The paper presents a novel and economic manufacturing process for microlens arrays (MLAs). This process uses micromilling machining, PDMS casting, and hybrid bonding between a glass substrate and PDMS membrane to create a microfluidic chip which is used for manufacturing MLAs on a PDMS substrates. MLAs of various diameters were fabricated for experiments, including 1000 μm, 500 μm, and 200 μm. The sag height of the MLAs is easily adjusted by controlling the pressure inside the microchannel to deform the PDMS membrane. Multiple experiments were conducted to characterize the performance of MLAs, the results of which demonstrate: (1) this fabrication process is able to manufacture MLAs with various dimensions and the diameter of an MLAs is determined by the size of micromilling bit and cutting path; (2) the sag height and curvature of MLAs can be controlled by the PDMS membrane thickness and the hydraulic pressure inside the microchannel; (3) an optical system was built to investigate the uniformity of MLAs and the experiment results showed uniform focal length of MLAs; (4) the resulting MLAs magnify tiny objects and significantly enhance the fluorescence signal emitted from the microchannel
© 2017 Optical Society of America
Expansive growth in the optoelectronic industry and intelligent manufacturing is fueling interest in methods for the efficient fabrication of precise optical components. Microlens arrays (MLAs) are widely applied in optical communication , flat panel display modules , optical storage , shack-hartmann wavefront sensors , beam shaping for illumination , and light gathering for solar collectors . Numerous fabrication methods have been developed for the fabrication of MLAs on polymeric substrates, including photoresist reflow [7–10], laser ablation [11,12], gray-scale photolithography [13–15], micro-jet fabrication , hot embossing [17–19], electroforming , surface-energy driven process , and electrically template dewetting .
A number of researchers have integrated adjustable microlenses into microfluidic platforms. Chronis et al.  reported an elastomer-based tunable liquid-filled microlens array fabricated using soft lithography. The focal length of the microlenses is pneumatically controlled via a microfluidic network to pressurize each microlens evenly, resulting in devices with a focal length ranging from hundreds of microns to several millimeters. Chen et al.  developed a controllable PDMS microlens on a PDMS substrate, wherein the PDMS top layer was deformed through the injection of pneumatic pressure from the bottom microchannel resulting in a spherical microlens with a diameter from 600 to 1400 μm. Experiment results reveal that the focal length of the microlens was from 3.82 mm to 10.64 mm. In a similar method, Jeong et al.  took advantage of the flexibility of PDMS to fabricate biconvex and meniscus lenses on a PDMS substrate using micromolding and photopolymer microdispensing technologies. They used this technique to create a microdoublet lens capable of minimizing optical aberrations for use in optical communications and medical imaging. Ren et al.  demonstrated a liquid lens that a focal length that can be controlled by a servo motor. The lens cell comprises an elastic membrane with planar glass plate, peripheral sealing ring, and fixed volume of liquid in a lens chamber. The volume of liquid injected into the lens cell alters the curvature and focal length to achieve a maximum focal length of approximately 6 m with maximum radius of approximately 2.7 m. Lee et al.  used an electromagnetic actuator to deform a PDMS membrane for use as a microlens. Changing the current input into the electromagnetic actuator proved effective in adjusting the focal length; however, the current tends to generate heat after extended operation.
While most of the methods for MLAs were based on photolithography and physical phenomenon, we herein demonstrated a novel and economic method for fabricating spherical MLAs using micromilling process and casting techniques with PDMS and ultraviolet (UV) curable adhesive. To make this approach closer to the commercialization of MLAs with injection molding machine, the final step of PDMS casting could be replaced with electroplating metal mold inserts for mass production. Micromilling is a rapid and cost-efficient tool for the fabrication of micro-scale features on polymeric substrates . In this study we used micromilling to fabricate the 1st mold insert on a poly(methylmethacrylate) (PMMA) substrate to initiate the manufacture of the MLAs. PDMS casting and UV adhesive casting were then used sequentially to manufacture a PDMS MLAs. Creating micro pillar features within the PMMA substrates and adjusting the pressure inside the bottom microchannel make it possible to deform the top PDMS membrane in order to produce a well-defined MLAs, in terms of the sag height, curvature, and focal length. To characterize the uniformity and performance of the resulting MLAs, multiple experiments were conducted to elucidate the following: (1) the relationships among membrane thickness, sag height, focal length, and pressure with the microchannel, as observed in MLAs with diameters of 1000 μm, 500 μm, and 200 μm; (2) the magnification and enhancement of fluorescence signals from the microchannel when the MLAs is mounted over a microchannel, and (3) understanding the uniformity of the MLAs with an home-built optical system.
Figure 1 illustrates the concept underlying the proposed fabrication process, in which a flexible PDMS membrane is used to form a spherical MLAs and the bottom microchannel is used to deliver the precise pressure required to deform the top PDMS membrane. Figure 1(a) shows the micromilled mold inserts used in the fabrication of the PDMS membrane. A cast PDMS membrane was formed from the top and bottom PMMA substrates and subsequently bonded to a glass substrate with the aid of plasma treatment [Fig. 1(b)]. Liquid is injected into the microchannel of the glass-PDMS hybrid microfluidic chip, such that the pressure accumulating within the chamber causes the PDMS membrane to deform into a spherical shape [Fig. 1(c)]. The deformed PDMS membrane is used as a 2nd mold for UV adhesive casting. The degree to which the PDMS deforms is easily controlled by adjusting the pressure inside the micro chamber. UV adhesive is poured over the membrane and quickly hardened under exposure to UV radiation for use as a 3rd mold in subsequent PDMS casting [Fig. 1(d)]. In the final step, the PDMS casting is used to transfer the spherical concave shape from the UV substrate to the PDMS substrate [Fig. 1(e)]. This PDMS MLAs [Fig. 1(f)] was then characterized in a number of experiments. In this proposed method, the diameter of the MLAs is determined by the dimensions of the standing structure, as shown in Fig. 1(a). The thickness of the PDMS membrane is determined by the height of the standing structure, as shown in the Fig. 1(a). The pressure accumulated inside the chamber is used to control the degree to which the PDMS membrane deforms [Fig. 1(c)], this step determines the physical dimension of the final MLAs.
3. Fabrication process
3.1 Micromilling 1st polymeric mold insert
Substrates of PMMA material were cut to 50 mm x 37.5 mm x 3 mm and then inscribed with micro features using a micromilling machine. The micromilling machine used in this study includes five major components: a spindle (E3000c, Nakanishi, Japan), a laser non-contact tool setting system to reset cutting datum following the exchange of milling bit (NC4, Renishaw, United Kingdom), a numerical controller (M515i, LNC Technology Co. Ltd., Taiwan), a compressed air/oil coolant system, and a milling bit holder for tool exchange. Air only or a mixture of oil and air was supplied to the cutting surface of the PMMA substrate via a nozzle throughout the cutting process. We employed a 2-flute end mill bit with a diameter of 200 μm (Taiwan Microdrill Co. Ltd., Taiwan) for manufacturing the micro pillar features on the PMMA substrates. Figure 1(a) presents the top and bottom PMMA mold inserts used for the manufacture of the PDMS membrane. Carefully designing the height of the micro pillar features and the gap distance between the two PMMA substrates made it possible to adjust thickness of the PDMS membrane. Figure 2(a) presents the micromilled bottom mold insert, Fig. 2(b) illustrates the micromilled bottom mold insert, and Fig. 2(c) presents an enlarged image from Fig. 2(a) showing the circular standing structures placed over the microchannel structure. As shown in Fig. 2(c), the diameter of the circular standing structures was 200 μm and the width and depth of the microchannel were 200 μm, respectively.
3.2 PDMS casting and bonding for microfluidic platform as 2nd mold insert
Following assembly of the top and bottom mold inserts, a standard PDMS casting process was used to manufacture PDMS membranes . The PDMS (Sylgard 184, Dow Corning) is prepared by mixing the pre-polymer and the curing agent in the weight ratio of 10:1. After degassing and the PDMS mixture was poured on the mold inserts, then the mold insert with PDMS mixture was put in an oven for 1.5 hours at 80°C. After demolding, the cast membrane was then bonded to a glass substrate to form a hybrid microfluidic chip. Prior to bonding, the PDMS membrane and the glass substrate were both exposed to plasma (PDC-32G, Harrick Plasma) for 2 minutes in a vacuum chamber. The PDMS was then brought in contact with the glass substrate to produce a hybrid chip suitable for experiments. Figure 2(d) presents the circular structures and microchannels on the PDMS membrane casting from the micromilled mold insert shown in the Fig. 2(c). Figure 3(a) shows the bonded hybrid microfluidic chip. The two standing structures located on the edge of the PDMS membrane were used for the connection of tubing and the wall structures around the microchannel were used to define a boundary for UV adhesive casting. Figure 3(b) presents the hybrid microfluidic chip in cross-section, clearly showing how liquid was injected into the chamber through the bottom microchannel. The pressure inside the chamber causes the PDMS membrane to deform into a spherical shape. As shown in Fig. 3(b), the thickness of the membrane is related to the degree of deformation and is therefore strongly related to the sag height and focal length of the resulting MLAs.
3.3 Replication using UV adhesive to produce 3rd mold inserts
The completed hybrid microfluidic chip was then used as a 2nd mold to produce a 3rd mold insert on a UV curable substrate [Fig. 1(c)]. A manual syringe pump was connected to the hybrid microfluidic chip and liquid was pumped to deform the PDMS membrane. Figure 4(a) shows red food dye injected into the hybrid microfluidic chip and Fig. 4(b) presents a close-up of the MLAs on the PDMS membrane. One can clearly make out the microchannels carrying red food dye beneath the MLAs, which caused the circular PDMS membrane to deform into a convex shape. UV curable adhesive (Slink P21-A, Slink Adhesive Technology) was then poured onto the PDMS membrane and sequentially exposed to UV radiation for 108 seconds under the irradiation intensity of 50 mW/cm2 (Xlite 500, OPAS, Taiwan). Following solidification, the UV substrate was demolded from the PDMS membrane for use as a 3rd mold insert for the final PDMS MLAs.
3.4 PDMS casting of microlens array (MLAs)
The UV mold insert was fixed within a fixture [Fig. 1(e)] in order to control the thickness of the final PDMS MLAs. Micromilling was used to form cavities in the PMMA substrate, into which was placed the UV mold inserts for subsequent PDMS casting. Again, standard PDMS casting was applied on the PMMA fixture with multiple UV mold inserts to manufacture PDMS MLAs [Fig. 1(f)]. Figure 4(c) shows the final MLAs on the PDMS substrate. As shown in Fig. 4(d), this fabrication method enables the production of novel MLAs with various curvatures and sag heights in various locations of the same substrate. In other words, this approach is applicable to the creation of localized MLAs.
This focus of this paper is to introduce this rapid and efficient fabrication method, elucidate the limitations for MLAs, and evaluate the uniformity and overall performance of MLAs. Several experiments were conducted to characterize the MLAs, including the following: (1) the relationships among membrane thickness, hydraulic pressure inside the microchannel, sag height, and focal length; (2) the effectiveness of the MLAs in magnifying tiny objects and enhancing fluorescence signals emitted from microchannels; and (3) the uniformity of MLAs, which is crucial to optical performance.
4.1 Characterization of microlens
As shown in Fig. 1(c), the sag height of the final PDMS MLAs was largely determined by the deformation of PDMS on the hybrid microfluidic chip. To elucidate this effect, we built a system by which to measure the deformation of the PDMS on the hybrid microfluidic chip, as shown in Fig. 5(a). The system included a sensor (PS100-10 Bar, Yalab, Taiwan) to measure the pressure accumulated within the microchannel, a manual syringe pump to push the liquid into the microfluidic chip and deform the PDMS membrane, and a microscope to record the deformation of the PDMS (as viewed from the side) for the estimation of sag height. Figure 5(b) presents the actual system used in the experiments. Our experiment results help to illustrate the relationship between hydraulic pressure and sag height as a function of membrane thickness and diameter.
4.2 Magnification and characterization of fluorescence by MLAs
The second experiment dealt with optical performance, wherein the MLA was used to magnify tiny objects and enhance the fluorescence signal from the microchannel when integrated atop a microfluidic chip. To observe the effects of magnification, the MLAs was placed over a paper covered with tiny black dots. A tool microscope (SAGE 2600, SAGE, Taiwan) was then used to capture images above the MLAs. To quantify the degree to which the MLAs enhances fluorescence signals, we micromilled a straight microchannel on a polymeric substrate, which was then bonded to another piece of PMMA substrate using ethanol solution. This fabrication process was described in a previous paper . The difference between fluorescence signals obtained with/without an MLAs atop the microfluidic chip was determined by placing the PDMS MLAs over the microchannels and introducing a fluorescence solution into the microchannel using a syringe pump. The fluorescence solution contained fluorescence particles (Leuchtstoffwerk, Breitungen GmbH, Breitungen, Germany) and DI water (the fluorescent particles were undissolved). The excitation wavelength of the fluorescence particles was 465 nm and the emission wavelength was 553 nm. A fluorescence microscope (Axio Imager A1, Zeiss) was used to measure the fluorescence signal emitted from the microchannel. After capturing the image, Image J software was used to quantify the intensity of the fluorescence signal and illustrate any differences.
4.3 Uniformity of MLAs
To understanding the uniformity of the MLAs fabricated from this proposed method, an optical system was built to investigate the focal length of MLAs. Figure 6(a) shows the components of this optical system, including the laser, lens combination for beam expansion, polarizer, and CCD camera. The fabricated MLAs is fixed on a x-y table, hence the MLAs can move either far away or closer to the CCD camera, to determine the focal length of the MLAs. Figure 6(b) shows the experiment setup from the top view. Once the image was captured by the CCD camera, the image will be imported into Matlab for intensity analysis. The analyzed result can be used as information to understand the uniformity of the MLAs in terms of the focal length or curvature.
5. Experiment results and discussion
5.1 Relationships among sag height, focal length, pressure, and membrane thickness
Figure 7(a) presents images of a PDMS microlens with a diameter of 200 μm, which was deformed through the application of various amounts of pressure. When the accumulated pressure was increased from 10 kpa to 120 kpa, the sag height increased from 5 μm to 96 μm. Figure 7(b) presents profiles of sag height as a function of pressure. With pressure of less than 90 kpa, deformation occurred around the central area of the PDMS membrane, such that the diameter of the microlens was smaller than 200 μm. When the pressure was increased to 90 kpa, the diameter of the microlens increased to 200 μm. At pressure values above 90 kpa, the diameter of the microlens exceeded 200 μm, due to deformation of the area surrounding the PDMS microlens. The experiment results clearly illustrated that diameter and sag height of the MLAs are both affected by the applied pressure, and the diameters of the final lenses are not always equal to the diameters of the standing posts on the micromilled mold inserts. Figure 8 presents the sag height of the microlens produced using PDMS membranes of various thicknesses and various pressure values. Figures 8(a)-8(c) respectively present the results obtained from microlenses with diameter of 1000 μm, 500 μm, 200 μm. The three figures clearly demonstrate the following: (1) The maximum sag height in each case approached 50% of the microlens diameter. Thus, microlenses with a larger diameter presented a higher sag height than did microlenses with a smaller diameter. The maximum height of the 1000 μm microlens was 500 μm, whereas the maximum height of 200 μm microlens was 100 μm; (2) Regardless of the microlens diameter, a thicker membrane resulted in less deformation. For example, microlenses with a membrane 122-μm thick presented a lower sag height than did those with membranes with thickness of 35 μm; (3) In all cases, thinner membranes presented a higher curve slope indicating a faster response to pressure. Figure 9 presents a comparison of three microlenses of various diameter but the same thickness. Clearly, a microlens with a larger diameter was more responsive to changes in pressure, resulting in more pronounced deformation. Currently the PDMS membrane thickness and the level of PDMS deformation is highly related to the design of mold insert, material of mold insert, and precision of the machining, hence more research with industry-level machines has to be conducted to increase the reproducibility. Based on the diameter of the microlens, the measured sag height, and the optical properties (index of refractive is 1.4), we used the equations in  to convert the data in Fig. 8 to render the focal length of the microlens as a function of lens diameter and pressure, as shown in Fig. 10. A microlens with diameter of 1000 μm has a maximum focal length of approximately 11 mm, corresponding to pressure of 5 kpa in Fig. 8(a). In all three cases, an increase in pressure resulted in a decrease in focal length. The focal lengths in Figs. 8-10 were estimated from the measured sag height and curvature. We then used the system in Fig. 6 to measure the actual focal length of each MLA (1mm diameter and various sag heights). Measurements were obtained from ten microlenses on a single substrate. As shown in Table 1, the difference in focal length between the estimates and measured values was less than 3%.
5.2 Magnification of tiny dots and enhancement of fluorescence signals using microlens
Figure 11 shows three PDMS MLAs with different sag heights and magnifications, all of which were fabricated from the same hybrid microfluidic chip. Changing the pressure inside the microchannel altered the degree to which the PDMS membrane on the hybrid microfluidic chip was deformed, resulting in PDMS MLAs with different sag heights. We used a paper with tiny black dots (diameter is 3.84μm) as a background to illustrate the difference in optical performance between the three MLAs. Clearly, the MLAs with the highest sag height (400 μm) in Fig. 11(c) produced the greatest magnification (diameter is 6.25μm, magnification is 1.62) whereas the MLAs with the smallest sag height (150 μm) in Fig. 11(a) produced the least magnification (diameter is 4.86μm, magnification is 1.25).
Figure 12 presents the results of fluorescence experiments. Figures 12(a) and 12(b) respectively present images of fluorescence in the microchannel captured using the microlens with diameter of 1 mm. The sag height in Fig. 12(a) is 235 μm whereas the sag height in Fig. 12(b) is 459 μm, corresponding to numerical aperture of 1.24 in Fig. 12(a) and 1.6 in Fig. 12(b). The images were captured using a fluorescence microscope and then analyzed to determine the intensity of the fluorescence signal. Figure 12(c) presents the analysis results (the lines shown in the Fig. 12(a) and 12(b) are the locations for analysis). The fluorescence signal from the microlens is clearly stronger than the direct signal from the microchannel (1000 a.u.). As indicated by the two peaks in Fig. 11(c), a microlens with a higher sag height of 459 μm greatly increased fluorescence intensity to 3500 a.u., whereas the microlens with a sag height of 235 μm increased fluorescence intensity to 2600.
5.3 Uniformity of MLAs
After fabricating MLAs on the PDMS substrate, the MLAs was fixed on the optical system shown in the Fig. 6(b). The x-y table was adjusted to ensure the imaging had the smallest circles on the CCD camera, meaning the distance between the MLAs and the CCD was the focal length. Figure 13(a) shows the captured image from the CCD, which had multiple dots due to the imaging of the MLAs. The captured image was imported into Matlab for intensity analysis, the results of which are presented in Fig. 13(b). The number next to each red dot in Fig. 13(a) indicates the maximum intensity of each peak, which clearly demonstrates that this manufacturing method could fabricate MLAs of high uniformity.
This paper reports a novel, straightforward, and cost-efficient method for the manufacture of MLAs in various dimensions. The proposed method involves micromilling and bonding to produce a hybrid microfluidic chip, followed by UV adhesive casting and PDMS casting. To make this manufacturing method closer to the commercialization of MLAs, the final step of PDMS casting could be replaced with electroplating metal mold inserts for mass production. In this study, we manufactured MLAs with diameters of 1000 μm, 500 μm, and 200 μm for characterization tests. Multiple experiments were conducted, which led to the following conclusions: (1) the specific sag height or focal length of MLAs can be achieved by carefully designing the diameter of the micro pillar structure, membrane thickness of the MLAs, and controlling the pressure inside the microchannel; (2) the resulting MLAs can be used to magnify tiny objects or help concentrate fluorescence signals from the microchannel to improve detection sensitivity; (3) an optical system was built to investigate the uniformity of the fabricated MLAs, and the result clearly showed that this proposed method can manufacture uniform MLAs; (4) this method can be potentially used for manufacture large-area MLAs with high density at a lower cost and shorter manufacturing period.
Ministry of Science and Technology (MOST 105-2221-E-011-061).
We appreciate Prof. Hung-Lin Hsieh from Mechanical Engineering Department of National Taiwan University of Science and Technology to help setup an optical system to measure the uniformity of MLAs.
References and links
1. M. He, X. C. Yuan, N. Q. Ngo, J. Bu, and S. H. Tao, “Single-step fabrication of a microlens array in sol–gel material by direct laser writing and its application in optical coupling,” J. Opt. A-Pure Appl. Opt. 6(1), 94 (2003).
3. K. Kurihara, I. D. Nikolov, S. Mitsugi, K. Nanri, and K. Goto, “Design and fabrication of microlens array for near-field vertical cavity surface emitting laser parallel optical head,” Opt. Rev. 10(2), 89–95 (2003). [CrossRef]
5. J. W. Pan, C. M. Wang, H. C. Lan, W. S. Sun, and J. Y. Chang, “Homogenized LED-illumination using microlens arrays for a pocket-sized projector,” Opt. Express 15(17), 10483–10491 (2007). [CrossRef] [PubMed]
6. J. H. Karp, E. J. Tremblay, J. M. Hallas, and J. E. Ford, “Orthogonal and secondary concentration in planar micro-optic solar collectors,” Opt. Express 19(104), A673–A685 (2011). [CrossRef] [PubMed]
8. C. P. Lin, H. Yang, and C. K. Chao, “Hexagonal microlens array modeling and fabrication using a thermal reflow process,” J. Micromech. Microeng. 13(5), 775–781 (2003). [CrossRef]
9. Y. S. Cherng and G. D. J. Su, “Fabrication of polydimethylsiloxane microlens array on spherical surface using multi-replication process,” J. Micromech. Microeng. 24(1), 015016 (2013). [CrossRef]
10. V. Lin, H.-C. Wei, H.-T. Hsieh, J.-L. Hsieh, and G.-D. J. Su, “Design and fabrication of long-focal-length microlens arrays for Shack-Hartmann wavefront sensors,” Micro & Nano Lett. 6(7), 523–526 (2011). [CrossRef]
11. K. Naessens, H. Ottevaere, P. V. Daele, and R. Baets, “Flexible fabrication of microlenses in polymer layers with excimer laser ablation,” Appl. Surf. Sci. 208, 159–164 (2003). [CrossRef]
14. W. Yu and X. Yuan, “UV induced controllable volume growth in hybrid sol-gel glass for fabrication of a refractive microlens by use of a grayscale mask,” Opt. Express 11(18), 2253–2258 (2003). [CrossRef] [PubMed]
15. M. H. Wu, C. Park, and G. M. Whitesides, “Fabrication of arrays of microlenses with controlled profiles using gray-scale microlens projection photolithography,” Langmuir 18(24), 9312–9318 (2002). [CrossRef]
16. D. L. MacFarlane, V. Narayan, J. A. Tatum, W. R. Cox, T. Chen, and D. J. Hayes, “Microjet fabrication of microlens arrays,” IEEE Photonics Technol. Lett. 6(9), 1112–1114 (1994). [CrossRef]
17. S. D. Moon, N. Lee, and S. Kang, “Fabrication of a microlens array using micro-compression molding with an electroformed mold insert,” J. Micromech. Microeng. 13(1), 98–103 (2002). [CrossRef]
18. C. Y. Chang, S. Y. Yang, M. S. Wu, L. T. Jiang, and L. A. Wang, “A novel method for fabrication of plastic microlens array with aperture stops for projection photolithography,” Jpn. J. Appl. Phys. 46(5A), 2932–2935 (2007). [CrossRef]
19. S. M. Kuo and C. H. Lin, “The fabrication of non-spherical microlens arrays utilizing a novel SU-8 stamping method,” J. Micromech. Microeng. 18(12), 125012 (2008). [CrossRef]
20. H. Sun, X. Wang, Y. Xiong, G. Liu, and K. Wang, “Fabrication of microlens based on overplating in electroforming,” J. Micromech. Microeng. 26(5), 055007 (2016). [CrossRef]
21. C. Jiang, X. Li, H. Tian, C. Wang, J. Shao, Y. Ding, and L. Wang, “Lateral flow through a parallel gap driven by surface hydrophilicity and liquid edge pinning for creating microlens array,” ACS Appl. Mater. Interfaces 6(21), 18450–18456 (2014). [CrossRef] [PubMed]
22. X. Li, H. Tian, Y. Ding, J. Shao, and Y. Wei, “Electrically templated dewetting of a UV-curable prepolymer film for the fabrication of a concave microlens array with well-defined curvature,” ACS Appl. Mater. Interfaces 5(20), 9975–9982 (2013). [CrossRef] [PubMed]
24. J. Chen, W. Wang, J. Fang, and K. Varahramyan, “Variable-focusing microlens with microfluidic chip,” J. Micromech. Microeng. 14(5), 675–680 (2004). [CrossRef]
27. S. W. Lee and S. S. Lee, “Focal tunable liquid lens integrated with an electromagnetic actuator,” Appl. Phys. Lett. 90(12), 121129 (2007). [CrossRef]
28. D. J. Guckenberger, T. E. de Groot, A. M. Wan, D. J. Beebe, and E. W. Young, “Micromilling: a method for ultra-rapid prototyping of plastic microfluidic devices,” Lab Chip 15(11), 2364–2378 (2015). [CrossRef] [PubMed]
29. K. Kim, S. Park, J. B. Lee, H. Manohara, Y. Desta, M. Murphy, and C. H. Ahn, “Rapid replication of polymeric and metallic high aspect ratio microstructures using PDMS and LIGA technology,” Microsyst. Technol. 9(1–2), 5–10 (2002). [CrossRef]
30. P. C. Chen and L. H. Duong, “Novel solvent bonding method for thermoplastic microfluidic chips,” Sens. Actuator B-Chem. 237, 556–562 (2016). [CrossRef]
31. H. T. Hsieh, V. Lin, J. L. Hsieh, and G. D. J. Su, “Design and fabrication of long focal length microlens arrays,” Opt. Commun. 284(21), 5225–5230 (2011). [CrossRef]