Open Access
Add to BibSonomy
Abstract
We propose a simple autofocusing technique that can be introduced into conventional two-photon lithography systems without additional devices. Autofocusing is achieved by image processing using transmission images of photopolymerized voxels. The signal-to-noise ratio of transmission images was improved by optimal low-pass filtering to detect voxels in them. The focal point was detected with an accuracy of about 250 nm from the difference images. Further, we demonstrated mass-fabrication of a 5 × 5 spiral square array with an area of 900 × 900 µm2 using this method. The method has potential application in constructing low-cost, compact and versatile two-photon lithography apparatus.
© 2020 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
In recent years, various types of three-dimensional (3D) printing technologies have been developed and widely used for prototyping 3D models and in the manufacture of products made from polymers [1,2], metals [3,4], ceramics [5,6] and others [7–9]. 3D printing has evolved into a manufacturing technology called “additive manufacturing” and the importance of automation and mass production is increasing. Among these technologies, two-photon lithography, which involves laser direct writing of photopolymerizable materials, provides the highest resolution of submicron order owing to the nonlinearity of two-photon absorption of photopolymerizable materials [10–14]. Such high-resolution 3D printing techniques have widely been applied to produce functional microdevices such as micro optical components [15,16], waveguides [17], photonic metamaterials [18], mechanical metamaterials [19], micromachines [20], microfluidic devices [21] and scaffolds [22]. These 3D microstructures are fabricated by scanning the laser focus inside photocurable resins and curing the resin to obtain cross-sectional layers stacked sequentially from the surface of the substrate. The focal spot at the starting point of fabrication must be on the surface of the substrate with a minimum accuracy of less than 1 µm due to submicron resolution in depth [23,24]. This is also necessary for the mass production and large-scale fabrication of the 3D microstructures in a large area over the field of view of an objective lens. Thus, autofocusing techniques are necessary for practical applications of large-scale two-photon lithography systems.
Various autofocusing methods have been developed to realize automation for microscopic observation and microscale fabrication using optical microscopy [25–27]. For example, Liu et al. developed an optical-based autofocusing microscope achieving autofocus accuracy less than 1 µm with a large linear autofocusing range and a rapid response using additional two optical paths [25,26]. Zhang et al. developed an automatic focus detection system with multi-point laser differential confocal autofocus method [27] for improving the focusing accuracy. In this method, a reference laser reflected by the sample is detected by multiple sensors around the confocal point to improve the autofocus accuracy. Their method provides an accuracy of 0.2 µm with an objective lens of numerical aperture (NA) of 0.55. However, since these autofocusing methods require additional devices such as image sensors and lasers, adding an autofocus device increases both the size and the cost of the two-photon lithography system.
As an alternative approach to implement autofocus function into a two-photon lithography system, the direct detection of a polymerized voxel was employed to realize simple, low-cost fabrication systems with autofocusing [28–33]. For example, Woggon et al., utilized the fluorescence signal from a photosensitive material (Ormocomp, Microresist Technology GmbH, Berlin, Germany) to find the position of the interface between the glass substrate and the Ormocomp layer for laser patterning of grating structures [28]. They also demonstrated a hybrid lithographic technique combining ultraviolet (UV) exposure and two-photon lithography by means of 3D position alignment of a femtosecond pulsed laser beam using the fluorescence signal from a preformed 3D structure doped with a fluorescent dye [29]. Furthermore, Jung et al. developed an autofocus method by detecting the two-photon fluorescence emitted from a cured photopolymer (SCR-500, JSR Corp., Tokyo, Japan) around the two-photon absorption region [31]. In this system, the photopolymer emits two-photon fluorescence from the area where the resin is cured by two-photon absorption above the polymerization threshold. Autofocusing is demonstrated by detecting the fluorescence with an accuracy of -100 nm to +200 nm relative to the substrate surface. Jeon, et al. improved the positioning accuracy to ±0.045 µm using normalized image size, which was calculated using the second momentum radius (SMR) of two-photon induced fluorescence (TPIF) [32]. The fluctuation of the TPIF intensity is reduced via the use of the normalized fluorescence images. Although these techniques for the direct detection of a voxel using the fluorescence signal of photosensitive materials has high accuracy without problems arising from the flatness and thickness deviation of substrates, they can be applied only to materials that produce two-photon induced fluorescence high enough to be detected by a charge-coupled device (CCD). In addition, since the fluorescence intensity depends on irradiated laser power, it has to be normalized to achieve high accuracy. Zheng et al. proposed and demonstrated an autofocusing method using transmission optical microscopic images of four spiral squares fabricated with different heights at corners in fabrication area [33]. In their method, the focus position is determined by the number of visible lines consisting of each spiral square at the corners. Thus, the photosensitive materials used in this method are not limited to specific materials suitable for TPIF. Using this versatile autofocusing method, they demonstrated fabrication of microstructures with the average deviation of 2.3% in height. However, since the transmission optical microscopic images are unclear due to the non-uniform light intensity of the light-emitting diode (LED) illumination, the accuracy of this approach using transmission optical microscopic images could be further improved using optimal image processing.
Here, we propose a simple autofocusing method using an image processing of the transmission optical microscope images of a voxel. In this method, focal point is determined by the difference image of the transmission images of a voxel cured by multiple point exposure while changing the focal position along the optical axis. When the focal point reaches the surface of the substrate by lowering the sample stage, a cured photopolymer is created and appears in the transmission image due to the change in refractive index. The back ground noise of transmission images is improved by low pass filtering to detect the tip of the cured voxel. The optimal low-pass filtering enables us to detect the position of focal point with an accuracy of about 250 nm from the difference image of the transmission images whose noise was reduced via optimal low-pass filtering. Additionally, we demonstrated the mass-production of micropatterns with an area of 900 × 900 µm2 by repeating both autofocusing and laser drawing. Unlike the conventional methods, since this method allows autofocusing without additional devices, it is useful for constructing low-cost, compact and versatile two-photon lithography apparatus.
2. Autofocusing by imaging processing
Our autofocusing technique for a two-photon lithography system is based on a straightforward algorithm using frequency filtering and difference image processing in a region-of-interest (ROI) of the obtained transmission images. First, as shown in Fig. 1, the focal point of the femtosecond (fs) laser beam, located inside the glass substrate, is moved towards the surface on the substrate stepwise while the fs laser beam was irradiated at each position for a certain exposure time. Transmission images were taken at each position. When the focal point is approaches the surface, the photopolymer is cured by two-photon absorption. The cured photopolymer (voxel), located on the surface of the substrate appears in the transmission image. Here, the total intensity difference (subtraction of images between before and after moving the focal point) in ROIs of the images captured increases, and when the intensity difference is larger than a predetermined threshold, focusing at the surface of the substrate is detected. Because the change in refractive index of the photopolymer before and after polymerization is as small as approximately 0.04 [34], the background noise of the transmission image of the voxel is too high to be detected directly by the appearance of the tip of the voxel using original transmission images without any image processing. This is due to the low contrast of the voxel image. This was addressed by using a low-pass filter to detect the voxel less than 1 µm in size as accurately as possible. This process is critical in the detection of voxels because the transmission images often have high frequency noise from image sensor, especially in case of low intensity illumination. This additional image processing enables the highly sensitive detection of voxels using a simple algorithm.
Fig. 1. Schematic diagram of autofocusing by simple imaging processing. The focal point is moved from inside the substrate through the photopolymer stepwise, and optical microscopic images are obtained at each position. The position of the surface of the substrate is detected by the intensity difference in ROI between before and after stepwise moving. To reduce the background noise in the obtained images, a low-pass filter was applied to each image.
Transmission images were obtained at each position while moving the focal spot by 100 nm stepwise. The voxel exposure time was 500 ms. For the low-pass filter, the discrete two-dimensional Fourier transform function in OpenCV 2.4.13.4 was used to remove high frequency noise in the images. The extraction of the spatial frequency components in the image was carried out using cos 2π (ux + vy) (u, v ≦ 1/5). Since a smaller ROI is better for removing unnecessary noise during the subtraction of two images, the ROI was set to 50 × 50 pixel square. In our current setup, the processing time of autofocus judgement of each image needs about 40 ms. For example, it takes 33 seconds for total autofocusing process including exposure time, image capture, and stage movement within the range of 5 µm along the optical axis.
3. Two-photon lithography system with autofocusing function
A laboratory-made two-photon lithography system was used for the demonstration of our autofocusing method. As shown in Fig. 2, the light path for fabrication is represented by the red solid line, while the light path of the observation part is represented by the dotted line.
Fig. 2. Optical setup for a two-photon lithography system for demonstrating autofocusing. The polymer structure was fabricated on a cover glass by scanning the fs laser in the photopolymer, and the transmitted images were captured by a CCD camera.
A fs pulse laser having a pulse width of 100 fs with a wavelength of 752 nm (Mai Tai HP, Spectra-Physics, Inc.) was used for the polymerization of the photopolymer (SR499 Sartomer Inc.) including the photoinitiator (Diphenyl(2,4,6-trimethylbenzoyl) phosphine oxide, Sigma-Aldrich Corp.). The incident laser power before the objective lens was set to 20 mW with a neutral density filter. The laser beam, which was expanded using a beam expander, was focused into the glass substrate using an objective lens whose NA is 1.4 (UPLSAPO100XO, Olympus Corp.). The focal point in the glass substrate was moved towards surface of the substrate using a 3D piezoelectric stage (P-563, E-727, PI Corp.). This system also includes observation optics equipped with a charge-coupled device (CCD) camera (XC-ST50, Sony Imaging Products & Solutions Inc.) to observe the cured voxel. A light emitting diode (LED) light source (SZ2-CLS, Olympus Corp.) with a color filter (R-60, Hayashi Watch Ind., Corp.) was used to prevent the curing of the photopolymer, and a laser cut filter (#47-290, Edmund Optics Inc.) was used after the imaging lens (focal length: 400 mm). The images captured by the CCD camera were transferred to a computer by an industrial analog frame grabber board (DFG/SV1/PCIe, The Imaging Source Asia Corp.). The fabrication system was controlled using homebuilt software. A 3-axis linear motor stage (ONE-XY100HA and VP-5ZA, Newport Corp.) was used for mass fabrication.
4. Results and discussion
4.1 Detection of voxels from optical microscopic images
We investigated the size of the voxel that could be detected by the autofocusing method. First, we prepared the voxels at every position at intervals of 100 nm along the optical axis, and transmission images of the voxel were observed by the observation optics included in our two-photon lithography system shown in Fig. 2. The exposure time of each focal position was 500 ms at a laser power of 20 mW. It was confirmed that the resultant voxel was approximately 0.5 µm wide and 1.4 µm high under this exposure condition. After washing off the unpolymerized resins with an alcohol solvent (SOLFIT, 3-methoxy-3-methyl-1-butanol, Kuraray CO., Ltd., Tokyo, Japan), the voxels formed are also observed by scanning electron microscopy (SEM; JSM-6390LV, JEOL Ltd.) (Fig. 3). The heights of the voxels calculated by 45° tilted SEM images were shown in the SEM images. Here, when the Z position of the focus moves from 1100 nm to 1200 nm, the curing height was larger when compared to other steps. This is mainly caused by effects such as self-focusing due to the lens effect of the previously cured voxel [34] and the threshold of two-photon polymerization [24]. As a result, when the height of the voxel is 200 nm or less, no voxel could be detected in the transmission images without a low-pass filter. However, with a proper low-pass filter, the background noise of images decreased, and the 200 nm voxel was slightly observed in the image. The SEM and transmission images show that voxels higher than about 200 nm appear in transmission image under our experimental conditions.
Fig. 3. Optical microscopic images and SEM images of voxels obtained at different positions of the focal spot. In the SEM images, the heights of the voxels are shown. The voxel was prepared by moving the laser beam upward by steps of 100 nm and the irradiation time of the laser beam was 500 ms. Although the background noise of the original optical images was high, it was improved with low-pass filter.
Next, we confirmed that the appearance of voxels could be determined automatically by comparing the total intensity difference in ROIs at each focal position. The total light intensity was obtained by integrating the light intensity value of each pixel at each position, and the total intensity difference was calculated as the difference between the total light intensity value at the current position and the reference value when no voxel appeared at the surface of the glass substrate when focusing the fs laser beam inside the glass substrate. Figure 4 shows the total intensity difference at each focal position, with and without low pass filtering. Without the low-pass filter, the difference of the total intensity value in the ROIs hardly changed even if the voxel appeared on the substrate. However, with the low-pass filter, the difference clearly increased when the voxel height was 200 nm or greater. In this case, when the height of the voxel was 250 nm, it could be detected automatically by monitoring the difference between the total light intensity values with the low-pass filter. This demonstrates that autofocusing with an accuracy of about 250 nm, applicable to two-photon lithography, could be achieved by simple image processing.
Fig. 4. Total light intensity difference as a function of the position of the focal spot. SEM images of the voxel formed at each position are also shown in the figure. The reduction of high-frequency noise by the low-pass filter improves the background noise of the image, and increases the difference in intensity of ROI before and after movement, thereby realizing autofocusing.
In addition, based on the total light intensity difference, we set the threshold for determining the autofocus position, and confirmed the autofocus algorithm via the following procedure. At each step, after exposure to the fs laser for a certain exposure time, a transmission image of a voxel is obtained at each focal position, and the total light intensity value of the ROI is calculated with the low pass filter. Next, the total light intensity value is compared with the first one. This procedure is repeated while lowering the sample stage (moving the focal spot upward). Finally, when the difference of the total light intensity value reaches the threshold, the autofocusing procedure is stopped. At this time, it is recognized that the voxel appears, and the 3D fabrication of a microstructure has been achieved. The threshold used was determined to be about 1.5 to 2.0 × 104 (Signal intensity of CCD) from the average of several experimental results.
4.2 Demonstration of fabrication with autofocusing
As the first demonstration of the autofocusing, we fabricated square structures at an autofocused position as shown in Fig. 5. Each side of the three squares is 100, 200 and 270 µm, respectively. Autofocusing was performed at the center of the square. The laser power for autofocusing and fabrication was set to 20 mW, and fabrication speed was set to 100 µm/s. As a result, we succeed in fabrication at the autofocus position. In Fig. 5, the voxel created during autofocusing cannot be observed at the center of the squares because the size of the voxel is too small to observe at this magnification. In our method, the position of the voxel can be adjusted to a place that has little effect on the target structure. The voxel can be also included in a part of the target structure when the size of the voxel is smaller than the target structure unlike previous method using a spiral structure [33]
In experiments, success of the autofocus strongly depends on the image contrast of voxel without the low-pass filter. Stray light entering the CCD camera reduces the image contrast and prevents voxel detection. In such cases, the low-pass filter does not improve the contrast sufficiently and the difference in intensity does not reach the threshold. Consequently, it is important to ensure uniform transmission illumination without stray light to successfully use autofocusing for fabrication. In our experiments, to prevent entering stray light in image sensor, the light pass from the imaging lens to the CCD camera was covered.
To demonstrate the validity of our autofocusing technique for the mass-production of microstructures at large scale, we fabricated a spiral square array by repeated autofocusing while moving the focal spot in-plane direction. Each spiral square was fabricated by single scanning of the fs laser beam with the accumulation of two layers at a distance of 0.5 µm. For this experiment, a 3-axis linear motor stage (ONE-XY100HA and VP-5ZA, Newport Corp.) was used for a large-scale fabrication instead of a 3D piezoelectric stage. The linear motor stage can be moved over a wider range (XY stroke: 90 mm, Z stroke: 4.8 mm) than piezoelectric stage systems (XYZ stroke: 300 µm). The detailed geometry of the array structure is shown in Fig. 6(a), and we succeeded in fabricating a 5 × 5 structure array in 900 × 900 µm2 area, as shown in Fig. 6(b). In this experiment, we used a cover glass that is normally used in clinical tests (C024401, Matsunami Glass Ind. Ltd., size: 24 × 40 mm, thickness: 0.13–0.17 mm, flatness: < 0.010 mm,). Because the cover glass has thickness errors and insufficient flatness when compared to the voxel size of two-photon photopolymerization, many structures in a large area could not be fabricated without autofocusing. Figure 6(c) compares the fabrication results of 5 × 5 structure arrays with and without autofocusing. The results demonstrate that the use of our autofocusing technique enables the production of multiple microstructures at a large scale of approximately 1 mm2 square without peeling off from the substrate. By using the current experimental setup and image processing parameters such as threshold and cut-off frequency, we experimentally confirmed that the success rate of autofocusing when fabricating a 10 × 10 structure array is 98%.
Fig. 6. Microstructures produced by mass production. (a) The detailed geometry of the spiral square array. It was fabricated along the route indicated by the red arrow. (b) Optical microscope image of the fabricated array. (c) Demonstration of fabrication of microstructure array with and without autofocusing.
5. Conclusion
We have thus proposed and demonstrated a simple autofocusing method for large-scale two-photon lithography. The voxels which appeared at the surface of a glass substrate by the movement of the focal point was detected by monitoring the subtraction of the total intensity values of the ROI in the images using the current position and the first position. In addition, a low-pass filter has been applied to the transmitted image to improve the contrast of the images and detect even smaller voxels. In the experiment, we confirmed the formation of voxels by moving the focal point stepwise, from inside the glass substrate to the surface of the photocurable resin. Subsequently, using an appropriate threshold for voxel detection based on the difference in intensity values, the appearance of the voxel was successfully detected. As a result, we succeeded in the automatic detection of voxels with an accuracy of about 250 nm. In addition, the use of the developed autofocusing technique and a linear motor stage system made it possible to fabricate 5 × 5 structural arrays of spiral squares over a large area of 900 × 900 µm2. This method allows autofocusing without any additional devices, which helps to build a low-cost, compact two-photon lithography system. In the near future, this autofocusing method could be utilized to correct the inclination of substrate by measuring at multiple points at the corners of a large fabrication area. Thus, this simple method will be useful for large-scale two-photon lithography systems not only for the mass-production of 3D microstructures but also fabrication of large-scale 3D structures.
Funding
Japan Society for the Promotion of Science (18K13667, 19H02039); Core Research for Evolutional Science and Technology (JPMJCR1905).
Disclosures
The authors declare no conflicts of interest.
References
1. F. P. W. Melchels, J. Feijen, and D. W. Grijpma, “A review on stereolithography and its applications in biomedical engineering,” Biomaterials 31(24), 6121–6130 (2010). [CrossRef]
2. S. C. Ligon, R. Liska, J. Stampfl, M. Gurr, and R. Mulhaupt, “Polymers for 3D Printing and Customized Additive Manufacturing,” Chem. Rev. 117(15), 10212–10290 (2017). [CrossRef]
3. W. E. Frazier, “Metal Additive Manufacturing: A Review,” J. Mater. Eng. Perform. 23(6), 1917–1928 (2014). [CrossRef]
4. L. Hirt, A. Reiser, R. Spolenak, and T. Zambelli, “Additive Manufacturing of Metal Structures at the Micrometer Scale,” Adv. Mater. 29(17), 1604211 (2017). [CrossRef]
5. Z. W. Chen, Z. Y. Li, J. J. Li, C. B. Liu, C. S. Lao, Y. L. Fu, C. Y. Liu, Y. Li, P. Wang, and Y. He, “3D printing of ceramics: A review,” J. Eur. Ceram. Soc. 39(4), 661–687 (2019). [CrossRef]
6. N. Travitzky, A. Bonet, B. Dermeik, T. Fey, I. Filbert-Demut, L. Schlier, T. Schlordt, and P. Greil, “Additive Manufacturing of Ceramic-Based Materials,” Adv. Eng. Mater. 16(6), 729–754 (2014). [CrossRef]
7. J. Y. Lee, J. An, and C. K. Chua, “Fundamentals and applications of 3D printing for novel materials,” Appl Mater Today. 7, 120–133 (2017). [CrossRef]
8. R. D. Farahani, M. Dube, and D. Therriault, “Three-Dimensional Printing of Multifunctional Nanocomposites: Manufacturing Techniques and Applications,” Adv. Mater. 28(28), 5794–5821 (2016). [CrossRef]
9. F. Momeni, S. M. M. Hassani.N, X. Liu, and J. Ni, “A review of 4D printing,” Mater. Des. 122, 42–79 (2017). [CrossRef]
10. S. Maruo, O. Nakamura, and S. Kawata, “Three-dimensional microfabrication with two-photon absorbed photopolymerization,” Opt. Lett. 22(2), 132–134 (1997). [CrossRef]
11. C. N. LaFratta, J. T. Fourkas, T. Baldacchini, and R. A. Farrer, “Multiphoton fabrication,” Angew. Chem., Int. Ed. 46(33), 6238–6258 (2007). [CrossRef]
12. C. Barner-Kowollik, M. Bastmeyer, E. Blasco, G. Delaittre, P. Muller, B. Richter, and M. Wegener, “3D Laser Micro- and Nanoprinting: Challenges for Chemistry,” Angew. Chem., Int. Ed. 56(50), 15828–15845 (2017). [CrossRef]
13. M. Malinauskas, M. Farsari, A. Piskarskas, and S. Juodkazis, “Ultrafast laser nanostructuring of photopolymers: A decade of advances,” Phys. Rep. 533(1), 1–31 (2013). [CrossRef]
14. J. Fischer and M. Wegener, “Three-dimensional optical laser lithography beyond the diffraction limit,” Laser Photonics Rev. 7(1), 22–44 (2013). [CrossRef]
15. T. Gissibl, S. Thiele, A. Herkommer, and H. Giessen, “Two-photon direct laser writing of ultracompact multi-lens objectives,” Nat. Photonics 10(8), 554–560 (2016). [CrossRef]
16. P. I. Dietrich, M. Blaicher, I. Reuter, M. Billah, T. Hoose, A. Hofmann, C. Caer, R. Dangel, B. Offrein, U. Troppenz, M. Moehrle, W. Freude, and C. Koos, “In situ 3D nanoprinting of free-form coupling elements for hybrid photonic integration,” Nat. Photonics 12(4), 241–247 (2018). [CrossRef]
17. N. Lindenmann, G. Balthasar, D. Hillerkuss, R. Schmogrow, M. Jordan, J. Leuthold, W. Freude, and C. Koos, “Photonic wire bonding: a novel concept for chipscale interconnects,” Opt. Express 20(16), 17667–17677 (2012). [CrossRef]
18. M. S. Rill, C. Plet, M. Thiel, I. Staude, G. Von Freymann, S. Linden, and M. Wegener, “Photonic metamaterials by direct laser writing and silver chemical vapour deposition,” Nat. Mater. 7(7), 543–546 (2008). [CrossRef]
19. T. Buckmann, N. Stenger, M. Kadic, J. Kaschke, A. Frolich, T. Kennerknecht, C. Eberl, M. Thiel, and M. Wegener, “Tailored 3D Mechanical Metamaterials Made by Dip-in Direct-Laser-Writing Optical Lithography,” Adv. Mater. 24(20), 2710–2714 (2012). [CrossRef]
20. S. Maruo, K. Ikuta, and H. Korogi, “Force-controllable, optically driven micromachines fabricated by single-step two-photon micro stereolithography,” J. Microelectromech. Syst. 12(5), 533–539 (2003). [CrossRef]
21. S. Maruo and H. Inoue, “Optically driven micropump produced by three-dimensional two-photon microfabrication,” Appl. Phys. Lett. 89(14), 144101 (2006). [CrossRef]
22. E. D. Lemma, B. Spagnolo, M. De Vittorio, and F. Pisanello, “Studying Cell Mechanobiology in 3D: The Two-Photon Lithography Approach,” Trends Biotechnol. 37(4), 358–372 (2019). [CrossRef]
23. S. Kawata, H. B. Sun, T. Tanaka, and K. Takada, “Finer features for functional microdevices - Micromachines can be created with higher resolution using two-photon absorption,” Nature 412(6848), 697–698 (2001). [CrossRef]
24. X. Q. Zhou, Y. H. Hou, and J. Q. Lin, “A review on the processing accuracy of two-photon polymerization,” AIP Adv. 5(3), 030701 (2015). [CrossRef]
25. C. S. Liu, P. H. Hu, and Y. C. Lin, “Design and experimental validation of novel optics-based autofocusing microscope,” Appl. Phys. B: Lasers Opt. 109(2), 259–268 (2012). [CrossRef]
26. C. S. Liu and S. H. Jiang, “Design and experimental validation of novel enhanced-performance autofocusing microscope,” Appl. Phys. B: Lasers Opt. 117(4), 1161–1171 (2014). [CrossRef]
27. X. B. Zhang, F. M. Fan, M. Gheisari, and G. Srivastava, “A Novel Auto-Focus Method for Image Processing Using Laser Triangulation,” IEEE Access 7, 64837–64843 (2019). [CrossRef]
28. T. Woggon, T. Kleiner, M. Punke, and U. Lemmer, “Nanostructuring of organic-inorganic hybrid materials for distributed feedback laser resonators by two-photon polymerization,” Opt. Express 17(4), 2500–2507 (2009). [CrossRef]
29. C. Eschenbaum, D. Grossmann, K. Dopf, S. Kettlitz, T. Bocksrocker, S. Valouch, and U. Lemmer, “Hybrid lithography: Combining UV-exposure and two photon direct laser writing,” Opt. Express 21(24), 29921–29926 (2013). [CrossRef]
30. M. Malinauskas, A. Žukauskas, and K. Belazaras, “Employment of fluorescence for autofocusing in direct laser writing micro-/nano-lithography,” Proc. SPIE 9192, 919212 (2014). [CrossRef]
31. B. J. Jung, H. J. Kong, B. G. Jeon, D. Y. Yang, Y. Son, and K. S. Lee, “Autofocusing method using fluorescence detection for precise two-photon nanofabrication,” Opt. Express 19(23), 22659–22668 (2011). [CrossRef]
32. B. G. Jeon, B. J. Jung, H. J. Kong, and Y. H. Cho, “Precise autofocus method employing normalized fluorescence image size in a two-photon polymerization nanofabrication system,” Appl. Opt. 54(24), 7323–7329 (2015). [CrossRef]
33. X. Zheng, K. Cheng, X. Q. Zhou, J. Q. Lin, and X. Jing, “A method for positioning the focal spot location of two photon polymerization,” AIP Adv. 7(9), 095318 (2017). [CrossRef]
34. A. S. Kewitsch and A. Yariv, “Self-focusing and self-trapping of optical beams upon photopolymerization,” Opt. Lett. 21(1), 24–26 (1996). [CrossRef]
