1. Introduction

There are increasing demands for 3D microstructures in the fields of cell culture [1], microfluidics [2–5], sensors [6], optics [7–10], architectural materials [11–13], and energy storage [14,15]. The major strategies for additive manufacturing include fused deposition modeling (FDM) [16], direct ink writing (DIW) [17], selective laser sintering (SLS) [18], computed axial lithography (CAL) [19], two-photon polymerization (TPP) [20,21], and stereolithography (SLA) [22,23]. Among them, digital light processing (DLP)-based projection stereolithography 3D printing has been praised for its superior capability in microstructuring.

However, there are limitations that need to be addressed when printing micropatterns in a vat by PSL. First, photopolymers with high viscosity can hardly be printed using PSL at high throughput. Based on vat polymerization, the fabrication speed of PSL depends not only on the photopolymerization process but also on rheological factors such as the viscosity and surface tension of the photopolymer [24,25]. Photopolymers with a high viscosity require long residence and precipitation times; thus, it is difficult to diffuse them into uniform ultrathin layers within a reasonable time, and the printing speed will be significantly reduced. Liquid bridge µSLA was proposed to decrease the usage of material and to enable the fabrication of highly viscous material within a limited printing volume (∼0.25 mL) [26]. CAL supports high-viscosity materials and even solids, which broadens the selection of light-assisted printing materials [19]. Second, the tuning range of PSL z-slicing is not an independently tunable parameter. The tuning range is affected by various factors, such as the focal depth of the projection lens, oxygen inhibition, viscosity of the material and fabrication rate [27–29]. Finally, multimaterial 3D printing is challenging due to the difficulties of material exchange in a vat [30]. The printing process has to be stopped periodically to manually change the material in the vat. To solve this problem, a fluidic cell or microfluidic chamber has been integrated with PSL for material exchange and chemical cleaning at the cost of increased material consumption [31,32]. Again, highly viscous materials can hardly be used in such multimaterial 3D printers.

In light of the aforementioned considerations, we propose an R2P-PSL 3D printer where layers of photopolymer are transported and polymerized on a flexible membrane instead of in a vat. R2P-PSL is capable of printing highly viscous materials and multimaterials with limited material consumption, high throughput, large printing volume (∼cm scale), and high resolution (5 µm). There is a wide tuning range of vertical printing depth created by decoupling the z-slicing of the photopolymer layer from the polymerization process parameters. To achieve this result, we characterize the z-slicing consistency, the printing resolution, and the material consumption of the system. The 3D printing process is described by structuring an ear prosthesis, a Great Wall, and architectural structures. In addition, the z-slicing of each layer can be dynamically changed from 1–200 µm during the printing process to speed up the fabrication rate by increasing the layer thickness in regions with large feature sizes. Finally, we printed complex 3D architectures with multiple materials by dispensing various polymers on the membrane successively.

2. Schematic of R2P-PSL 3D printer

The R2P-PSL system is built with the following major components (Fig. 1(a); Fig. 8 in Appendix): a UV LED light source, a projection lens, a sample stage, syringes filled with photopolymer, a doctor blade, rollers, and a flexible membrane. The schematic process of the proposed R2P additive manufacturing process can be described as “coat-expose-peel”, as in the R2P-UV nanoimprinting process (Fig. 1(b)).

 

Fig. 1. Schematic of R2P-PSL system. (a) The schematic setup of R2P-PSL system. (b) The printing procedure of R2P-PSL.

Download Full Size | PDF

In the coating process, a few drops of photopolymer are dispensed and then coated uniformly on a flexible membrane by a nozzle and a doctor blade. The thickness of the photopolymer layer varies in the range of 1–200 µm by adjusting the distance between the doctor blade and the membrane. The guide roller keeps the photopolymer-coated membrane slightly stretched and conveys the uncured photopolymer layer to the polymerization region. Then, a sample stage is moved downward, while a press roller is lifted to attach the photopolymer layer to the sample stage. To avoid crosstalk between layers, the photopolymer layer is attached to the sample without pressing.

During exposure, the uncured photopolymer layer is sandwiched by the sample stage and a quartz plate with a 1.2 µm thickness to maintain an optical flatness. The photopolymer is solidified frame-by-frame by a digital light processing based projection system, as shown in Fig. 2. The PSL system consists of a projection system and a monitoring system. There are three parts in the projection system: a light-emitting diode light for photopolymerization (405 nm, 3 W, CUN06B1B, SEOUL VIOSYS), a DMD device as a digital reflective photomask (0.55 inches, Texas Instruments), and a demagnifying lens (10X, Nikon). The DMD chip has 1920 × 1080 pixels with a pixel size of 10.8 µm; thus, the total area for one exposure is 2.1 × 1.1 mm² and the lateral printing precision can theoretically reach 1.08 µm. The monitoring system is comprised of illuminating light (650 nm, 3 mW) and a CCD camera (Daheng Optics). The printing surface can hardly be perfectly flat, so the camera is used to adjust the distance between the projection lens and the photopolymer layer prior to each exposure.

 

Fig. 2. Schematic optical design for photopolymerization

Download Full Size | PDF

After photopolymerization, the quartz plate is detached from the membrane. Then, the press roller is lowered to detach the flexible membrane from the printed structure by peeling force. The surface energy of the membrane is carefully engineered so that the solidified resin adheres to the sample, while the uncured photopolymer residue remains on the flexible membrane. Finally, the uncured photopolymer residue on the membrane is collected for recycling, while the membrane is rolled forward for the photopolymerization of the next layer. The above process is repeated until molding.

The details of the fabrication procedure are described in Fig. 3. The fabrication process starts with an STL file of a 3D model. Then, an optimized z-slicing is calculated depending on the feature size. In the next step, each slice is converted into an 8-bit PNG image file, where white corresponds to areas of exclusions and black corresponds to voids. The images are further divided into a sequence of subimages for printing. We apply a photosensitive polymer layer on a flexible membrane with a controllable thickness. When the rollers conveyed the resin layer to the light exposure region, the sample stage moved down, and a quartz plate moved up to ensure that the uncured resin was sandwiched in between. Finally, DMD loads the sequential subimages, and the exposure process starts. To avoid stitched lines between two adjacent prints, we developed an overlapping exposure method. Each print has significant structural overlap with the previous print. As a result, the structures are exposed four times, which is attributed to uniform light projection on each feature of the sample.

 

Fig. 3. Fabrication procedure

Download Full Size | PDF

A flexible membrane is adopted here as a critical material to transfer layered photopolymers. Unlike vat-based PSL, distortion or damage of the printed structures may occur during the demolding process as a result of adhesion at the surface, friction due to surface roughness or trapping of the photopolymer at the cavity. The risk of defects is increased when the size of the printed structure or the complexity of the structure increases. The demolding force between two flat, elastic, cylindrical solids in contact can be described as follows [33]:

The design of the surface energy of the conveying membrane is critical. Mismatching of the surface energies may lead to the release failure of the solidified photopolymer from the flexible membrane. Quartz glass, as a polar material with high surface energy (60.20 mJ·m⁻²), is used as the substrate for the sample stage. A nonpolar PET membrane with low surface energy (46.93 mJ·m⁻²), high chemical resistance, and high UV transmittance is adopted to convey the layered photopolymer. The typical size of the PET membrane is 500m(x) × 6.1cm(y). The adopted PET membrane is a highly transparent biaxially stretched mylar membrane with a minimum thickness of 17 µm and adequate mechanical strength. The PET membrane is further coated with a 25 nm thick water-based wax (montan wax) with low surface energy (27.96 mJ·m⁻²) to increase hydrophobicity.

In the current setup, the dispensing and coating process of the photopolymer is decoupled from the photopolymerization process. Moreover, the web speed for the flexible membrane is 20 cm/s, which means that the layered photopolymer is transported from the coating region to the polymerization region within 1∼2 sec. The duration of each exposure for photopolymerization is usually in the range of 0.02–0.2 sec. However, when the cross-section of the device is larger than the write field, patterns are written frame-by-frame. There are 4 overlapping exposures on each feature to avoid stitching error. Considering the mechanical movement of the projection lens, the fabrication speed for devices is approximately 3.25 mm²/min.

Compared with vat-based PSL, the R2P 3D printer consumes a limited amount of photopolymer. The volume of the dispensed photopolymer in each layer can be adjusted according to the cross-section of the 3D model, which means that the material consumption for microscale structures can be significantly reduced. Moreover, the unsolidified photopolymer on the membrane can be recycled by a doctor blade.

The z-slicing of each layer is mechanically controlled by adjusting the distance between the doctor blade and the membrane. The layer consistency was within 90% for a thickness larger than 7 µm (Fig. 4(a)). An increased variation was noticed when the thickness of the layered photopolymer was smaller than 7 µm. The coating inconsistency for thin photopolymer layers can be solved by the airbrush coating method or optimized mechanical design [34]. In vat-based PSL, photopolymers with low viscosity are generally employed to maintain sufficient fluidity. High-viscosity photopolymers hinder accurate layer thickness control and require more time to settle during the printing process, resulting in prolonged manufacturing time [24,26]. In R2P-PSL, the material dispensing and coating region is decoupled from the photopolymerization region. This eliminates the dwelling time and settling time in vat-based PSL. The layer consistency, depth resolution, and printing speed are all independent of the viscosity of the material. As shown in Fig. 4(b), the fabrication of a highly viscous material is enabled in R2P-PSL with high throughput and high resolution.

 

Fig. 4. Complex geometric objects fabricated with the R2P-PSL printer. (a) The dependence of layer consistency on z-slicing depth. (b) Comparison diagram of fabrication rates of photosensitive polymers of different viscosities between the R2P-PSL and vat-based SLA systems. (c) SEM micrograph of the printed kelvin foam structure. (d-f) SEM micrographs of the Great Wall. (g) Optical microscopic image of hollow cylinders with a height of 470 µm and a wall thickness of 10 µm. (h) SEM micrograph of the microneedle arrays. (i) Ear prosthesis with dimensions of 15(x) × 8(y) × 10(z) mm³ fabricated by the R2P-PSL system.

Download Full Size | PDF

To demonstrate the fabrication capability of R2P-PSL, we printed multiple samples. We designed a porous structure with a tetrakaidecahedron unit cell architecture (known as Kelvin foams), as shown in Fig. 4(c), where each layer was set to 5 µm. A porous structure with a clean appearance can be used in applications that require extremely lightweight structural elements [35]. As shown in Figs. 4(d)–4(f), the Great Wall has an overall size of 0.86(x) × 2.34(y) × 1.11 (z) mm³. The architecture began with a 100 µm thick flat pedestal fabricated at 100% LED power to ensure tight adhesion with successive building blocks. The exclusion of the z-slicing was set to 10 µm/layer. The overall fabrication time was 2 h. After exclusion, the sample was developed in absolute ethyl alcohol baths for 5 s, followed by blowing dry to remove the remaining solvent from the surface. Figure 4(g) illustrated hollow cylinders with a height of 470 µm and a wall thickness of 10 µm. One can observe a clean bottom, suggesting no residue was stained inside the hollow cylinders. Figure 4(h) illustrated microneedle arrays with a height of 570 µm and the width of 250 µm, which have good-straightness and high-sharpness (tip radius less than 6 µm). A 3D printed ear prosthesis with dimensions of 15(x) × 8(y) × 10(z) mm³ is shown in Fig. 2(i), suggesting the capability of fabricating devices with a large format.

3. Adaptive printing depth for high throughput

In general, the printing resolution is inversely related to the fabrication rate. Our printing system can reconcile this problem by providing a wide tuning range of thickness for z-slicing. As shown in Fig. 5, we enable another architecture called the crown structure with a volume of 1.4×1.4×1.1 mm³. The designed CAD model is shown in Fig. 5(a). When printed at an increased slicing depth (7 µm in Fig. 5(b), 30 µm in Fig. 5(c), and 50 µm in Fig. 5(d)), the resolution of the structure decreases, but the fabrication time was shortened from 160 min, 45 min to 31 min (blue curve in Fig. 3(e)). Alternatively, the architecture can be patterned within 105 min if the height of each layer varied from 5 µm to 50 µm in the so-called adaptive slicing printing process. The digital model was sliced at a variety of slicing depths depending on the feature size. The region with fine features in the z direction, such as balls on the crown, was sliced into much thinner layers (1–5 µm), while the parts with large features were sliced into thicker layers (5−50 µm). Figure 5(f) presented the crown that was printed at various slicing depths. It consumed 55 min less than the crown that was printed at 7 µm/layer with preserved details, such as the balls on the crown. When the slicing depth was set between 5 µm and 70 µm, there was a small deformation (<0.968). The tuning range of the slicing depth can be further increased by adopting an objective lens with a long working distance. The adaptive slicing printing process is of special interest in the fabrication of microfluidic devices, where channels can be printed at high precision and inlets/outlets can be printed at high throughput.

 

Fig. 5. Adaptive slicing printing by R2P PSL. (a) The 3D model of the crown. (b-d) SEM micrographs of the crown with z-slicing depths of 7 µm, 30 µm and 50 µm. (e) The deformation increases as the z-slicing depth increases (green line). The printing time decreases as the z-slicing depth increases (blue line). (f) The distribution of z-slicing depth at different layers for (g). (g) SEM micrograph of the crown with varied z-slicing depths (5 µm−50 µm).

Download Full Size | PDF

4. Multimaterial printing

The multinozzle dispensing mode of R2P-PSL provides a novel method for multimaterial printing with a fast speed and low material consumption compared to prior techniques [32]. Figures. 6(a)–6(d) shows a schematic diagram of the multimaterial printing process. Since layers of various curable polymers were transferred to the polymerization region by a flexible membrane, we eliminated the need for material exchange and chemical cleaning procedures in the vat. Although most of the unsolidified photopolymer was detached from the sample, there was still a small amount of residue around the newly printed structure. To avoid contamination, we adopt an additional cleaning and drying step before UV irradiation of the successive layer. Since the residue was only on the surface of the printed sample and the volume of the residue was quite limited, the cleaning/drying process can be completed within 7 s for most materials and 15 s for highly viscous materials (>3000 cP).

 

Fig. 6. Multimaterial printing by R2P-PSL. (a-d) Schematic illustration of the multimaterial printing process. (e-f) Microscope images of the structures made of transparent photopolymers stained with dyes at different concentrations. Scale bar: 500 µm. (g) Microscope image of the bagua symbol made of red and brown dye loading (2 vol. %) of transparent resin. Scale bar: 1 mm.

Download Full Size | PDF

Three types of photosensitive polymers were used in the experiments: eResin-LD1002 (eSUN); MICIOE-47S (TMTCTW); and FLFLGR02 (Formlabs). The corresponding viscosities of the polymers were 148 cP, 760 cP, and 3500 cP at 23.8°C, respectively. To demonstrate the feasibility of multimaterial 3D printing, photopolymers were stained with 2% w/w pigment for visualization. We demonstrated the flexibility of R2P-PSL for the manufacturing of multimaterial parts with both vertical and horizontal interfaces. A tree stained with three colors (white, red, dark red) was printed, as shown in Fig. 6(e), suggesting the feasibility of multimaterial interlayer printing. The height of the tree was 1.74 mm, with 360 µm for each color. As shown in Fig. 6(f), a transparent photopolymer was printed around the red cylinder to mimic the backbone and tissue structure. The famous bagua symbol, composed of two photopolymers, was intralayer fabricated, as shown in Fig. 6(g). The diameters of the symbol and the small circle inside are 3.2 mm and 400 µm, respectively. Two distinctive colors with clear boundaries were observed.

Finally, we generated a butterfly structure by photopolymers, which, when cured, exhibited different Young’s moduli as measured by a nanomechanical tester (Nano Indenter G200, Agilent): 0.11 ± 0.08 MPa for the flexible joint and 4.40 ± 0.12 GPa for the rigid region. The viscosity of the rigid photopolymer in white is 148 cP at room temperature, while the viscosity of the flexible photopolymer in gray is 3500 cP, which is more than 20-fold higher than that of the rigid photopolymer (Fig. 7). As a result, the wings of butterfly can be bent up to ±45° at the joints. By combining multiple materials with various mechanical or biological properties, one can fabricate functional devices for sensing, robots, or medical care.

 

Fig. 7. Multimaterial printing of butterflies by R2P-PSL. (a) Design of a butterfly pattern. (b) Microscope image of a butterfly made by two different materials. Scale bar: 2 mm. (c) Nanoindentation of the butterfly across the interface region. Error bars represent the standard deviation from n=3 measurements. Scale bar: 500 µm. (d) A bar chart of the viscosities of the stiff and flexible photopolymers. (e-f) A forward and backward bending image of a butterfly.

Download Full Size | PDF

5. Conclusion

The drop-dispense coating mechanism adopted in the proposed R2P-PSL system can be used to apply a noncontinuous thin film of highly viscous materials. In addition to the drop-dispense mechanism, roll coating, valve jet, or spray coating, which are commonly used in R2P nanoimprinting processes, can also be adopted in R2P-PSL systems for more efficient resist deposition with an ultrathin thickness (smaller than 100 nm).

In conclusion, we demonstrate a novel R2P-PSL approach in which layers of photopolymer are transferred and photopolymerized through a flexible membrane. Benefitting from the “coat-expose-peel” procedure, the material consumption is significantly reduced, especially when we print microstructures. Moreover, highly viscous materials can be printed quickly and with good vertical resolution. With a wide tuning range of slicing depths, the fabrication rate can be increased while maintaining reasonable resolution by optimizing each z-slicing procedure before 3D printing. Most importantly, the multinozzle dispensing method enables the fabrication of multimaterial architectures with the features of high throughput, low material consumption, and low cross-contamination. R2P-PSL provides a unique balance for flexible 3D printing in terms of minimum resolvable feature size, maximum printing volume, fabrication throughput, and material complexity. For this purpose, we envision infinite scenarios involving potential applications in bionics, biotechnology, microcircuit graphics, photonic devices, microfluidics and materials science.

Appendix

• 1. Photo of a designed R2P-PSL system.

 

Fig. 8. Photo of R2P-PSL system.

Download Full Size | PDF

Funding

Natural Science Foundation of Suzhou City (SYG201930); Basic Research Program of Jiangsu Province (BK20192003); National Natural Science Foundation of China (61975140, 62075145); Priority Academic Program Development of Jiangsu Higher Education Institutions.

Disclosures

The authors declare the following competing financial interest: W. Qiao, L. Chen, M. Zhu, and D. Pu are co-inventors on a related pending patent application.

Data availability

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

References

1. A. Accardo, M. C. Blatché, R. Courson, I. Loubinoux, C. Thibault, L. Malaquin, and C. Vieu, “Multiphoton direct laser writing and 3D imaging of polymeric freestanding architectures for cell colonization,” Small 13(27), 1700621 (2017). [CrossRef]  

2. B. Y. Ahn, E. B. Duoss, M. J. Motala, X. Guo, S. I. Park, Y. Xiong, J. Yoon, R. G. Nuzzo, J. A. Rogers, and J. A. Lewis, “Omnidirectional printing of flexible, stretchable, and spanning silver microelectrodes,” Science 323(5921), 1590–1593 (2009). [CrossRef]  

3. M. A. Luzuriaga, D. R. Berry, J. C. Reagan, R. A. Smaldone, and J. J. Gassensmith, “Biodegradable 3D printed polymer microneedles for transdermal drug delivery,” Lab Chip 18(8), 1223–1230 (2018). [CrossRef]  

4. V. Saggiomo and A. H. Velders, “Simple 3D printed scaffold-removal method for the fabrication of intricate microfluidic devices,” Adv. Sci. 2(9), 1500125 (2015). [CrossRef]  

5. S. Waheed, J. M. Cabot, N. P. Macdonald, T. Lewis, R. M. Guijt, B. Paull, and M. C. Breadmore, “3D printed microfluidic devices: enablers and barriers,” Lab Chip 16(11), 1993–2013 (2016). [CrossRef]  

6. J. T. Muth, D. M. Vogt, R. L. Truby, Y. Mengüç, D. B. Kolesky, R. J. Wood, and J. A. Lewis, “Embedded 3D printing of strain sensors within highly stretchable elastomers,” Adv. Mater. 26(36), 6307–6312 (2014). [CrossRef]  

7. A. Zolfaghari, T. Chen, and A. Y. Yi, “Additive manufacturing of precision optics at micro and nanoscale,” Int. J. Extrem. Manuf. 1(1), 012005 (2019). [CrossRef]  

8. W. Wan, W. Qiao, D. Pu, R. Li, C. Wang, Y. Hu, H. Duan, L. J. Guo, and L. Chen, “Holographic sampling display based on metagratings,” iScience 23(1), 100773 (2020). [CrossRef]  

9. J. Shi, W. Qiao, J. Hua, R. Li, and L. Chen, “Spatial multiplexing holographic combiner for glasses-free augmented reality,” Nanophotonics 9(9), 3003–3010 (2020). [CrossRef]  

10. F. Zhou, J. Hua, J. Shi, W. Qiao, and L. Chen, “Pixelated blazed gratings for high brightness multiview holographic 3d display,” IEEE Photonics Technol. Lett. 32(5), 283–286 (2020). [CrossRef]  

11. S. Shan, S. H. Kang, J. R. Raney, P. Wang, L. Fang, F. Candido, J. A. Lewis, and K. Bertoldi, “Multistable architected materials for trapping elastic strain energy,” Adv. Mater. 27(29), 4296–4301 (2015). [CrossRef]  

12. J. A. Jackson, M. C. Messner, N. A. Dudukovic, W. L. Smith, L. Bekker, B. Moran, A. M. Golobic, A. J. Pascall, E. B. Duoss, K. J. Loh, and C. M. Spadaccini, “Field responsive mechanical metamaterials,” Sci. Adv. 4(12), eaau6419 (2018). [CrossRef]  

13. Q. Wang, J. A. Jackson, Q. Ge, J. B. Hopkins, C. M. Spadaccini, and N. X. Fang, “Lightweight mechanical metamaterials with tunable negative thermal expansion,” Phys. Rev. Lett. 117(17), 175901 (2016). [CrossRef]  

14. F. Zhang, M. Wei, V. V. Viswanathan, B. Swart, Y. Shao, G. Wu, and C. Zhou, “3D printing technologies for electrochemical energy storage,” Nano Energy 40, 418–431 (2017). [CrossRef]  

15. K. Sun, T. Wei, B. Y. Ahn, J. Y. Seo, S. J. Dillon, and J. A. Lewis, “3D printing of interdigitated Li-Ion microbattery architectures,” Adv. Mater. 25(33), 4539–4543 (2013). [CrossRef]  

16. F. Ning, W. Cong, J. Qiu, J. Wei, and S. Wang, “Additive manufacturing of carbon fiber reinforced thermoplastic composites using fused deposition modeling,” Compos. B. Eng. 80, 369–378 (2015). [CrossRef]  

17. M. A. Skylar-Scott, J. Mueller, C. W. Visser, and J. A. Lewis, “Voxelated soft matter via multimaterial multinozzle 3D printing,” Nature 575(7782), 330–335 (2019). [CrossRef]  

18. N. K. Roy, D. Behera, O. G. Dibua, C. S. Foong, and M. A. Cullinan, “A novel microscale selective laser sintering (µ-SLS) process for the fabrication of microelectronic parts,” Microsyst. Nanoeng. 5(1), 1–14 (2019). [CrossRef]  

19. B. E. Kelly, I. Bhattacharya, H. Heidari, M. Shusteff, C. M. Spadaccini, and H. K. Taylor, “Volumetric additive manufacturing via tomographic reconstruction,” Science 363(6431), 1075–1079 (2019). [CrossRef]  

20. Y. Liu, H. Wang, J. Ho, R. C. Ng, R. J. H. Ng, V. H. Hall-Chen, E. H. H. Koay, Z. Dong, H. Liu, C. W. Qiu, J. R. Greer, and J. K. W. Yang, “Structural color three-dimensional printing by shrinking photonic crystals,” Nat. Commun. 10(1), 4340 (2019). [CrossRef]  

21. S. Kawata, H. B. Sun, T. Tanaka, and K. Takada, “Finer features for functional microdevices,” Nature 412(6848), 697–698 (2001). [CrossRef]  

22. M. M. Emami, F. Barazandeh, and F. Yaghmaie, “Scanning-projection based stereolithography: Method and structure,” Sens. Actuators, A 218, 116–124 (2014). [CrossRef]  

23. X. Chen, W. Liu, B. Dong, J. Lee, H. O. T. Ware, H. F. Zhang, and C. Sun, “High-speed 3D printing of millimeter-size customized aspheric imaging lenses with sub 7 nm surface roughness,” Adv. Mater. 30(18), 1705683 (2018). [CrossRef]  

24. Y. Luo, G. Le Fer, D. Dean, and M. L. Becker, “3D Printing of poly(propylene fumarate) oligomers: evaluation of resin viscosity, printing characteristics and mechanical properties,” Biomacromolecules 20(4), 1699–1708 (2019). [CrossRef]  

25. S. Y. Song, M. S. Park, D. Lee, J. W. Lee, and J. S. Yun, “Optimization and characterization of high-viscosity ZrO2 ceramic nanocomposite resins for supportless stereolithography,” Mater. Des. 180, 107960 (2019). [CrossRef]  

26. J. Lee, Y. Lu, S. Kashyap, A. Alarmdari, M. F. Emon, and J. Emon, “Liquid bridge microstereolithography,” Addit Manuf. 21, 76–83 (2018). [CrossRef]  

27. J. R. Tumbleston, D. Shirvanyants, N. Ermoshkin, R. Janusziewicz, A. R. Johnson, D. Kelly, K. Chen, R. Pinschmidt, J. P. Rolland, A. Ermoshkin, E. T. Samulski, and J. M. DeSimone, “Continuous liquid interface production of 3D objects,” Science 347(6228), 1349–1352 (2015). [CrossRef]  

28. Y. Li, H. Mao, P. Hu, M. Hermes, H. Lim, J. Yoon, M. Luhar, Y. Chen, and W. Wu, “Bioinspired functional surfaces enabled by multiscale stereolithography,” Adv. Mater. Technol. 4(5), 1800638 (2019). [CrossRef]  

29. D. A. Walker, J. L. Hedrick, and C. A. Mirkin, “Rapid, large-volume, thermally controlled 3D printing using a mobile liquid interface,” Science 366(6463), 360–364 (2019). [CrossRef]  

30. Q. Ge, A. H. Sakhaei, H. Lee, C. K. Dunn, N. X. Fang, and M. L. Dunn, “Multimaterial 4D printing with tailorable shape memory polymers,” Sci Rep 6(1), 31110 (2016). [CrossRef]  

31. D. Han, C. Yang, N. X. Fang, and H. Lee, “Rapid multi-material 3D printing with projection micro-stereolithography using dynamic fluidic control,” Addit Manuf. 27, 606–615 (2019). [CrossRef]  

32. F. Mayer, S. Richter, J. Westhauser, E. Blasco, C. Barner-Kowollik, and M. Wegener, “Multimaterial 3D laser microprinting using an integrated microfluidic system,” Sci Adv. 5(2), eaau9160 (2019). [CrossRef]  

33. H. M. Pollock, D. Maugis, and M. Barquins, “Force of adhesion between solid-surfaces in contact,” Appl. Phys. Lett. 33(9), 798–799 (1978). [CrossRef]  

34. W. Qiao, W. Huang, Y. Liu, X. Li, L. S. Chen, and J. X. Tang, “Toward scalable flexible nanomanufacturing for photonic structures and devices,” Adv. Mater. 28(47), 10353–10380 (2016). [CrossRef]  

35. X. Zheng, H. Lee, T. H. Weisgraber, M. Shusteff, J. DeOtte, E. B. Duoss, J. D. Kuntz, M. M. Biener, Q. Ge, J. A. Jackson, S. O. Kucheyev, N. X. Fang, and C. M. Spadaccini, “Ultralight, ultrastiff mechanical metamaterials,” Science 344(6190), 1373–1377 (2014). [CrossRef]