.1. Choosing wavelength and exposure settings to mitigate light scattering and bleed-through

Internal features such as enclosed microchannels are imperative to many 3D printed microfluidic devices. Most commercial resin materials yield features at the scale of millimeters or hundreds of micrometers on standard printers with 30-40 μm pixel size, and it is possible to improve the print resolution of internal features through strategic selection of resins, light sources and exposure settings, and number of repeat exposures.³² The central requirement is to avoid unintentional crosslinking of uncured resin inside the feature, which would lead to blocked features.

Light source wavelength, intensity, and exposure time have a large impact on resolution by modulating on the rate and extent of crosslinking.^31,32,60 Common light sources include a laser, LED, or UV lamp, and typically emit at 385 nm or 405 nm in commercial printers.⁶¹ Though most commercial resins can be printed at either wavelength, more efficient reactions are achieved by matching the wavelength of the light source with the excitation and absorbance peaks of the photoinitiators and photoabsorbers, respectively. Doing so decreases the exposure time and intensity required to achieve crosslinking and reduces unwanted scattered and transmitted light through print layers, as documented by Nordin and Wooley³² as well as Pontoni.⁶¹ To briefly demonstrate the impact of wavelength on resolution of internal features, test pieces containing six internal channels of decreasing square cross-section (0.2 – 1.2 mm side-length; Fig. 2A) were printed in FL Clear resin, using either a 405-nm or 385-nm printer (Fig. 2B vs 2C). The X,Y-resolution (effective pixel resolution) was similar between the two printers, at 30 and 40 μm for the 405- and 385-nm printers, respectively. In this resin, the crosslinking reaction was more efficient with the 385 nm light source, enabling reduced light dosage (shorter times and lower intensities; Table S1), which assisted in diminishing bleed-through light allowing uncured resin to drain more easily from channels. Consistent with this, channels were printable at sizes ~0.2 mm smaller with the 385 nm printer (Fig. 2C) versus the 405 nm source (Fig. 2B).

Figure 2. The resolution of internal features can be increased by making changes to the light source, resin viscosity, and print layer height. (A) A test piece had six internal channels, 9-mm long with 0.75 mm diameter inlets and varied channel cross-sections as noted. It was printed with (B-D) (i) 50 μm layer height or (ii) 100 μm layer height, as follows: FormLabs Clear resin was printed with a (B) 405 nm or (C) 385 nm light source; (D) for comparison, a low viscosity resin, BV007a, was printed at 385 nm. Other settings were left unchanged, with the exception of increasing exposure times slightly for some 100-μm prints (Table S1). Resolution was determined visually by observing the smallest channel cross-section that could be printed fully (pink arrows) or partially open (purple arrows).

An additional print setting that affects resolution is layer height, which sets the thickness and number of layers that must cured directly above hollow features, known as overhang layers. Each overhang layer, though required to close off the top of the feature, is a chance for light to unintentionally penetrate or scatter into the uncured resin that is trapped in the hollow space, potentially crosslinking it. Increasing the layer height increases layer thickness and reduces the number of overhang layers mitigating some bleed-through curing. This setting can be modified on most printers during the file slicing step when converting a design file into a printable file and works well for designs without strong diagonal features in the z-direction.⁶³ Using the FL Clear resin, doubling from 50-μm (Fig. 2i) to 100-μm layer height (Fig. 2ii) improved the print resolution of interior channels in the test piece to partially open the next smaller channel (an improvement of <0.2 mm, Fig. 2ii, purple arrows). Therefore, simply decreasing the number of overhang layers decreased the degree of overexposure or bleed-through light and improved print resolution, though not as much as changing the light source.

3.2.2. Lower viscosity improves drainage from internal channels

As noted in Section 3.1, the need to drain uncured resin out of hollow features during both printing and cleaning means that resin viscosity has a major impact on the resolution of internal features. To demonstrate this, we compared the resolution of FL Clear resin (Fig. 2C) to MC BV007a (Fig. 2D), which have viscosities of ~900 mPa s and ~100 mPa s, respectively (Table 1), using the 385 nm light source. The FL Clear resin retained uncured resin in the channels during printing (visible when the device was removed from the printer), and produced open channels only down to 0.6 mm under these conditions. In contrast, no residual BV007a resin was observed prior to rinsing the channels, and the print yielded a resolution of 0.2 mm, the smallest size tested, thus confirming the significant benefit of low viscosity for channel resolution.

In summary, with its ability to print efficiently at 385 nm and to drain easily with low viscosity, BV007a provided the best print resolution, with 0.2-mm channels printing cleanly at a 50-μm layer height. Printing with the Clear resin, however, achieved nearly 0.4 mm channels if used with a 385 nm light source and 100-μm layer height. While the specifics of printability will change for each design, we found that the light source, material viscosity, and layer height each provide opportunities to increase print resolution of interior microchannels.

Apparatus Used

Clear Microfluidic Resin

Learn More

PR110
3D Printer

Legacy

3.2.3. Influences of print environment on photopolymerization

Finally, the humidity and temperature within or surrounding the printer can also impact how well the photopolymerization reactions take place. Many resins are recommended to be printed at relatively low humidity (20-40%), as photopolymers are often hygroscopic.⁶⁴ We have observed that when the humidity increases (averaging 45% in the summer in our laboratory, sometimes higher than 50-60%; versus ~30-35% in winter), print failures became more common: print layers delaminated or base layers did not remain attached to the baseplate. Keeping printers and resins in an environment with regulated humidity should be considered, e.g. dehumidification. We also found that increasing the power of the light source by 5-10% partially compensated for humidity increases.

Ambient temperature can also influence how well a material prints as it directly impacts the viscosity of the material and reaction times. This topic was explored in depth by Steyrer, et al. in an evaluation of hot vs. room temperature lithography.⁶⁵ In general, many resins tend to produce better print resolution in a warmer environment (above 20°C) due to lower viscosity and more efficient polymerization.^65,66 For this reason, some printers allow for control over vat temperature. For printers without such control, we have found that decreases in ambient temperature (e.g. from 23°C to 20°C) can be compensated by increasing exposure time, usually by a few seconds depending on the resin and other factors, to overcome the decreased reaction efficiency. Care should be taken, though, as increasing exposure times may lead to overcuring.⁶⁵

3.3. Print and design considerations for common microfluidic features

In this section, we discuss design principles for troubleshooting common print failures and improving the resolution of 3D printed microfluidic channels, along with the role of light exposure and printer settings in controlling print quality.

3.3.1. Selecting a print orientation

The build orientation and design of a part impact how the print experiences gravitational strain and other mechanical stress during printing. Print orientation techniques are a common topic for 3D printing blogs and education resources.^67–69 In general, considerations for choosing a suitable build orientation include baseplate adhesion, location of part functionalities, the use of build supports, and fabrication time. Baseplate adhesion is usually strongest when the largest surface area of a design can be placed flush against the baseplate (Fig. 3A–D, i), which increases the overall stability of piece during the printing process. A slightly rough surface allows prints to adhere better to the plate while printing with lower base layer curing times, although too much adhesion makes removal difficult and can lead to breakages. Location of part features also influence how to orient a print. In this example, orienting the reservoir sideways (ii, iv) caused distortion of its walls or damage from supports, and inhibited resin drainage from the channel. Surface roughness or clarity were also influenced by the direction in which the layers were built: clarity and smoothness are maximized by orienting so that the face of the chip is printed in one layer (i, iii), and reduced by gravitational strain (Fig. 3C, iii) or too many print supports (Fig. 3D, iii). The latter can also contribute to overexposure and light scattering, causing features to fill in (Fig. 3D, ii-iv). Finally, as each additional layer adds cure time to the print, and supports add post-processing time, it is useful to minimize the use of supports while printing if possible, and to select print orientations with fewer layers to create a more efficient printing process.

Figure 3. Orientation of prints on a build plate impacts the print resolution and optical clarity. A microfluidic design containing a reservoir, channel, and chamber is shown in the MiiCraft Utility slicing software in four orientations (A) without supports and (B) with supports: (i) base-down, (ii) vertical, (iii) top-down, and (iv) horizontal. Blue grid represents the base plate. Supports were added to all areas using default settings in the slicing software. (C, D) Photos of pieces printed in BV007a resin with the MiiCraft Ultra 50 printer, (C) printed without supports, and (D) printed with supports, with 2 minutes of support removal. Surface roughness and clarity variation is evident between the 4 orientations. Defects were noted as follows: In C: (i) no defects, (ii) overhang drooping, (iii) delamination (purple) and stress fracturing (pink), (iv) blocked channel (purple), overhang drooping (pink); in D: (i) no defects, (ii) stress fracturing, (iii) damage from supports (purple) and filled chamber (pink), (iv) filled channel (purple) and support damage (pink).

3.3.2. Reducing mechanical strain in wells, cups, or ports

In microfluidic design, particularly for bioassays, wells or cup-like features are often used as reservoirs or ports. If not properly designed, these wells often fall victim to print failure.^70–72 Like other photocrosslinkable polymers, many SL resins shrink 1-3% by volume upon crosslinking which may induce mechanical strain.^55,71,73 If the walls of a print are thinner at some points and thicker at others, the thicker regions experience greater shrinkage and may generate defects.⁷¹ Weak structural points may form when thin walls, sharp corners, and sharp edges (~90° angles) are used in a design, and these may not hold up well in the printing process.^74,75 Wells or cup-like features, for example, may experience “cupping,” which occurs when hollow features become damaged during printing due to the formation of a pressure differential (Fig. 4). This effect is due to a low pressure region, or suction, formed within the feature as the print is peeled or pulled away from the Teflon sheet after each layer is exposed, leading to cracks and/or holes at weaker structural points in the design as it caves inward under the surrounding pressure (Fig. 4B).⁷⁰ Such defects may cause leaking when the printed well is later filled with fluid.

Figure 4. Schematic of “cupping” damage during the printing of a hollow, cup-like feature. (A) During printing, a UV light source (LED) forms a new crosslinked layer of resin flush against the Teflon sheet. This design is an inverted bowl shape; supports are not shown for clarity. (B) After a layer is finished printing, the print is peeled away from the Teflon sheet, e.g. by pulling the vat down and/or the baseplate up. This process creates a region of suction within the hollow cup feature; the surrounding pressure, now higher than the pressure within piece, may form a stress fracture on the print.

As a demonstration, a hollow well printed in a square base with 90° square angles and thin walls broke routinely under the pressure build up from cupping (Fig. 5A). We tested the impact of various strategies to reduce mechanical strain in the design, drawing upon engineering principles.^74,75 The square exterior, with its varied wall thickness around the radially symmetric well, contributes to an uneven stress distribution during resin shrinkage. Increasing the thickness of the base and wall and filleting the connecting edge at the base of the well reduced the risk of cracking the base of the print, but not the appearance of small holes at the base of the well (Fig. 5B). Making the thickness of the walls more uniform around the well-like feature, by rounding the exterior corners of the feature either partially (Fig. 5C) or fully (Fig. 5D) reduced the strain unequal shrinkage as expected, with full rounding producing a well feature with no holes or other leaks.

Figure 5. Strengthening the design around well-like features decreases the impacts of cupping and resin shrinkage. (i) Illustrated computer-aid designs, and (ii) photos of the top and bottom views of the corresponding print. All pieces were printed in FormLabs Clear resin. Well A had thin walls, thin base, and strain from 90□ connections at the bases and the sides. Well B had thicker surrounding walls and a thicker, filleted base, but retained 90□ outer corners. Wells C and D further reduced the strain by rounding out the external edges. All pieces were qualitatively evaluated, with the absence of cracks (arrowheads) and pinholes (shown by arrows) indicating a strong design.

3.3.3. Improving print resolution of interior channels via device design

Similar to the effect of layer height during printing, we predicted that changing the device design itself to reduce the number of UV exposures over the channel or otherwise facilitate drainage of uncured resin would improve resolution. To demonstrate, we again selected the FL Clear resin due to its high viscosity and transparency, properties that result in frequent bleed-through curing. The number of overhanging layers was controlled by repositioning the channels in the z-direction (Fig. 6), using a fixed layer height of 50 μm. Square channels that were printed near the base of the print, with 1.5 mm thick overhangs and thus 30 overhang layers, printed cleanly only down to 1.0 mm width (Fig. 6B). Decreasing the overhang thickness to 0.5 mm (10 overhang layers) improved the resolution by ~0.2 mm (Fig. 6C), consistent with the prediction. Additionally, adding drainage holes further improved print resolution of long channels that otherwise did not drain (Figure S1).

Figure 6. Reducing the number of overhang layers in the chip design improved the resolution of internal features. (A) Schematic of the test piece with six internal channels, in (i) top view and (ii) side view. (B,C) Channels were printed with (B) 1.5 mm or (C) 0.5 mm overhang thickness, and imaged from (i) top and (ii) side. The square channel is traced with dashed outlines; the features visible above the channel in the side view are inlets and outlets. All pieces were printed in FL Clear resin with a 405 nm light source (settings in Table S1). Resolution was determined visually by observing the smallest channel cross-section that could be printed fully (pink arrows).

3.4. Cytocompatibility of resins with primary cells

Within this section, we discuss actionable steps towards improving resin cytocompatibility, as it is typically the most challenging issue for 3D printing microfluidics to be used with live cultures and tissues. Many recent publications have addressed this issue, with most solutions focused on preventing potentially toxic resin components from reacting with cell and tissue cultures. Methods include coating the prints to reduce direct contact with cells;²⁴ minimizing unbound toxins via autoclaving, overcuring, or pre-leaching toxic substances;²⁹ or producing resins with more biocompatible photoabsorbers.^38,76,77 Leaching in particular is convenient and consistent, especially for treating internal print features that may not be directly accessible to over-curing. Therefore, here we first compared biocompatibility across the representative commercial resins, with and without a standardized, leaching-based post-treatment in heated saline (see Methods; Fig. 7A).^18,29 Primary, naïve murine splenocytes were chosen for the viability tests as they are typically more sensitive to small changes in their culture environment than hardier, immortalized cell lines.^78,79

Figure 7. Viability of primary murine splenocytes in contact with 3D printed materials. Primary splenocytes from male and female mice (Nmice = 2 per experiment) were evaluated by flow cytometry after live/dead staining with Calcein-AM/7AAD. (A) Cell viability after 4 hr of direct physical contact with PBS+incubation treated (T) or non-treated (N) resins compared with well plate control (C). Treated BioMed and Clear prints maintained relatively high viability compared to the well plate, while BV007 did not. Bars show mean ± SD. One-way ANOVA per data set, ns > 0.06, *p > 0.01, **p = 0.01, ****p < 0.0001. (B) Multiple post-treatment methods (PBS, incubation at 37 °C, PBS+incubation, and autoclavation) were evaluated using BioMed resin and culturing for 4 hours with direct contact. Bars show mean ± SD. One-way ANOVA, ns > 0.2, *p > 0.02. ****p < 0.0001. (C) Viability of direct contact (i.e. culture with resin) and indirect contact (i.e. culture with resin leachate) of cells with treated BioMed prints at 45 min, 4 hours, and 24 hours showed a decrease in viability over time. Values show mean ± SD. One-way ANOVA at final time point, ****p < 0.0001. — Figure 7. Viability of primary murine splenocytes in contact with 3D printed materials. Primary splenocytes from male and female mice (Nmice = 2 per experiment) were evaluated by flow cytometry after live/dead staining with Calcein-AM/7AAD. (A) Cell viability after 4 hr of direct physical contact with PBS+incubation treated (T) or non-treated (N) resins compared with well plate control (C). Treated BioMed and Clear prints maintained relatively high viability compared to the well plate, while BV007 did not. Bars show mean ± SD. One-way ANOVA per data set, ns > 0.06, *p > 0.01, **p = 0.01, ****p 0.2, *p > 0.02. ****p < 0.0001. (C) Viability of direct contact (i.e. culture with resin) and indirect contact (i.e. culture with resin leachate) of cells with treated BioMed prints at 45 min, 4 hours, and 24 hours showed a decrease in viability over time. Values show mean ± SD. One-way ANOVA at final time point, ****p < 0.0001.

We found that for experiments under 4 hours, overnight heat and saline leaching treatment was sufficient to increase viability of primary murine splenocytes in contact with 3D printed materials (Fig. 7A). Splenocytes cultured in direct physical contact with either of the treated FL resins retained viability high enough (>60%) after 4 hr to enable short on-chip experiments, whereas untreated materials resulted in lower viability on average. For the MC BV007a resin, the treatment did improve viability over the non-treated resin, but the resin was still largely cytotoxic for these cells at 4 hours. The higher cytotoxicity is consistent with the greater quantity of potentially toxic additives in BV007a compared to other resins (10-15%, Table 1). Furthermore, BV007a had lower heat stability and could not be leached at the same temperature as the FL resins without mechanical damage (Section 3.1). Therefore, for short experiments with primary cells in suspension, we concluded that the Biomed or Clear resins are more suitable than BV007a and should be pre-treated to minimize release of toxic components into the culture.

Next, we compared a number of common leaching treatments head-to-head: heat (37°C incubation, 24 hr), saline (sterile PBS soak at room temperature, 24 hr), a combination of the two (sterile PBS soak at 37°C, 24 hr), or autoclaving (gravity, 120°C, 30 min) to leach unbound toxins out of solid prints (Fig. 7B). Using the BioMed resin, we found all overnight saline (PBS) and incubation (37°C heat) treatments increased splenocyte viability at 4 hours, with non-significant differences between the live plate control and treated materials. Autoclaving was slightly less effective, and though this treatment takes less time (~45 minutes), it is not recommended, at least for this resin.

Finally, as many experiments require longer term culture than just 4 hr, we tested splenocyte viability over time when cultured in direct physical contact with treated FL BioMed resin or in “indirect contact,” i.e. cultured in contact with media that was pre-conditioned by incubation with the resin. Direct contact with resin is likely to occur in a fully integrated bioanalytical microchip with on-chip cell culture, while indirect contact may occur when microdevices are used to prepare media or drug solutions (e.g. mixers or droplet microfluidics) or to deliver media components to cultures (e.g. flow channels or bubble traps). We again found that splenocyte viability was high for both contact conditions within 0-4 hr, but decreased overnight (24 hrs) for both contact conditions compared to plate controls (Fig. 7B). Others have reported longer culture times for hardier cell lines.^{18,25,27–29,33,38,80,81} This result indicates that at least for primary murine splenocytes and perhaps for other fragile cells, more work is needed to identify best practices for post-print treatments and/or more biocompatible resin formulations.

3.5. Optical components of resins

Microfluidics is frequently integrated with on-chip imaging. Autofluorescence is a potential limitation of polymeric chips, whereas PDMS is often praised for its optical compatibility. Therefore, we quantified the autofluorescence of each of the representative resins in four standard fluorescence channels. In the Cy5 and Rhodamine channels, all three resins showed relatively low autofluorescence and were comparable to PDMS (Fig. 8). However, UV excitation (DAPI channel) elicited high levels of autofluorescence from the two FormLabs resins. The Clear resin also showed moderate autofluorescence in the GFP channel. Autofluorescence at short wavelengths is consistent with the use of photoabsorbers intended to absorb at short wavelengths.⁷⁶ On the other hand, MC BV007a had negligible autofluorescence in all channels, similar to PDMS. Thus, optical compatibility in the intended fluorescence channels should inform material choice for microfluidics devices.

Figure 8. Autofluorescence depended on resin composition and fluorescence filter set. BioMed, Clear, and BV007a prints were evaluated in comparison to PDMS for autofluorescence in Cy5, Rhodamine, EGFP, and DAPI fluorescent channels. Background subtracted mean grey values were analyzed with ImageJ and used to determine the fluorescent intensity of each material. Saturation values were at 65,000 AU. Bars show mean ± SD, N=3 intensity measurements per print. Two-way ANOVA, comparisons shown between resin results versus PDMS for each respective fluorescent channel, ****p<0.0001.

Optical clarity (transparency) is also important for microscopy and imaging. Various approaches for improving this property have been suggested in published papers as well as on vendor and hobbyist websites (r\3Dprinting, All3DP, etc.).^37–39 Here we compared 5 of these methods head to head, using the FL Clear resin as a base, to determine which would be best for increasing the transparency of a printed piece (Fig. 9). All methods were compared to a glass slide as a positive control. Pieces that were printed flush against the rough base plate (i.e. without the use of supports or rafts) came off the base plate slightly cloudy in appearance (Fig. 9VI). This cloudiness was intensified when attempting to buff both the base end and the vat end of the print with a typical nail file (Fig. 9VII). Since the nail file did not have a fine sanding grain, it ended up doing more harm than good by producing new scratches on the surface of the print.

Figure 9. Optical clarity of clear resins was enhanced with post-treatments that achieved material transparency similar to glass. Round prints of FormLabs Clear resin were printed with the MiiCraft Ultra 50 405 printer and post-treated as listed in the legend, labels II -VII. (A) The test pieces were positioned over the Pompano Lab logo for qualitative, visual comparison (captured with phone camera). (B) Transmittance was evaluated using an upright microscope. Relative transmittance was compared to the average light transmittance through a 0.17 mm thick glass slide (I). Bars show mean ± standard deviation, n = 3. One-way ANOVA, Tukey’s post-hoc test; *p = 0.0284, **p = 0.0067, and ns > 0.05. (C) To show reproducibility for different resins, technique II was applied to similar round prints in FL BioMed, FL Clear, and MC BV007a resins and evaluated similarly to panel B. All samples were statistically similar compared to the glass slide control. Two-way ANOVA, Tukey’s multiple comparison test; ***p=0.0003, ****p0.8.

Several methods proved successful in improving optical clarity. The quickest method that produced glass-like clarity was the nail polish coating of both the base and vat sides of the print (Fig. 9II). The nail polish only took a few seconds to apply and approximately 10 minutes to dry on each side, which introduced a challenge to keep dust out of the polish while drying on an open countertop. Sanding the piece with micron grain sandpaper avoided the issue of dust and the pieces emerged smooth (Fig. 9III) instead of scratched like the nail-filed (“buffed”) piece (Fig. 9VII). Sanding the printed part on at least the baseplate-side is especially recommended if it will not disrupt any print features, as it more easily maintained a uniform surface than the other methods. Printing directly on glass³⁸ (Fig. 9IV) had similar impacts to sanding and is recommended for smaller prints that would be easier to remove from a glass slide. This approach requires some caution, though, as attaching glass to a baseplate can cause increased wear and tear to a printing vat. It also requires resetting the initial layer height so that the printer does not lower the baseplate too far into the vat. Resin coating (Fig. 9V) was also helpful for increasing transparency, but it was difficult to achieve without overcuring the rest of the print or under-curing the additional coat, which was also prone to capturing dust. Relative to the glass slide control, most of the post-treated 3D prints were determined to have acceptable transparency. The nail polish technique (Fig. 9II) was tested on all three representative resins and consistently improved the relative transparency of each resin (Fig. 9C). We expect that each of methods I-V will provide improved optical clarity to most transparent or semi-transparent resins.

4. Conclusion

In using 3D printing for production of microfluidic devices, compromises and strategic design choices are often required to best match the material and design to the required experiment. After identifying priorities based on the planned experiments, a resin should be chosen that best fits the requirements of print resolution, mechanical stability, cytocompatibility, and optical compatibility, informed by a foundational understanding of material components. If needed, printer settings and device designs can be modified to increase the integrity of printed parts and resolution of interior channels, and post-treatment methods can be used to increase the cytocompatibility and optical clarity of a printed piece. Print stability can be improved by reducing mechanical stress in the design of a piece, and internal feature resolution can be increased by ensuring adequate resin drainage and minimizing the photoexposure of trapped resin, e.g. by reducing the number of overhanging layers. Viability can be improved upon by leaching toxins out of prints prior to application with cells, though there is still a need for more biocompatible options, especially for sensitive primary cells. Optical clarity of parts printed with clear resins can be improved via polishing methods to achieve glass-like transparency. In the future, resins that are high-resolution, cytocompatible, and optically clear will certainly be an area of continued commercial development, and promising PEG-DA based resin formulas have been reported that can be made in the laboratory.^38,76,77 Meanwhile, the considerations and best practices recorded here can help researchers begin to integrate SL 3D printing fabrication with commercially available products into their microfluidics research. We envision that this guide and its head-to-head comparison of conditions will help streamline the fabrication workflow for researchers who are new to 3D printing within the biomicrofluidic community.

Rapid Softlithography Using 3D-Printed Molds

March 12, 2026August 12, 2019 by CadWorks3DAdmin

Rapid Softlithography Using 3D-Printed Molds

by Sajad Razavi Bazaz, Navid Kashaninejad, Shohreh Azadi, Kamal Patel, Mohsen Asadnia, Dayong Jin and Majid Ebrahimi Warkiani

Abstract: Polydimethylsiloxane (PDMS) is a long-standing material of significant interest in microfluidics due to its unique features. As such, rapid prototyping of PDMS-based microchannels is of great interest. The most prevalent and conventional method for fabrication of PDMS-based microchips relies on softlithography, the main drawback of which is the preparation of a master mold, which is costly and time-consuming. To prevent the attachment of PDMS to the master mold, silanization is necessary, which can be detrimental for cellular studies. Additionally, using coating the mold with a cell-compatible surfactant adds extra preprocessing time. Recent advances in 3D printing have shown great promise in expediting microfabrication. Nevertheless, current 3D printing techniques are sub-optimal for PDMS softlithography. The feasibility of producing master molds suitable for rapid softlithography is demonstrated using a newly developed 3D-printing resin. Moreover, the utility of this technique is showcased for a number of widely used applications, such as concentration gradient generation, particle separation, cell culture (to show biocompatibility of the process), and fluid mixing. This can open new opportunities for biologists and scientists with minimum knowledge of microfabrication to build functional microfluidic devices for their basic and applied research.

Keywords: 3D-printed molds, 3D-printing, microfluidic resin, microfluidics, soft lithography

We kindly thank the researchers at the University of Technology Sydney and Macquarie University for this collaboration, and for sharing the results obtained with their system.

1. Introduction

In recent years, there has been a new surge of interest in 3D printing, which is defined as building successive layers of materials to form a desired object.^[1,2] The interest in 3D printing methods is twofold. First, the advent of 3D printing has triggered the creation of numerous intricate designs, whether in micro or macro scale, often implausible through conventional fabrication methods. Second, 3D printing enables quick evaluation of ideated solutions, often within the same day. Feature-wise selection of printing parameters and multistep printing processes enable users to pay extra attention to the tiny details of their objects.^[3] In addition, material specifications (e.g., Young modulus or transparency) can be adjusted based on the printing method. It is estimated that the market size of 3D printing will triple in the next half-decade, growing from 7.3 billion dollars in 2017 to 23 billion dollars by 2022.^[4] As structures manufactured by 3D printing methods can be in the range of micrometers to centimeters, a new challenge emerges for microfabrication.^[5]

The miniaturization of high-cost, resource demanding, and time-consuming lab processes into a high-efficient, multifunctionalized, and integrated microchip has been considered as a revolution across many fields of science.^[6] Microfluidics, the commercial name for this revolution, is ubiquitous in fluid mechanics, reagent mixture, cell biology, particle and cell separation, metabolomics and proteomics, forensic, and genetic analysis.^[7,8] Microfluidic devices enjoy the proficiency of low reagent consumption, parallelization, portability, integrated several biological assays, small footprint, accurate measurement, and live feedback.^[9] Compared to macroscale fluid handling, microfluidics provides end-users with an economical and ready-to-use microchip with faster reaction time and prompt analysis.^[10,11]

There is a growing body of literature that recognizes the significance of lithography in the fabrication of PDMS-based microchannels. However, lithography is limited in its ability to fabricate nonstraight microchannels. For instance, for vascular behavior imitation, fabrication of 3D complex vessel branches is mandatory.^[12] Moreover, there are major limitations in the fabrication of angular designs, such as a microchannel with a trapezoidal cross-section.^[13] Furthermore, nonplanar structures as well as 2D and 3D nanolithography always introduce more complexity to the fabrication process.^[14] In addition, advanced equipment and an adroit operator are essential for microfabrication processes, especially when the surface coating of the device is of interest.^[15] For these reasons, research groups have tried to provide alternative methods for the fabrication of molds used in softlithography processes.^[16] One such alternative is the use of 3D printing technology for the fabrication of softlithography molds. Among all 3D printing methods, stereolithography apparatus (SLA) and digital light processing (DLP) offer great advantages, making them ideal candidates for microfluidics and biomedical applications.^[17] However, one of the limitations of 3D printed SLA/DLP master molds for softlithography is the requirement for tedious pretreatments prior to PDMS casting. The pretreatment of the resin is necessary to ensure the complete curing of the PDMS in contact with the resin. Otherwise, the surface of the PDMS replica in contact with the resin cannot be polymerized due to the presence of residual catalyst and monomers, and its transparency would be also compromised.^[18] It has been observed that the effects of pretreating the master mold are more significant in channels with smaller feature sizes,^[19] and, in the case of relatively larger 3D printed parts, this challenge is not significant.^[20] To address this issue, many researchers have proposed various pretreatment protocols to treat the 3D printed master mold before PDMS casting.^{[18,19,21–24]} As one of the first attempts, Comina et al. proposed to cover the 3D printed template with a specialized ink via airbrushing.^[21] However, the effectiveness of that method depended largely on the thickness of the ink. Four procedures are commonly used among other proposed postprinted protocols: 1) UV curing; 2) surface cleaning (e.g., ethanol sonification and soaking); 3) preheating; 4) surface silanization. Waheed et al. introduced an efficient yet time-consuming pretreatment protocol for PDMS softlithography.^[24] The postprocessing included a 5 min UV treatment followed by 6 h of soaking in an ethanol bath. Following the air plasma treatment for 1 min, the surface of the 3D printed template was silanized by perfluorooctyl triethoxysilane for 3 h.

Nevertheless, there is still no consensus about the optimum protocol for treating 3D printed templates for PDMS casting. In addition, the proposed protocols are time-consuming, laborintensive, and lacking reproducibility. Furthermore, the treatment parameters, such as UV curing time, preheating temperature, and duration, seem to be a function of the feature size; thus, differ from one experiment to another.^[24] Also, preheating in particular is a common step in many procedures and often induces high levels of material strain, resulting in the formation of cracks within microstructures.^[18,25] Most importantly, surface silanization of the 3D printed templates is essential to ensure the PDMS peels off, correctly. Some silanizing agents such as perfluorooctyl triethoxysilane are cytotoxic and are not suitable for biological applications. Thus, development of a resin suitable for master mold fabrication will reduce all these time-consuming steps.

To address the aforementioned issues, herein, we optimize the use of a new resin developed by Creative CADworks (CCW Master Mold for PDMS devices) (i.e., made of methacrylated oligomers and monomers) for the fabrication of master molds directly by the DLP 3D printing method. We show that the 3D printed templates obtained using this resin can be immediately casted with PDMS without any pretreatment or surface modification. By way of explanation, the process of master mold design to microchip fabrication has been reduced from a time frame of several days (for a conventional softlithography process) to less than 5 h. In order to showcase the functionality of this resin, four different microfluidic devices have been developed. Each device represents a specific application, including separation, micromixing, concentration gradient generation, and cell culturing. The surface of the PDMS replica obtained from the 3D printed mold is also evaluated to investigate the bonding quality of PDMS.

Apparatus Used

Master Mold for PDMS

Learn More

Ultra 50
3D Printer

Legacy

2. Results and Discussions

2.1. PDMS Characterization

It is well-known that the quality of the PDMS casted on the mold can affect the whole performance of the microfluidic device.^[26] Hence, its quality must be analyzed before use. After fabricating the 3D printed molds and removal of any residual resin, PDMS was casted on the master molds. For the sake of comparison, two different molds were fabricated, one with a conventional DLP resin and the other with the newly developed microfluidic resin. The main challenge with conventional DLP resin is that due to the presence of unreacted monomers, complete polymerization of PDMS cannot occur, resulting in the presence of residual material on both the PDMS and the mold, as shown in Figure 1A. The comparison between the mold fabricated via conventional resin and the microfluidic resin reveals that these two molds have identical surface roughness, and the smallest channel height for the fabrication of molds can be achieved with a thickness layer of 30 µm. However, for this thickness layer, the curing time of each layer for the newly developed resin is 6.5 s, which is more than the conventional one which is 1.3–1.5 s; as more time must be devoted to the methacrylated resins to be completely polymerized and cured. All in all, the fabrication time for both molds took less than an hour which is much faster than other methods. Also, the inset in Figure 2A shows the contact angle of the 3D printed molds. The contact angle measurement reveals that both surfaces are hydrophilic; however, the microfluidic resin is slightly more hydrophilic than the conventional one.

By substituting the acrylate components with methacrylated monomers and oligomers (Figure 1B), we are able to create a clean temporary binding site between the PDMS and the 3D printed master mold. To demonstrate this, we applied both of the conventional DLP resin and the newly proposed microfluidic resin to a single design and investigated the boding properties of PDMS. Both molds were subjected to the same experimental procedure.

Figure 1. A) Schematic illustration of how acrylated DLP resins impact the surface finish of casted PDMS pieces. Residual catalysts and monomers present at the interface between the resin and PDMS impede the polymerization of PDMS components, leaving behind residual material. B) Demonstrating the improved performance of methacrylated resin over conventional acrylates in providing a smooth surface finish with no residual material. This is due to a lack of unreacted monomers and oligomers impeding the complete polymerization of PDMS.

It has been proven that in UV-cured systems, cracks developed as a result of shrinkage forces between and after curing.^[27,28] In the methacrylated systems, this shrinkage has an inverse relation to the initial viscosity.^[28,29] As the modified resin is more viscous than the conventional ones, the chance of cracks appearing and propagating reduced significantly during the curing process. As such, the mold fabricated via the microfluidic resin has better stability and a very smooth surface compared to those fabricated by conventional resin. As Figure 2A indicates, in the conventional DLP resin, PDMS surfaces in contact with the surface of the resin were not properly cured, and uncured PDMS layers remained on both surfaces. It can be clearly seen that the casted PDMS fails to adopt the pattern of the mold, thoroughly. In addition, during the detachment of PDMS from the mold, PDMS tends to stick to the resin, confirming that the surface of the conventional DLP resin is not appropriate for PDMS casting. By analyzing the materials constituting the conventional DLP resin, it is believed that this problem is related to the chemical composition of the resin. We hypothesized that the remaining catalyst and monomers on the surface of the printed mold disrupt the complete polymerization of a thin layer of PDMS in contact with the mold. This can be clearly seen upon the removal of the PDMS replica from the mold (Figure 2A). As such, the “acrylate group” in the resin’s chemistry is not a suitable choice for PDMS casting; this has urged different scientists to explore time-consuming strategies for the surface treatment of DLP printed molds. Through extensive research conducted by Creative CADworks, a new resin which contains methacrylated monomers and oligomers has been developed. Casted PDMS does not react with the methacrylated monomers because the surface of the mold is free of residual monomer units that may impede PDMS polymerization. As Figure 2B illustrates, once a blade cuts through the PDMS layer down to the mold, the PDMS replica detaches easily. The operation of each device and the quality of bonding were also analyzed for a wide range of flow rates (to check the simulations results of surface roughness and bonding quality, see Section 2.2) with the experimental setup shown in Figure 2C. The results, as shown in Figure 2D, confirmed that there was no leakage observed between flow rates ranging from 0.1 to 5 mL min⁻¹, which indicates that the proposed method for fabricating PDMS-based microdevice is an ideal candidate for a variety of applications.

Figure 2. PDMS casting process in A) conventional DLP resin and B) microfluidic resin. The insets depict the contact angles on the surface of molds. In conventional resin, PDMS in touch with the surface of the mold cannot provide a temporary bonding, and the surface of the PDMS cannot replicate the pattern used in resin. In microfluidic resin, as soon as the blade reaches the surface of the mold, PDMS start to detach from the surface, and it can easily peel-off. The mold after PDMS casting in microfluidic resin clarifies that there is not any residual of PDMS on its surface, while in conventional DLP resin, residuals are on the surface. C) Experimental setup used in these series of experiments is illustrated. D) No leakage was seen during the experiments after bonding of PDMS by plasma surface treatment method.

2.2. Simulation Studies of Surface Characterization

Here, the effects of surface roughness on the velocity and shear rate distribution along the length of microchannel were investigated through simulation study by COMSOL 5.3a. For a smooth surface, S_a was set as 0.3 µm, and for a rough surface, S_a was assigned to be 1 µm. Different flow rates of 0.1, 1, 1.7, and 3 mL min⁻¹ were tested to investigate the shear rates present in the devices. Figure 3A shows velocity and shear rate distribution along the length of the smooth microchannel (S_a = 0.3 µm). The two insets (Figure 3AI,AII) depict shear rate distribution across the bottom surface of the microchannel at flow rates of 0.1 and 3 mL min⁻¹; and by increasing the flow rate from 0.1 to 3 mL min⁻¹, the order of the shear rate increased 100 times. Furthermore, the shear rate distribution illustrates that in the middle of the microchannel, due to the high shear rate, there is a higher probability for the quality of surface bonding to be disrupted than at the edge of the microchannel. Moreover, shear rate distribution 50 µm from the inlet was investigated at heights of 2, 5, 10, and 15 µm (half of the channel height) from the bottom surface for four flow rates of 0.1, 1, 1.7, and 3 mL min⁻¹ (Figure 3AIII–AVI). The trend observed illustrates that the shear rate is focused halfway across the width of the channel at the height of 2 µm; as the height increases, the focus is drawn away from the center of the channel.

Figure 3. Velocity and shear rate distribution along the length of microchannel for A) Sa = 0.3 µm and B) Sa = 1 µm. Part I and II of each section (i.e., A and B) stand for the shear rate distribution at the bottom of the microchannel for velocity of 0.1 and 3 mL min−1 (black arrows are first principal curvature of surface). In the smooth channel, the peak of shear rate focuses at the center of the microchannel, where, in the other one, it relocates to the edges of microchannel. Shear shear distribution along the width of microchannel at 2, 5, 10, and 15 µm for velocities of 0.1, 1, 1.7, and 3 mL min−1 are illustrated by parts III to VI, respectively. It shows that in rough microchannel shear rate is uneven.

For the rough channel, although the applied flow rates were the same as the smooth channel, the shear rate distribution was noticeably greater. There is more variation in the bottom surface of the microchannel, (identified by the black arrow) when S_a = 1 µm compared to 0.3 µm. The bottom layer of the shear rate distribution also illustrates that the shear rate focuses more on the edges of the microchannel rather than in the middle (compared to the smooth surface). Thus, the probability of bonding disruption will be relocated to the edge of the channel instead of the middle of the channel. Figure 3BIII–BVI show the flow rates of 0.1, 1, 1.7, and 3 mL min⁻¹ for S_a = 1 µm at a point 80 µm after the inlet. These figures demonstrate that the shear rate distribution is uneven along the width of the microchannel. Also, the shear rate values for S_a = 1 µm are higher than those for S_a = 0.3 µm for all heights and all magnitudes of velocity. Thus, surface roughness in microfluidic devices must be small enough so as to not impact upon the performance of the device, and the bonding quality as well as measurement performed within a microchannel were not influenced by the surface roughness of the microchannel.

2.3. Microfluidic Devices for Liquid Handling

Particle sorting and separation have become important processes within diagnostics and biological sample handling.^[30] The unique properties of fluids at the microscale can be exploited to provide a perfect platform for handling fluid samples. For instance, fluid inertia is often used for focusing randomly dispersed particles into a particular location for the aim of collection or separation.^[31,32] Spiral microchannels require relatively high flow rates which needs strong permanent bonding. In order to achieve strong bonding, the surface of PDMS layers must be ultrasmooth to facilitate plasma bonding of the PDMS and withstand the high shear stress generated by the input velocity.

Figure 4A shows the whole-chip layout of a spiral microchannel used in this study. Surface characterization depicts that the S_a is around 0.3 while R_a is approximately 0.2. As R_a is evaluated randomly in a line, it is reasonable that its value be less than that of S_a which covers the whole selected area. The function of the spiral microchip was examined with 15 µm fluorescent particles to verify the bonding and blocking of the microchannel. Flow rates from 0.5 to 3 mL min⁻¹ (with an increment of 0.5 mL min⁻¹) were tested to examine the bonding between the microchannel and its base, as shown in Figure 4B. It was illustrated that the flow behavior for these particles was the same as those reported in literature, where flow rates below 1.5 mL min⁻¹ dispersed particles at the inner wall. However, at flow rates more than 1.7 mL min⁻¹ , particles were focused at the outer wall and could be easily isolated for further use.^[33]

Micromixers have become an essential tool in the preliminary stages of many lab-on-a-chip processes. Previously, by gaining the efficiency of proximity field nanopatterning and 3D nanolithography, Jeon et al. proposed a micromixer by implanting 3D nanostructures within the channel to enhance mixing efficiency, especially at low Re where diffusion mixing is dominant.^[34] It has been proven that the combination of mixing units in micromixers improves the mixing efficiency.^[35] As such, two different planar mixing units (without obstacles) were selected and connected to form a hybrid micromixer, as depicted in Figure 4C. The results of this micromixer design illustrated the efficient mixing of two fluids to give a high mixing efficiency suitable for many applications. Moreover, height profile of the channel is similar to the input CAD file. The values of R_a and S_a for this micromixer were measured to be 0.248 and 0.596, respectively. As the flow regime in microfluidic mixers usually exists at a R_e of less than 100,^[36] indicating laminar flow, the surface roughness does not adversely affect the function of the device.

Microfluidic devices can be integrated to act as modular components of a larger process. A decrease in the turnover time between designs as well as increased design flexibility makes 3D printing a perfect candidate for the future modularization of microfluidic devices.^[37,38]

2.4. Biological Applications

In vitro cell culture platforms play a crucial role in cell biology, cancer research, regenerative medicine, pharmacy, and biotechnology. Although 2D cell culture in planar dishes is still widely used, this oversimplified model fails to mimic the actual cellular microenvironment. Alternatively, 3D cell culture platforms (mostly in the form of multicellular spheroids) are far more realistic models, which can better mimic in vivo responses.^[39] However, these static 3D systems are still sub-optimal and lack many of the critical features essential to a complex tissue microenvironment. Additionally, these systems cannot precisely control the chemical and nutrient concentration gradients over time and space. Furthermore, the oxygen tension and shear stress experienced by the cells are different from in vivo conditions.^[40] To address these shortcomings, microfluidic 2D and 3D cell culture platforms have emerged recently, progressing along with the rapid advances in microfabrication techniques.^[41] Such platforms offer several advantages to engineering a physiologically relevant biomimetic tissue.

Here, we chose a pear-shaped microchamber similar to the design proposed by Chong et al.^[42] The authors used the pearshaped design to minimize the shear stress during continuous perfusion. To fabricate the arrays of the microchambers, Liu et al. used standard dry etching on a silicon substrate followed by PDMS softlithography. The dimensions and characteristics of the 3D printed microchamber are shown in Figure 5A. The total printing time starting from the initial design to the final product took only 45 min. MCF-7 cells with a concentration of 10⁶ cells mL⁻¹ in culture media (Roswell Park Memorial Institute (RPMI) 1640 with 10% fetal bovine serum (FBS) and 1% streptomycin–penicillin) were introduced into PDMS microchamber. The device was incubated for 24 h at 37 °C with 5% CO₂. To evaluate the cell viability in the PDMS microchamber, live/dead cell double staining was performed. As shown in Figure 5B,C, more than 98% of the cells remained viable in the microchamber 24 h after the initial cell seeding. This confirms that no cytotoxic residual material had been left on the PDMS from casting on the 3D printed resin. Also, in cell culture platforms, flow rates exist in the order of µL min−1,^[43] and the values of R_a and S_a, as shown in Figure 5A, indicate that the device is functional within its flow regime. Therefore, it can be concluded that the newly developed resin for 3D printing master molds is suitable for cell culture applications and does not compromise cellular viability. Currently, lung-on-a-chip studies using 3D printed microfluidic resin molds are under investigation in our group; these studies demonstrate long-term cell viability (more than a week).

The gradient of biomolecules plays a crucial in controlling various biological activities, including cell proliferation, wound healing, and immune response. One of the most popular types of CGGs that produces discontinuous concentrations is the tree-like CGG. This type of CGG is based on the fact that one can divide and mix the flow through bifurcations and pressure differences downstream. This type of CGG is usually used for cancer cell cultures, as these CGGs transfer more oxygen and nutrients to cells as they develop a convective mass flux. Among various tree-like CGGs proposed in the literature, we chose the S-shaped CGG design developed by Hu et al.^[44] The authors used micromilling to fabricate the CGG on a polymethylmethacrylate substrate. Here, we developed the same structure in PDMS using softlithography based master mold fabrication from our new microfluidic resin. Figure 4D shows the characteristics of the fabricated CGG. The device has two inlets and six outlets to produce six different concentration ranges. To examine the performance of the device, we used two colors of food dyes (please refer to the Supporting Information for dye preparation protocol). The concentration profile of the fabricated CGG is illustrated in Figure 5D, which is similar to those reported by the literature.^[44] Since the velocity in CGG devices is small,^[45] surface roughness cannot impose problems on the binding of PDMS. For printing of planar structures, 3D printing can be performed with higher slice thickness, as a result of which, printing time will be reduced.

In summary, the microfluidic resin for 3D printing is an ideal candidate for fabricating different bio-microfluidic devices and can replace all cost-intensive and time-consuming fabrication methods.

3. Conclusion

In this study, we introduced a microfluidic resin for direct fabrication of master molds for PDMS softlithography, which can substitute other time-consuming master mold fabrication methods. Conventionally, the main components of SLA/DLP resins are acrylated monomers and oligomers. These materials cannot provide a temporary attachment to PDMS without leaving uncured PDMS on the surface of the mold, indicating that the PDMS cannot replicate the mold pattern. In the proposed master mold microfluidic resin, methacrylated monomers and oligomers have been used to facilitate PDMS casting, the proof of which was illustrated by fabrication of four benchmark microfluidic devices, including separator, micromixer, cell culture device, and a concentration gradient generator. In addition, the effects of velocity and shear rate distribution on the total performance of the microfluidic device were investigated numerically. It was shown that the surface roughness has to be small enough so as to not create extra shear stress endangering PDMS bonding. As the fabricated devices were tested in wide ranges of Re, we showed that there was not any leakage in these microfluidic devices. The height profile also confirmed that there was not any major discrepancy between the CAD geometry and the fabricated part. The results of the spiral microchannel for flow rates from 0.5 to 3 mL min⁻¹ illustrated that the behavior of particles in spiral microchannel was in line with those reported in the literature, and the microchip could withstand high flow rates. The characterization of the micromixer also demonstrated that the proposed microfluidic resin was able to fabricate microchannels with different geometries, and the mixing result was appealing so that two tested color dyes mixed completely. In the conventional softlithography process, silanization is necessary to prevent the attachment of PDMS to the master mold, which can be detrimental for cellular studies. The 3D printed mold obtained from the microfluidic resin proposed here does not require any silanization, and the cellular studies in the PDMS-based cell culture device confirmed the biocompatibility of the resin. The 3D printed CGG device produced a stable gradient profile, implying the application of such a versatile 3D printing technique for effective drug delivery. As PDMS-based microchannels are ubiquitous in microfluidic devices, the present study can be considered as a milestone in the microfluidic field which can reduce the brainstorming-to-production from a time frame of several days (including the time required for conventional master mold fabrication and post-treatment) to less than 5 h (with the new proposed microfluidic resin).

Apparatus Used

Master Mold for PDMS

Learn More

Ultra 50
3D Printer

Legacy

4. Experimental Section

Resin: As SLA/DLP printing process has risen in popularity, concern over its compatibility with PDMS is now an issue. The commercial resins used for DLP 3D printing of microfluidic devices were acquired from Creative CADworks company are BV-003 and BV-007 (manufactured by MiiCraft, Taiwan), which have been broadly used in the literature^[46–48] (please refer to the Supporting Information for a detailed description of these two resins). However, these resins proved to be not effective for PDMS casting. As previously mentioned, although certain surface treatments for 3D printed molds (prior to PDMS casting) have been trialed, all are either time-consuming, nonreplicable, or not effective. These two resins are composed of acrylated monomers and oligomers. However, the required surface treatment for PDMS casting impedes their further applications in microfluidic devices. Thus, methacrylated monomers and oligomers were substituted to form a microfluidic resin, which is suitable for direct PDMS casting without any post-treatment. In conventional DLP resins, COCHCH₂ exists in their functional groups. These components are not proper for the PDMS casting (i.e., incomplete cure of PDMS), and several groups tried to come up with a surface treatment strategy to mitigate this issue.^[18] This problem is attributed to the acrylate groups, resulting in the utilization of methacrylated monomers and oligomers instead of them. Indeed, hydrogen (H) in the chemical formulation of acrylate components was replaced by methyl (CH₃) to form the COCCH₂CH₃ group. The resultant resin possesses a viscosity in the range of 175–230 cps.

The polymer network of the methacrylate composites was shaped by the so-called process of “free-radical addition polymerization” of the corresponding methacrylate monomers. The process of polymerization happens in three stages, which are initiation, propagation, and termination. In this process, usually volume shrinkage is observed as a result of Van der Waals volume or the free volume reduction.^[49] This volume reduction can be minimized by either adding the prepolymerized resins to the monomer resins, utilizing methacrylate monomers with high molecular mass, or increasing the percentage of inorganic filler. These monomers modify the final surface of the resin and eliminate the uncured layer in contact with the PDMS, making it appropriate for PDMS casting.^[50] The exact formulations and chemical compositions of the developed microfluidic resin are proprietary to Creative CADworks.

3D Printer Specifications, Printing Parameters, and PDMS Casting: In this study, to create the molds, a MiiCraft Ultra 50 3D printer (MiiCraft, Hsinchu, Taiwan) was used, which has a printing area of 57 × 32 × 120 mm³ and XY resolution of 30 µm. The UV wavelength used in this device is 385–405 nm, which projects from the bottom of the resin bath filled with microfluidic resin. The operating temperature is 10 to 30 °C, and the operating humidity is 40% to 60%. The desired geometries were drawn in Solidworks 2016, a commercial CAD/CAE software, and then exported with the STL file format suitable for 3D printers. The STL file is imported into the Miicraft software (MiiCraft 125, Version 4.01, MiiCraft Inc), a software for preprocessing of design models. The imported file must be sliced to shape the desired geometry. The slicing in Z direction can be adjusted from 5 to 200 µm (with an increment of 5 µm). Reducing the thickness layer increased the final quality of the product. Since the modified resin has a high viscosity, the curing time of each layer is a challenging factor. In addition, the base and buffer layers must be carefully adjusted to allow the part to adhere to the picker without falling. When selecting a slice thickness of 10 µm for smaller features, it was better to set the curing time for each layer between 5 and 6 s. For slice thicknesses of 30 and 50 µm, the optimum curing times were found to be 6.5 and 9.5 s, respectively. The base layer is the layer that accounts for the bonding of the part to the picker. The curing time for the base layer was set to 60 s. The buffer layer was used to reduce the curing time between the base layer and subsequent part layers. As the UV light cures each layer, the Z-axis stepper motor displaced the sample one slice upward, before curing the next layer. This process continued until the whole geometry was printed. Once printed parts were removed from the picker, they were rinsed thoroughly with isopropanol. Next, an air nozzle was used to remove residual resin from the edges and in between extremely fine features. Eventually, the mold was postcured by exposing each part to the UV light in a curing chamber with a wavelength of 405 ± 5 nm. Upon fabrication of master molds, the PDMS prepolymer and the curing agent (Sylgard 184 from Dow Corning, MI, USA) were mixed in the ratio (W/W) of 10:1. This process was followed by degassing in a vacuum chamber for 15 min and pouring the liquid PDMS onto the 3D printed microfluidic mold without any surface treatment process. Afterward, it was kept in an oven to complete the curing of PDMS. Subsequently, the cured PDMS was peeled off, and the inlet and outlet holes were punched. The PDMS-based microchannel was then bonded onto either a glass slide or another PDMS substrate by plasma activation to form a closed channel. The schematic illustration of microchip fabrication based on the proposed resin is illustrated in Figure 6.

Figure 6. The workflow of the master mold preparation by DLP/SLA 3D printing method and microfluidic resin. A) The desired master mold is drawn. The beauty of microfluidic devices is that they require neither intricate geometries nor professional CAD drawer. Thus, the CAD drawing process will not take a long time. B) The design is then printed using a DLP/SLA 3D printer, and the residuals are removed from the surface of the mold. C) Afterward, PDMS is poured in the master mold, and D) in the final stage, PDMS is peeled-off, bonded to a glass or PDMS layer, and the finalization followed by the installation of inlets and outlets.

Benchmark Microfluidic Devices: In order to investigate the microchips fabricated via the 3D printed microfluidic mold, four benchmark
devices were tested. Generally, microfluidic devices are classified into two categories, those for liquid-handling and those for biological application.^[51] To showcase liquid handling using the proposed 3D printing resin, a spiral microchip for separation and a micromixer for mixing two fluids were fabricated.

It has been shown that spiral microchannels with a trapezoidal cross-section are useful in particle/cell separation for a wide range of flow rates.^[52] However, the fabrication of the mold which was mainly conducted by micromilling is a challenging process and not suitable for fabrication of complex cross-sections. By testing this device (please refer to the Supporting Information for sample preparation), the feasibility of fabricating a 3D-direct-printed spiral mold with a trapezoidal cross-section was evaluated, and the surface profile of the microchip and the bonding quality were assessed.

Mixing is an essential step in most chemical processes, and micromixer is an integral part of micro total analysis systems (µTAS). As such, the feasibility of producing planar micromixers has been showcased with a combination of two different mixing units adopted from Hossain and Kim^[53] and Bhopte et al.^[54] using the aforementioned technique (please refer to the Supporting Information for dye preparation). Finally, a specific design for cell culturing and concentration gradient generation for preparation of a drug with different dosages were selected. The cell culture device was selected to investigate the biocompatibility of 3D printed devices for cell culture applications (please refer to the Supporting Information for cell viability assay). The schematics of these devices with their specific dimensions are drawn in Figure 7.

Surface Characterization: Surface characterizations of the 3D printed mold and PDMS were analyzed using 3D laser microscopy (Olympus LEXT OLS5000); to this end, an LMPLFLN 20× LEXT objective lens (Olympus) was selected. Arithmetic mean deviation (R_a), the arithmetic mean of absolute ordinate Z (x,y) documented along a sampling length, and arithmetical mean height (S_a), the arithmetic mean of the absolute ordinate Z (x,y) documented along an evaluation area were chosen to evaluate the surface characterization of the samples. In order to investigate the velocity profile and shear stress along the length of the microchannel with a rough-embedded surface, COMSOL Multiphysics 5.3a, a commercial software based on the finite element method, was used. By applying the parametric surface function within COMSOL, two different S_a values (0.3 (attributed to the measured surface roughness of the spiral microchannel) and 1 µm) were evaluated. To apply roughness on the bottom of the channel, Equation (1) was used

where x and y are spatial coordinates, N and M are spatial frequency resolutions. The spectral exponent is controlled by β, and g(m,n)
and ϕ(m,n) are zero mean Gaussian and uniform (in the interval between −π/2 and π/2) random functions, respectively. In this study, the values of M and N were set to 40, and β was set as 2. Thereafter, f(x,y) was scaled in the Z direction to get the desired value of surface roughness.^[55] Based on Equation (2), to identify the surface roughness, the amplitude parameter of S_a was used

where the mean-plane area is identified by A. A microchannel with dimensions 400 × 50 × 30 µm³ was considered, and the rough surface was applied at the bottom of the channel. In the simulations, flow was considered to be steady-state, incompressible, and Newtonian with the same properties as water. Uniform velocity was applied to the inlets, zero static pressure was assigned to the outlet, and all other walls were considered to be no-slip boundary condition.

Supplementary Materials

Download

References

Yenilmez, F. Ghaderinezhad, S. Katebifar, M. Messina, A. Khademhosseini, S. Tasoglu, Biofabrication 2016, 8, 022001.
B. C. Gross, J. L. Erkal, S. Y. Lockwood, C. Chen, D. M. Spence, Anal. Chem. 2014, 86, 3240.
S. Waheed, J. M. Cabot, N. P. Macdonald, T. Lewis, R. M. Guijt, B. Paull, M. C. Breadmore, Lab Chip 2016, 16, 1993.
I. Wagner, Spending on 3D printing worldwide in 2019 and 2022 (in billion U.S. dollars), https://www.statista.com/statistics/590113/ worldwide-market-for-3d-printing/ (accessed: August 2018).
A. I. Shallan, P. Smejkal, M. Corban, R. M. Guijt, M. C. Breadmore, Anal. Chem. 2014, 86, 3124.
C. M. B. Ho, S. H. Ng, K. H. H. Li, Y.-J. Yoon, Lab Chip 2015, 15, 3627.
A. Lashkaripour, A. A. Mehrizi, M. Goharimanesh, M. Rasouli, S. R. Bazaz, J. Mech. Med. Biol. 2018,18, 1850002.
A. Lashkaripour, C. Rodriguez, L. Ortiz, D. Densmore, Lab Chip 2019, 19, 1041.
M. Mollajan, S. R. Bazaz, A. A. Mehrizi, J. Appl. Fluid Mech. 2018, 11, 21.
M. Rasouli, A. A. Mehrizi, M. Goharimanesh, A. Lashkaripour, S. R. Bazaz, Chem. Eng. Process. 2018, 132, 175
A. Lashkaripour, M. Goharimanesh, A. A. Mehrizi, D. Densmore, Microelectron. J. 2018, 78, 73.
P. F. Costa, H. J. Albers, J. E. Linssen, H. H. Middelkamp, L. Van Der Hout, R. Passier, A. Van Den Berg, J. Malda, A. D. Van Der Meer, Lab Chip 2017, 17, 2785.
T. Kwon, H. Prentice, J. De Oliveira, N. Madziva, M. E. Warkiani, J.-F. P. Hamel, J. Han, Sci. Rep. 2017, 7, 6703.
J. Park, K. I. Kim, K. Kim, D. C. Kim, D. Cho, J. H. Lee, S. Jeon, Adv. Mater. 2015, 27, 8000.
Z. Mahmoodi, J. Mohammadnejad, S. R. Bazaz, A. A. Mehrizi, M. A. Ghiass, M. Saidijam, R. Dinarvand, M. E. Warkiani, M. Soleimani, Drug Delivery Transl. Res. 2019, 9, 707.
M. A. Raoufi, A. Mashhadian, H. Niazmand, M. Asadnia, A. Razmjou, M. E. Warkiani, Biomicrofluidics 2019, 13, 034103.
N. P. Macdonald, J. M. Cabot, P. Smejkal, R. M. Guijt, B. Paull, M. C. Breadmore, Anal. Chem. 2017, 89, 3858.
H. N. Chan, Y. Chen, Y. Shu, Y. Chen, Q. Tian, H. Wu, Microfluid. Nanofluid. 2015, 19, 9.
T. Dinh, H.-P. Phan, N. Kashaninejad, T.-K. Nguyen, D. V. Dao, N.-T. Nguyen, Adv. Mater. Interfaces 2018, 5, 1800764.
K.-i. Kamei, Y. Mashimo, Y. Koyama, C. Fockenberg, M. Nakashima, M. Nakajima, J. Li, Y. Chen, Biomed. Microdevices 2015, 17, 36.
G. Comina, A. Suska, D. Filippini, Lab Chip 2014, 14, 424.
B. Parker, R. Samanipour, A. Ahmadi, K. Kim, IET Micro Nano Lett. 2016, 11, 41.
P. H. King, G. Jones, H. Morgan, M. R. R. de Planque, K.-P. Zauner, Lab Chip 2014, 14, 722.
S. Waheed, J. M. Cabot, N. P. Macdonald, U. Kalsoom, S. Farajikhah, P. C. Innis, P. N. Nesterenko, T. W. Lewis, M. C. Breadmore, B. Paull, Sci. Rep. 2017, 7, 15109.
D. Karalekas, A. Aggelopoulos, J. Mater. Process. Technol. 2003, 136, 146.
X. Ye, H. Liu, Y. Ding, H. Li, B. Lu, Microelectron. Eng. 2009, 86, 310.
V. S. Voet, T. Strating, G. H. Schnelting, P. Dijkstra, M. Tietema, J. Xu, A. J. Woortman, K. Loos, J. Jager, R. Folkersma, ACS Omega 2018, 3, 1403.
C. Charton, V. Falk, P. Marchal, F. Pla, P. Colon, Dent. Mater. 2007, 23, 1447.
A. Ellakwa, N. Cho, I. B. Lee, Dent. Mater. 2007, 23, 1229.
A. Tay, A. Pavesi, S. R. Yazdi, C. T. Lim, M. E. Warkiani, Biotechnol. Adv. 2016, 34, 404.
A. Kulasinghe, T. H. P. Tran, T. Blick, K. O’Byrne, E. W. Thompson, M. E. Warkiani, C. Nelson, L. Kenny, C. Punyadeera, Sci. Rep. 2017, 7, 42517.
M. Rafeie, J. Zhang, M. Asadnia, W. Li, M. E. Warkiani, Lab Chip 2016, 16, 2791.
M. E. Warkiani, G. Guan, K. B. Luan, W. C. Lee, A. A. Bhagat, P. K. Chaudhuri, D. S. Tan, W. T. Lim, S. C. Lee, P. C. Chen, C. T. Lim, J. Han, Lab Chip 2014, 14, 128.
S. Jeon, V. Malyarchuk, J. O. White, J. A. Rogers, Nano Lett. 2005, 5, 1351.
S. R. Bazaz, A. A. Mehrizi, S. Ghorbani, S. Vasilescu, M. Asadnia, M. E. Warkiani, RSC Adv. 2018, 8, 33103
N.-T. Nguyen, Z. Wu, J. Micromech. Microeng. 2005, 15, R1.
A. K. Au, N. Bhattacharjee, L. F. Horowitz, T. C. Chang, A. Folch, Lab Chip 2015, 15, 1934.
N. Bhattacharjee, A. Urrios, S. Kang, A. Folch, Lab Chip 2016, 16, 1720.
Z. Koledova, 3D Cell Culture: Methods and Protocols 2017, p. 1.
N. Kashaninejad, M. J. A. Shiddiky, N.-T. Nguyen, Adv. Biosyst. 2018, 2, 1700197.
N. Kashaninejad, M. R. Nikmaneshi, H. Moghadas, A. Kiyoumarsi Oskouei, M. Rismanian, M. Barisam, M. S. Saidi, B. Firoozabadi, Micromachines 2016, 7, 130.
L. Chong, W. Lei, X. Zheng, L. Jingmin, D. Xiping, W. Qi, C. Li, J. Micromech. Microeng. 2012, 22, 065008.
M. Ni, W. H. Tong, D. Choudhury, N. A. A. Rahim, C. Iliescu, H. Yu, Int. J. Mol. Sci. 2009, 10, 5411.
Z. Hu, X. Chen, L. Wang, Chem. Eng. Technol. 2018, 41, 489.
A. G. Toh, Z. Wang, C. Yang, N.-T. Nguyen, Microfluid. Nanofluid. 2014, 16, 1.
] Y.-J. Park, T. Yu, S.-J. Yim, D. You, D.-P. Kim, Lab Chip 2018, 18, 1250.
E. Mattio, F. Robert-Peillard, L. Vassalo, C. Branger, A. Margaillan, C. Brach-Papa, J. Knoery, J.-L. Boudenne, B. Coulomb, Talanta 2018, 183, 201.
C.-K. Su, W.-C. Chen, Microchim. Acta 2018, 185, 1.
J. V. Crivello, E. Reichmanis, Chem. Mater. 2014, 26, 533.
R. Sakaguchi, J. Powers, Craig’s Restorative Dental Materials, 13th ed., Mosby, Saint Louis 2012, p.161.
N.-T. Nguyen, M. Hejazian, C. H. Ooi, N. Kashaninejad, Micromachines 2017, 8, 186.
M. R. Condina, B. A. Dilmetz, S. R. Bazaz, J. Meneses, M. E. Warkiani, P. Hoffmann, Lab Chip 2019, 19, 1961.
S. Hossain, K.-Y. Kim, Micromachines 2014, 5, 913.
S. Bhopte, B. Sammakia, B. Murray, presented at 2010 12th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm), Las Vegas, Nevada, USA June 2010.
M. Villegas, Z. Cetinic, A. Shakeri, T. F. Didar, Anal. Chim. Acta 2018, 1000, 248.

H Series

H Series

Master Mold for PDMS

Clear Microfluidic Resin

Cyto-Clear Photopolymer Resin

Master Mold for PDMS

Clear Microfluidic Resin

Cyto-Clear Photopolymer Resin

Research Highlights

Academic Articles

Podcast: Big Ideas at μScale

Research Highlights

Academic Articles

Podcast: Big Ideas at μScale

Room temperature roll-to-roll additive manufacturing of polydimethylsiloxane-based centrifugal microfluidic device for on-site isolation of riboneucleic acid from whole blood

We kindly thank the researchers at Sungkyunkwan University for this collaboration, and for sharing the results obtained with their CADworks3D system.

1. Introduction

Apparatus Used

Master Mold for PDMS

Curezone

PR1103D Printer

2. Material and methods

2.1. Fabrication of negative 3D-printed mold

2.2. Fabrication of positive PDMS mold

2.3. PDMS formulations for R2R imprinting process

2.4. Roll-to-roll imprinting process and chip assembly

2.5. Replication accuracy and material characterization

2.6. Nucleic acid design and reagents

3. Results

3.1. Design and 3D printing of the centrifugal microfluidic devices

3.2. The multi-depth macro-to-micro flexible polymer shim

3.3. The fast-room temperature-curing PDMS

3.4. Roll-to-roll replication accuracy

3.5. RNA extraction from whole blood on the R2R additive manufactured LoaD

4. Discussion

5. Conclusions

Supplementary Materials

References

Skin-interfaced microfluidic systems with spatially engineered 3D fluidics for sweat capture and analysis

by Chung-Han Wu, Howin Jian Hing Ma, Paul Baessler, Roxanne Kate Balanay and Tyler Ray

We kindly thank the researchers at University of Hawai'i at Mānoa for this collaboration, and for sharing the results obtained with their CADworks3D system.

Introduction

Apparatus Used

Clear Microfluidic Resin

Curezone

PR1103D Printer

Legacy

Results

Sweatainer system design

Design and DLP printing considerations for optimized fabrication of 3D printed epifluidic devices

Print process optimization to support colorimetric analysis in 3D printed epifluidic systems

3D CBV designs for sequential sweat analysis

Field studies of the sweatainer

Discussion

Materials and Methods

Fabrication of 3D printed epifluidic devices

Apparatus Used

Clear Microfluidic Resin

Curezone

PR1103D Printer

Legacy

Fabrication of ultrathin capping layer for microfluidic channels in hybrid devices

Measurement of evaporation rate for 3D printed microfluidic networks

Characterization of 3D CBVs

Measurements of transmission properties of 3D printed devices

Colorimetric assay for chloride

Standard color development and color reference marker preparation

Digital image analysis for the evaluation of sweat chloride concentrations

Human participant sweat analysis

Applied tutorial for the design and fabrication of biomicrofluidic devices by resin 3D printing

by Hannah Musgrove, Megan Catterto and Rebecca Pompano

We kindly thank the researchers at University of Virginia for this collaboration, and for sharing the results obtained with their system.

1. Introduction

2. Materials and Methods

2.1. 3D Design and Printing

Apparatus Used

Clear Microfluidic Resin

PR1103D Printer

Legacy

2.2. Viability testing with primary murine lymphocytes

PR110
3D Printer

PR110
3D Printer

PR110
3D Printer

PR110
3D Printer

PR110
3D Printer

Ultra 50
3D Printer

Ultra 50
3D Printer