Rapid Fabrication of Multilayer Microfluidic Devices Using the Liquid Crystal Display-Based Stereolithography 3D Printing System

Abstract

We present a generalized process to characterize a 3D printer for fabrication of microfluidic devices. With this process, researchers are able to determine the capability of SLA printers for a specific resin. We employed a liquid crystal display (LCD)-based SLA 3D printer to demonstrate the feasibility of the process and applied optimized parameters for fabricating multilayer 3D microfluidic devices. It has been found that the LCD-based SLA 3D printer can support fabrication of microfluidic devices with the features down to 400 μm for in-plane features and 800 μm for vertical and interconnection features. The optimized curing time of the 100-μm-thick layer is 5.5, 6.5, and 7.5 s for yellow, light green, and dark green resins, respectively. The 3D printed flow-focusing droplet generator worked properly and could generate droplets with sizes between 50 and 185 mm². Taken together, the presented strategy can be used to quantitatively analyze and understand the capabilities of SLA 3D printing systems, which greatly facilitate optimization of device design and fabrication processes.

Introduction

3D printing has been considered a key technique of the new industrial revolution.¹ It provides a rapid and cost-effective way to fabricate small quantities of customized products. 3D printing has become a promising technology for medical and biomedical applications that require personalized design and biocompatibility of materials.² For example, 3D printing technologies have been used to produce a personalized medical device for treatment of airway diseases,³ patient-specific dental implants,⁴ and customized hearing aids for patients.⁵ In addition, living cells were printed for regeneration of tissues and organs using 3D printing.^6,7 3D printed microfluidic devices have also recently been developed for many biomedical applications such as disease diagnosis and high-throughput biomolecular analysis.^8,9

Microfluidic platforms have significantly affected many biomedical engineering research areas, such as tissue engineering,¹⁰ organ-on-a-chip,¹¹ DNA and RNA analysis,^12,13 cancer cell detection,¹⁴ single-cell analysis,¹⁵ point-of-care testing,¹⁶ and wearable drug delivery.¹⁷ Microfluidic platforms are well suited for biomedical applications taking advantage of their miniaturization, automation, and integration capabilities over traditional laboratory platforms such as Petri dishes, well plates, and flasks.^18,19

For example, microfluidic platforms have greatly benefited the application of cell manipulation and analysis. Microfluidic platforms allowed the high-efficiency trapping²⁰ and rotation^21,22 of a single cell that help the physiological measurement of the single cell. In addition, microfluidic platforms were used for rapid and high-throughput cell trapping²³ and separation.²⁴ These capabilities recently enabled the capturing of white blood cells²⁵ and rare circulating tumor cells (CTCs)^26,27 from unprocessed whole blood in an extremely sensitive manner. As CTCs are considered the key elements of cancer invasion and metastasis through peripheral blood, the fast and accurate capturing of CTCs would greatly improve the diagnosis and treatment of cancerous tumors.²⁸

Microfluidic devices are generally fabricated through microfabrication processes, which start from fabricating a mold using the photolithography process and casting microfluidic channels over the mold using a soft lithography process.^29–33 Such microfabrication processes require expensive cleanroom facilities and harmful chemicals. Therefore, researchers have adopted conventional manufacturing methods, such as CO₂ laser engraving^34,35 and industrial etching techniques,³⁶ to fabricate microfluidic devices with a low cost. However, these methods usually lack versatility and work for only one type of material. For example, industrial etching is specially designed for the printed circuit board with a thick copper layer. Introducing 3D printing technologies provides us with a rapid, versatile, and cost-effective fabrication process for production of microfluidic devices. Currently, there are two different ways of fabricating 3D printed microfluidic devices. First, 3D printing fabricates a mold for the soft lithography process.^37–40 This method still requires the subsequent soft lithography process to fabricate desired microfluidic channels. Second, 3D printing directly fabricates the microfluidic channel. The direct printing method has attracted more interest among researchers since it offers the rapid prototyping process of complicated devices and quick feedback of device performance to facilitate the rapid development and application of microfluidic devices.⁴¹

Different 3D printing techniques have been utilized for fabrication of microfluidic devices, including extrusion-based printing⁴² and stereolithography printing.^41,43 It has been found that the method of extrusion-based printing can only print channels with the size down to 4 mm.⁴² Thus, it is not suitable for fabricating microfluidic channels less than 1 mm. Stereolithography (SLA) employs patterned light from a scanned laser, beam projector, or liquid crystal display (LCD). It is more suitable for fabrication of microfluidic channels due to its micrometer printing resolution. Many microfluidic devices, such as droplet generators and valves, have been successfully printed through beam projector-based SLA.^41,43,44 However, no similar studies using LCD-based SLA have been conducted so far. As LCD-based SLA offers several advantages over the other SLA techniques (i.e., less expensive and large printing area), it is desired to use the LCD-based SLA for microfluidic devices considering cost and mass production. Supplementary Table S1 (Supplementary Data are available online at www.liebertpub.com/3dp) in the supplementary information provides a comparison between different SLA techniques. In addition, in spite of the success of 3D printed microfluidic devices, details of the fabrication process and its limitations have not been discussed until now. Fabrication of microfluidic devices often requires customized materials based on the applications. In addition, printing conditions, such as light intensity, curing time, and thickness, vary according to the materials and applications. As a result, it is necessary for researchers to develop their own protocols.

In this work, we present a generalized process to characterize an LCD-based SLA 3D printer for fabricating microfluidic devices. With this process, researchers are able to investigate the capability of SLA 3D printers using specific resins. The proper curing time of each layer and the resolution in three axes for a typical geometry can be determined. We used a prototype of the LCD-based SLA 3D printer (3D Currax Solutions Inc., Kelowna, BC, Canada) to demonstrate the feasibility of the characterization process for fabricating multilayer 3D microfluidic devices.

Materials and Methods

Printing system and materials

The LCD-based SLA 3D printer is a bottom-up-type printer that has an array of light-emitting diodes placed underneath an LCD screen, which forms a 20 × 15-cm printing area of 200 μm pixel resolution. An acrylate-based low-viscosity resin (Spot-LV; Spot-A materials, Barcelona, Spain) was used for printing. The polymer resin can be cured by blue/ultraviolet light with wavelengths from 365 to 405 nm. Yellow, light green, and dark green-colored resins were prepared by mixing the clear resin with 0%, 0.5%, or 2% v/v green dye (Spot-A materials, Barcelona, Spain). 3D models for printing were designed using SolidWorks (Dassault Systèmes, France) software and exported to a stereolithography (STL) file format. The STL files were then imported into the 3D printing software (CW3D, 3D Currax) to complete printing.

Characterization of printing condition

As shown in Figure 1, the curing time of resin for a specific thickness was characterized by exposing a small resin spot to light from the printer for a controlled time using DLPMaskGen software (Copyright Henry T. Locke). We characterized 16 time points, which are 3–10.5 s in increments of 0.5 s. The residue of uncured resins after exposure was removed by microfiber tissues. The thickness of each small spot was measured using a digital caliper.

FIG. 1.

Characterization of curing time and optical transmittance. (A) Schematics to illustrate the importance of proper exposure to the correct formation of the microfluidic channel. During 3D printing, the thickness of cured layer needs to be well controlled. Under or overexposure will result in printing failure or infeasible device. (B) Small spots of cured layer under different exposure times (from 0.5% dye resin). Each sample was measured by a digital caliper to determine its thickness. (C) Relationship between curing time and cured thickness for different dye concentrations (n = 3). (D) Measured light intensity passing through different samples placed on the optical path of the microscope (n = 6). (E) Optical transmittance of printed samples with 0%, 0.5%, or 2% dye (n = 6).

Three-millimeter-thick samples were printed using resins with 0%, 0.5%, or 2% dye. The curing time of each layer with the thickness of 100 μm was determined by the aforementioned method. The samples were then placed under an inverted microscope (Carl Zeiss AG, Oberkochen, Germany) with light intensity of 16,384 (a.u.). The measurement was carried out with 1.5-ms exposure time and 5 × lens under the same bright-field illumination. The maximum light intensities in the field of view with/without printed samples were measured to calculate the optical transmittance of each sample using the following formula: \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\usepackage{upgreek}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}&{ \rm{Optical \ Transmittance}} \\ & = {{{ \rm{Maximal \ light \ intensity \ received \ with \ the \ sample}}} \over {{ \rm{Maximal \ light \ intensity \ received \ without \ the \ sample}}}}\end{align*} \end{document}

Printing resolution was characterized by adding 0% and 2% green dye to the resin and printing the designed 3D structures with various patterns, as shown in Figures 2 and 3. The optimized curing time of a 100-μm-thick layer was investigated using the aforementioned method. To determine an in-plane (x-y plane) printing resolution, we printed half of the 1-cm-long rectangular and circular channels with a width or diameter of 100, 200, 400, 600, 800, 1000, 1500, and 2000 μm, respectively. To determine the minimum size of a vertical hole, we printed 3-mm-deep square and circular holes with a width or diameter of 200, 400, 600, 800, 1000, 1500, and 2000 μm, respectively. Finally, to characterize a vertical printing resolution (z direction), we printed 3-cm-long square and circular channels with a width or diameter of 400, 600, 800, 1000, 1500, and 2000 μm, respectively. The printed structures were then examined under an optical microscope (Carl Zeiss).

FIG. 2.

Characterization of actual resolution in the horizontal direction. (A) Photograph and microscope image of printed step pattern to determine the minimal resolution of a single layer in the horizontal direction. (B) Comparison of designed width and printed width in the horizontal direction (n = 3). (C, D) Photograph and microscope images of the printed (C) square and (D) circular via-hole pattern (3 mm in thickness) to determine the minimal resolution for interconnection structures between layers.

FIG. 3.

Characterization of actual resolution in the vertical direction. (A) Photograph and microscope images of printed step pattern to determine the minimal resolution in the vertical direction for channel forming. The channels, if properly connected, were visualized by adding 5% red food dyes. (B) Microscope images of printed channels to determine the minimal resolution of the printable square and circular channels. (C) Comparison of designed width and printed width for square channel printing (n = 3). (D) Comparison of designed width and printed width for circular channel printing (n = 3). Color images available online at www.liebertpub.com/3dp

Microfluidic device fabrication and experiment

To demonstrate the feasibility of fabricating microfluidic devices using optimized parameters, we printed single-layer and multilayer flow-focusing droplet-generating devices and a multilayer flow-mixing device using resins with 0% dye. The size of the devices was 5.5 × 4 cm × 5 mm. We infused flows mixed with 5% v/v red and blue food dyes to visualize the channels for the devices.

A droplet generation experiment was conducted by using two different fluid phases of 5% v/v red food dye in deionized (DI) water and 20% v/v Span 80 surfactant (Sigma-Aldrich, St. Louis, MO) in a mineral oil (VWR International, Radnor, PA). The two fluids were infused into the droplet-generating device through syringe pumps (Kent Scientific, Torrington, CT). The flow rate of the water phase was fixed at 15 μL/min. We tested the droplet generation process with the flow rate of the oil phase at 15, 30, 60, 120, 240, and 480 μL/min. The size of droplets was quantified by the WCIF ImageJ bundle plug-ins (University Health Network, Toronto, ON, Canada). The experimental process was monitored by a digital camera with 1920 × 1080 resolution and 30 fps.

Results and Discussion

Curing time optimization

Figure 1A illustrates three different cases of bottom-up-type SLA 3D printing. During the printing process, a 3D model is sliced into several layers with a fixed thickness (usually around 50–100 μm). Each layer is then sequentially projected through an LCD screen. A photocrosslinkable resin is selectively cured on the building plate corresponding to the projected pattern of each layer. Once a layer is printed, the building plate moves up to release the printed pattern and then down to allow the resin to fill a gap between the previously printed layer and the bottom of the resin vat. The amount of resin between the bottom of the vat and the printed layer will be printed as a new layer, which is added to the previously printed layers for building a 3D structure. The bottom-up SLA 3D printing repeats such a stepwise process until entire models are completed. To fabricate embedded hollow structures (i.e., microfluidic channels) using the SLA 3D printer, curing time is important to precisely control the curing depth of each layer of resin.³⁰ According to the curing time, the printed structure can be categorized into three different cases, as shown in Figure 1A. First, undercuring time results in the thickness of newly cured layers being less than the gap between the previous layer and the bottom of the vat. Thus, the newly cured layer cannot adhere to the previous layer and fails to form an embedded channel structure. Second, proper curing time makes the newly cured layer adhere to the previous layer, and the unexposed area of resin on the previous layer remains uncured. Third, overcuring time leads an unexposed area of resin in the previous layer to be partly or fully cured during the subsequent curing process of the new layer. As a result, the channel disappears. Therefore, the curing time to control the curing depth of each layer needs to be well controlled in the region below the channel structure during SLA 3D printing.

The thickness of the cured layer corresponding to a curing time was characterized by printing spots, as shown in Figure 1B. Light intensity, illumination time, and dye concentration in the resin regulate the thickness of the cured layer for a given resin. The SLA 3D printer projects constant light intensity within a certain time period. Therefore, we can control the thickness of the cured layer by changing the illumination time and/or the dye concentration. The DLPMaskGen software offers an express algorithm to illuminate a group of small circular shapes with different exposure times. Thus, the software was used to determine the optimized curing time for a certain thickness of layers. The relationship between cured thickness and curing time of resins with 0%, 0.5%, and 2% of dye concentration was characterized as shown in Figure 1C. It revealed that the dye concentration downregulated the cured thickness within a fixed time. In addition, the relationship between the curing time and cured thickness was not linear. After a certain time of illumination, the thickness was not increased, rather it was saturated. To print a 100-μm-thick layer, the minimum curing times for 0%, 0.5%, and 2% resins were 5–5.5, 6–6.5, and 7–7.5 s, respectively. It is worth noting that a little overcuring of the new layer is more acceptable for 3D printing than separating from the previous layer. Therefore, we chose 5.5 s for 0% and 7.5 s for 2% resins, which are a little higher than the threshold value of the curing time, to maintain the robustness of printing. Using this method, the proper curing time for layers with different thicknesses (50–150 μm) can be determined to achieve fine vertical resolution or fast fabrication time that can benefit specific applications.

The transparency of printed microfluidic devices is important to visualize various fluidic phenomena inside the devices with a microscope. We measured the light intensity passing through the printed samples placed on the optical path of the inverted microscope. As can be seen in Figure 1D, more light was absorbed by the samples as the dye concentration was increased. Optical transmittance was quantified by dividing the light intensities with/without the printed samples (Fig. 1E). The optical transmittance of the samples printed with 0%, 0.5%, and 2% dye was 69%, 57%, and 25%, respectively.

Printing resolution

In addition to optimization of the curing time, characterizing the capability of printing the smallest feature is one of the important parameters for the SLA 3D printing system. The SLA 3D printer cannot usually reach the printing resolution at its pixel resolution level, or maintain the consistent resolution in a large area, because of the light diffraction from cross talk.⁴⁵ For printing 3D microfluidic devices, key features include the minimum in-plane width of a channel, the minimum height of a channel, and the minimum size of holes for the interconnection between each layer. Such features were characterized by printing several patterns with square and circular channels, respectively.

Figure 2 shows the printing resolution of the in-plane direction. As shown in Figure 2A, B, although the minimum pixel size of the LCD screen was around 200 μm, the minimum size of features that the printer can print in the in-plane direction was around 400 μm. The discrepancies between the designed size and measured size are maximum (25.1%) at 400 μm and minimum (1.4%) at 2000 μm. In addition, all printed features were slightly larger than the designed size, which may be explained by the high shrinkage rate of low-viscosity resin (∼6%) (see Supplementary Table S2 in the supplementary file for the technical specification of resins). The printed samples were shrunk after printing and thus holes on the printed samples were enlarged. The resolution of the 3-mm-deep through-holes was defined as shown in Figure 2C, D. The minimum feature size of through-holes was around 800 μm instead of 400 μm. The increased minimum feature size was possibly due to the slight shifting of the through-holes' center during printing. It was also observed that circular holes tend to become square as the size gets smaller since inherently square-shaped pixels do not fit into a small circle.

The vertical resolution (z direction) was characterized by printing 3-cm-long microfluidic channels as shown in Figure 3. The minimum printable channel width was 800 μm, which was similar to through-holes. The cross-sectional views of the channels demonstrated the feasibility of using the LCD-based SLA 3D printer to print various shapes of embedded channels (Fig. 3B). Circular-shaped channels are more preferable than square-shaped channels since they generate a more uniform flow velocity profile.³⁹ Although the LCD-based SLA 3D printer is capable of fabricating embedded circular channels, the circular channels are close to becoming square as the size gets smaller. As shown in Figure 3C, D, the printed widths in the vertical direction are close to the desired size. For the square channels, the maximal discrepancy between the designed size and measured size, which is less than 10% of the channel size, is from the case of 0% dye with the 800-μm channel (actual size 871.667 μm). All other cases have less than 5% discrepancy. For the circular channels, in general, the discrepancy is around 10% and is much bigger than that of square channels. The maximum discrepancy, which is around 18%, is also from the case of 0% dye with the 800-μm channel (actual size 1046.67 μm). A variety of printing patterns provided us a better understanding of the optimized combinatorial parameters between the LCD-based SLA 3D printer and resin for printing microscale channels with minimum feature sizes of 400 μm in-plane (x-y direction) and 800 μm in z direction through-holes. With this in mind, we optimized the system to print complex microfluidic devices.

Droplet generation experiment

To test the system's capability of printing working devices, a flow-focusing droplet-generating device was designed and fabricated (Fig. 4A, B). Circular channels on the device with a diameter of 800 μm were printed using the optimized SLA 3D printing system. Microfluidic droplet-generating devices have been used in bulk generation of microdroplets for several engineering applications, such as tissue engineering,⁴⁶ drug delivery,⁴⁷ and wireless imaging.⁴⁸ The microdroplets are formed by the laminar flow of the water-based dispersed phase when it is surrounded by the oil-based continuous phase at a cross junction. The viscous force of the continuous phase and the interfacial force of the dispersed phase play essential roles to generate microdroplets. Once the viscous force overcomes the interfacial force, water starts to break droplets to minimize its surface tension. By changing the flow rates of the two different phases, the size of droplets can be controlled. We inspected the microfluidic droplet-generating device using DI water mixed with 5% red food dye to make sure that the printed device is correctly sealed and the size of channels is uniform (Fig. 4B). Thanks to the good transmittance of resin (69%), the channel can be clearly observed under the camera.

FIG. 4.

3D printed flow-focusing microfluidic device and its applications for droplet generation. (A, B) Design and photograph of the planar flow-focusing droplet generator. (C) Images of the droplet generation process under different flow rates of the oil phase and droplet generation frequencies. (D) Relationship between the droplet generation frequency and the flow rates of the oil phase. (E) Relationship between the size of droplets and the droplet generation frequency (n = 5).

Subsequently, the flow rate of the water phase was kept constant at 15 μL/min. Various sizes of droplets were generated under varied flow rates of the oil phase (Fig. 4C). The droplets were properly generated at 30–240 μL/min. At 15 μL/min, the viscous force of the oil phase was not enough to break the water phase into droplets. On the contrary, at 480 μL/min, the droplet generation process became a jetting regime to form unstable droplets, which were not suitable for most of the applications.⁴⁹ Figure 4D presents the relationship between the droplet generation frequency and flow rate of the oil phase. At the flow rates of 30, 60, 120, and 240, the number of droplets generated per minute were 19.5, 30, 45, and 78, respectively, showing a strong linear correlation. A Supplementary Video clip shows the representative droplet generation process at the frequency of 78 droplets per minute. We quantified the size of droplets using ImageJ (Fig. 4E). The size of droplets was well controllable within a wide range of 49–186 mm². For the generation frequency of 19.5, 30, 45, and 78, the sizes of droplets were 185.4 ± 4.4, 158 ± 10.4, 101 ± 8.8, and 49 ± 4.0 mm², respectively.

Multilayer microfluidic devices

In spite of the fact that fabricated microfluidic devices have a single layer, 3D printing already has many advantages, such as a more rapid and cost-effective process over the traditional soft lithography process.⁹ 3D printing can benefit the fabrication of multilayer microfluidic devices as the soft lithography method for fabricating multilayer devices requires complicated fabrication steps involving the aligning and bonding of each layer.⁵⁰ We finally demonstrated the feasibility of using the LCD-based SLA 3D printer to fabricate multilayer microfluidic devices. Two different types of multilayer devices were printed. Figure 5A shows a flow-focusing droplet-generating device with two parallel cross junctions. The two cross junctions were able to generate microdroplets simultaneously. A 3D mixer, where two flows move periodically in the top- and bottom-layer channels and finally meet at a cross junction of a middle-layer channel, is shown in Figure 5B. It was clearly observed that red and blue fluids were moved through channels on different layers and mixed at the cross junction on the middle layer. Both printed devices had sharp and clear channels that functioned properly. The successful printing and application of multilayer microfluidic devices demonstrate that the LCD-based SLA 3D printing system and the optimized printing process provide a rapid and robust solution for fabricating a variety of microfluidic devices.

FIG. 5.

3D printed multilayer microfluidic devices for parallel droplet generation and mixing applications. (A) In the parallel droplet generator, the water phase was fed in from the upper layer and met the oil phase in the bottom layer for droplet generation. (B) In the mixer, red flow was fed in from the bottom layer and passed through to the top layer through a via-hole. Blue flow was fed in and passed through the layers, and vice versa. The two flows met at the junction on the middle layer at the end of the device and fused together. Color images available online at www.liebertpub.com/3dp

Conclusions

3D printing has been regarded as an emerging technique for fabricating microfluidic devices, which usually require an optimized combination of the 3D printing system and customized materials. In this article, we presented a generalized strategy to utilize an LCD-based SLA 3D printer to pattern a specific resin for printing various microfluidic devices. The simple and robust strategy included the characterization of curing time and the printing resolution in each direction. With this strategy, we characterized three different combinations of 3D printing parameters and resins to fabricate single-layer and multilayer microfluidic devices. The printed droplet-generating devices and mixers worked properly. Since there is limited information on optimization of the SLA 3D printing parameter, the presented strategy can be used to quantitatively analyze and understand the capabilities of SLA 3D printing systems. It also facilitates the optimization of device design and fabrication processes.

Footnotes

Acknowledgment

This work was supported by the Natural Sciences and Engineering Research Council of Canada Engage Grant (EGP 491617-2015).

Author Disclosure Statement

No competing financial interests exist.

References

Berman

. 3-D printing: The new industrial revolution. Bus Horiz, 2012; 55:155–162.

Ventola

. Medical applications for 3D printing: Current and projected uses. P T, 2014; 39:704–711.

Morrison

, Hollister

, Niedner

, et al. Mitigation of tracheobronchomalacia with 3D-printed personalized medical devices in pediatric patients. Sci Transl Med, 2015; 7:285ra64.

Khalyfa

, Vogt

, Weisser

, et al. Development of a new calcium phosphate powder-binder system for the 3D printing of patient specific implants. J Mater Sci Mater Med, 2007; 18:909–916.

Sandström

. The non-disruptive emergence of an ecosystem for 3D Printing—Insights from the hearing aid industry's transition 1989–2008. Technol Forecast Soc Change, 2016; 102:160–168.

Mandrycky

, Wang

, Kim

D-H

. 3D bioprinting for engineering complex tissues. Biotechnol Adv, 2016; 34:422–434.

Murphy

, Atala

. 3D bioprinting of tissues and organs. Nat Biotechnol, 2014; 32:773–785.

CMB

, Ng

, Li

KHH

, Yoon

Y-J

. 3D printed microfluidics for biological applications. Lab Chip, 2015; 15:3627–3637.

, Huynh

, Horowitz

, Folch

. 3D-printed microfluidics. Angew Chemie Int Ed, 2016; 55:3862–3881.

10.

Morimoto

, Hsiao

, Takeuchi

. Point-, line-, and plane-shaped cellular constructs for 3D tissue assembly. Adv Drug Deliv Rev, 2015; 95:29–39.

11.

Wang

, Samanipour

, Koo

, Kim

. Organ-on-a-chip platforms for drug delivery and cell characterization: A review. Sensors Mater, 2015; 27:487–506.

12.

Wang

, Lee

, Koo

, Kim

. Nanowire-based sensors for biological and medical applications. IEEE Trans Nanobioscience, 2016; 15:186–199.

13.

Fan

Y-Q

, Wang

, Gao

, Zhuang

, Tang

, Zhang

Y-J

. Recent development of droplet microfluidics in digital polymerase chain reaction. Chinese J Anal Chem, 2016; 44:1300–1307.

14.

Chen

, Li

, Sun

. Microfluidic approaches for cancer cell detection, characterization, and separation. Lab Chip, 2012; 12:1753.

15.

Huang

, Bian

, Cheng

, et al. Microfluidics cell sample preparation for analysis: Advances in efficient cell enrichment and precise single cell capture. Biomicrofluidics, 2017; 11:011501.

16.

Sia

, Kricka

. Microfluidics and point-of-care testing. Lab Chip, 2008; 8:1982–1983.

17.

Fan

, Gao

, Wang

, Zhuang

, Tang

, Zhang

Y-J

. Recent development of wearable microfluidics applied in body fluid testing and drug delivery. Chinese J Anal Chem, 2017; 45:455–463.

18.

Whitesides

. The origins and the future of microfluidics. Nature, 2006; 442:368–373.

19.

Mitchell

. Microfluidics—Downsizing large-scale biology. Nat Biotechnol, 2001; 19:717–721.

20.

Jin

, Deng

, Li

, et al. A microfluidic device enabling high-efficiency single cell trapping. Biomicrofluidics, 2015; 9:014101.

21.

Huang

, Tu

, Zeng

, Mi

, Li

, Wang

. Study of a microfluidic chip integrating single cell trap and 3D stable rotation manipulation. Micromachines, 2016; 7.

22.

Benhal

, Chase

, Gaynor

, Oback

, Wang

. AC electric field induced dipole-based on-chip 3D cell rotation. Lab Chip, 2014; 14:2717–2727.

23.

, Huang

, Li

, Xu

, Wu

, Wang

. Fluidic circuit based high-efficiency and large-scale single cell trap. Lab Chip, 2016; 16:4507–4511.

24.

Cheng

, Wang

, Ma

, Wang

, Ye

X-Y

. Bubble- and clogging-free microfluidic particle separation platform with multi-filtration. Lab Chip, 2016; 16:4517–4526.

25.

Cheng

, Ye

, Ma

, Xie

, Wang

. High-throughput and clogging-free microfluidic filtration platform for on-chip cell separation from undiluted whole blood. Biomicrofluidics, 2016; 10:014118.

26.

Poudineh

, Aldridge

, Ahmed

, et al. Tracking the dynamics of circulating tumour cell phenotypes using nanoparticle-mediated magnetic ranking. Nat Nanotechnol, 2016; 12:274–281.

27.

Poudineh

, Sargent

, Kelley

. Amplified micromagnetic field gradients enable high-resolution profiling of rare cell subpopulations. ACS Appl Mater Interfaces, 2017; 9:25683–25690.

28.

Jackson

, Witek

, Kamande

, et al. Materials and microfluidics: Enabling the efficient isolation and analysis of circulating tumour cells. Chem Soc Rev, 2017; 34:1037–1056.

29.

Xia

, Whitesides

. Soft lithography. Annu Rev Mat Sci, 1998; 28:153–184.

30.

Whitesides

, Ostuni

, Takayama

, Jiang

, Ingber

. Soft lithography in biology and biochemistry. Annu Rev Biomed Eng, 2001; 3:335–373.

31.

McDonald

, Duffy

, Anderson

, et al. Fabrication of microfluidic systems in poly(dimethylsiloxane), Electrophoresis, 2000; 21:41–54.

32.

Wang

, Samanipour

, Kim

. Organ-on-a-chip platforms for drug screening and tissue engineering. In: Biomedical Engineering: Frontier Research and Converging Technologies, Jo

, Jun

H-W

, Shin

, Lee

(Eds). Cham, Switzerland; Springer International Publishing: 2015, pp. 209–233.

33.

Bhatia

, Ingber

. Microfluidic organs-on-chips. Nat Biotechnol, 2014; 32:760–772.

34.

, Fan

, Conchouso

, Foulds

. CO2 laser-induced bump formation and growth on polystyrene for multi-depth soft lithography molds. J Micromech Microeng, 2012; 22:115037.

35.

, Fan

, Kodzius

, Foulds

. Fabrication of polystyrene microfluidic devices using a pulsed CO2 laser system. Microsyst Technol, 2011; 18:373–379.

36.

, Qiao

, Feng

, et al. PDMS-based microfluidic devices using commoditized PCBs as masters with no specialized equipment required. RSC Adv, 2017; 7:31603–31609.

37.

Kamei

, Mashimo

, Koyama

, et al. 3D printing of soft lithography mold for rapid production of polydimethylsiloxane-based microfluidic devices for cell stimulation with concentration gradients. Biomed Microdevices, 2015; 17:36.

38.

Chan

, Chen

, Shu

, Chen

, Tian

, Wu

. Direct, one-step molding of 3D-printed structures for convenient fabrication of truly 3D PDMS microfluidic chips. Microfluid Nanofluidics, 2015; 19:9–18.

39.

Ahmadi

, Samanipour

, Parker

, Kim

. Rapid fabrication of circular channel microfluidic flow-focusing devices for hydrogel droplet generation. Micro Nano Lett, 2016; 11:41–45.

40.

Wang

, Samanipour

, Kim

. The cleanroom-free rapid fabrication of a liquid conductivity sensor for surface water quality monitoring. Microsyst Technol, 2016; 22:2273–2278.

41.

Rogers

, Qaderi

, Woolley

, Nordin

. 3D printed microfluidic devices with integrated valves. Biomicrofluidics, 2015; 9:016501.

42.

Kitson

, Rosnes

, Sans

, Dragone

, Cronin

. Configurable 3D-Printed millifluidic and microfluidic “lab on a chip” reactionware devices. Lab Chip, 2012; 12:3267.

43.

, Lee

, Folch

. Mail-order microfluidics: Evaluation of stereolithography for the production of microfluidic devices. Lab Chip, 2014; 14:1294–1301.

44.

Morimoto

, Tan

, Takeuchi

. Three-dimensional axisymmetric flow-focusing device using stereolithography. Biomed Microdevices, 2009; 11:369–377.

45.

Sun

, Fang

, Wu

, Zhang

. Projection micro-stereolithography using digital micro-mirror dynamic mask. Sensor Actuat A Phys, 2005; 121:113–120.

46.

Cha

, Oh

, Kim

, et al. Microfluidics-assisted fabrication of gelatin-silica core—Shell microgels for injectable tissue constructs. Biomacromolecules, 2014; 15:283–290.

47.

Wang

, Samanipour

, Gamaleldin

, Sakthivel

, Kim

. An automated system for high-throughput generation and optimization of microdroplets. Biomicrofluidics, 2016; 10:054110.

48.

Jin

, Guerrero

, Klukas

, Holzman

. Microlenses with tuned focal characteristics for optical wireless imaging. Appl Phys Lett, 2014; 105:031102.

49.

Samanipour

, Wang

, Ahmadi

, Kim

. Experimental and computational study of microfluidic flow-focusing generation of gelatin methacrylate hydrogel droplets. J Appl Polym Sci, 2016; 133:43701.

50.

Love

, Anderson

, Whitesides

. Fabrication of three-dimensional microfluidic systems by soft lithography. Mrs Bull, 2001; 26:523–528.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

7.61 MB

0.00 MB

0.08 MB