Open Access
Add to BibSonomy
Abstract
We report fabrication of three-dimensional (3D) high aspect ratio microchannels from the rear-surface of thick silica glass with femtosecond laser pulses. To avoid nonlinear self-focusing in the conventional focusing scheme which leads to intensity clamping in the interaction of the ultrafast laser pulses with the silica glass, simultaneous spatiotemporal focusing is employed which gives rise to high aspect ratio (∼30) straight and curved microchannels in 10 mm thick silica glass substrates.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Femtosecond lasers have been proven an effective means for micromachining of transparent materials [1,2]. As one of the key applications, fabrication of microchannels with femtosecond laser pulses has attracted much attention for its potential in enabling innovative microfluidic devices composed of embedded three-dimensional (3D) fluidic networks, such as liquid or gas chromatography, electrophoresis, microreactors, and micro-Total Analysis System (micro-TAS) systems [3]. To this end, high-aspect-ratio 3D microchannels are frequently required [4]. In certain types of transparent materials including silica glass and YAG crystal, the microchannels can be fabricated by femtosecond laser assisted chemical etching (FLACE) [5–10]. Nevertheless, the FLACE is material dependent and inherently of a limited etching selectivity which gives rise to a tapered feature for long channels. As an alternative to the FLACE, 3D microfluidic channels can also be fabricated in transparent materials by femtosecond laser 3D drilling in a single step without any post treatment [11–14]. This approach is material insensitive as it does not require an etching selectivity induced by the femtosecond laser irradiation. Previously, Li et al. introduced water-assisted, rear-surface microfabrication of 3D channels in silica glass and drilled high aspect ratio structures on the back surface of 1 mm thick glass using an objective lens with a relatively high numerical aperture (NA) of 0.55 [11]. Later on, Hwang et al. applied liquid and ultrasonic waves to drill straight and 3D microchannels in a 1.25 mm thick glass at the NA = 0.42 [14]. The high NA objectives can help achieve a high peak intensity to induce strong ablation which is necessary for generating gas bubbles in water for clearing up the debris produced during the laser drilling.
With the development of microfluidics, the devices tend to become more and more complex to achieve higher functionalities, making it desirable to realize high-aspect-ratio 1D and 3D channels in thick samples. Unfortunately, as low NA focal lenses must be employed for focusing deeply into the thick samples, nonlinear self-focusing takes effect resulting in decrease of the peak intensity due to the intensity clamping [15]. It has been shown that the intensity clamping can be effectively overcome with the simultaneous spatiotemporal focusing (SSTF) scheme [16]. The working mechanism of SSTF has been described in detail elsewhere [17,18]. Briefly, in the SSTF, the incident femtosecond pulse must first pass through a pair of gratings, separating the different frequency components from each other in space before impinging on the objective lens. After the objective lens, the different spectral components overlapped spatially and temporally only at the geometric focus of the focusing lens, which restores the shortest pulse duration to generate the highest peak intensity only at the focus. The extended pulse durations before and after the geometric focus effectively reduce the peak intensity and in turn suppress the intensity clamping caused by the nonlinear self-focusing. In the current work, we demonstrate drilling of high-aspect-ratio 1D and 3D microchannels in 10 mm thick silica glass with the SSTF scheme, which is unable to achieve with the conventional focusing (CF) scheme.
2. Experimental
Figure 1 schematically illustrates the experimental setup. A femtosecond laser amplifier (Libra-HE, Coherent Inc.) delivered uncompressed 800 nm femtosecond pulses with a spectral bandwidth of 27 nm and a maximum pulse energy of ∼4 mJ at 1 kHz repetition rate. The initial pulse duration was ∼200 ps. The average laser power was controlled by a variable neutral density filter (VF). A telescope system consisting of a convex lens and a concave lens was used to reduce the beam width for implementing the SSTF scheme. The beam then passed through a single-pass, double grating compressor consisting of two σ = 1500 grooves/mm gratings, blazed for the incident angle of 53°. The distance between the two gratings was adjusted to be ∼730 mm for compensating the temporal dispersion of the uncompressed pulses. After being dispersed by the grating pair, the spatially dispersed laser pulses were reflected by a dichroic mirror and then focused into a 10-mm-thickness fused silica glass sample using a long-working-distance objective lens (Leica, 2 ×, NA = 0.35, working distance, 20 mm). The pulse duration at the focus in SSTF scheme was ∼40 fs, corresponding to the transform limited duration of a pulse with a spectral bandwidth of 27 nm. The sample was optically polished on both the top and bottom sides. In the 3D laser drilling process, the glass sample was translated by a PC-controlled XYZ stage (Zaber Technologies Inc.) with a resolution of 1 µm. The rear surface of the glass sample was immerged in purified water contained in a glass cuvette. The fabrication process was monitored by a CCD camera in real time.
Fig. 1. Schematic illustration of the experimental setup. VF: variable neutral density filter. G1-2: gratings. L1, L2, L3: lenses. DM: dichroic mirror. OL: objective lens. GC: glass cuvette. SA: sample. LA: lamp.
3. Results and discussions
Previously, Vitek et al. have shown that a better confinement of pulse energy along the axial direction can be achieved in a 6 mm thick silica glass with the SSTF scheme than the CF [19]. We observe the similar phenomenon as shown in Fig. 2. The experiment was carried out by focusing the femtosecond laser beam at the rear-surface of a 10 mm thick silica glass at the same pulse energy of 100 µJ and an NA of 0.02. As shown in Fig. 2(a) which is obtained with the CF scheme, a bright light filament caused by the balance between nonlinear self-focusing and plasma defocusing is clearly visible in silica glass. In contrast, by replacing the CF with the SSTF scheme, the formation of filament is suppressed and plasma formation is localized near the focal plane at the rear surface of silica glass sample as shown in Fig. 2(b). The filament generated with the CF scheme was quite long along the propagation axis, which hamper the precise energy deposition and material ablation. On the contrary, with SSTF scheme, the formation of filament was suppressed and plasma was well localized at focal region, benefiting the efficiently ablation on the back surface of the sample. Below, we compare the water-assisted drilling in silica glass with femtosecond laser pulses underwent the two focusing schemes.
Fig. 2. Images and schematic illustrations (Inserts) of the optical breakdown with (a) CF scheme and (b) SSTF scheme when femtosecond laser pulses are focused on the rear surface of a 10 mm thick silica glass at 0.02 NA. Pulse energy: 100 µJ.
3.1 Drilling of the 1D microchannel along the vertical direction from the rear surface
To experimentally demonstrate the capability of efficient drilling in thick silica glass with the SSTF scheme, we compared the laser drilled channels generated with the CF (Fig. 3(a-f)) and SSTF (Fig. 3(g-l)) schemes for the same machining parameters. In both schemes, the beam had a nearly Gaussian profile with a diameter of ∼9 mm (1/e2) and the effective NA estimated in both schemes was about 0.06. At the beginning of the experiment, the femtosecond laser beam was focused at the rear surface of silica glass which was in direct contact with purified water. We then translated the focal point in the upward direction at a constant speed using the motorized micro-stage. The writing speed was fixed at 5 µm/s and the pulse energy was varied from 6 µJ to 80 µJ. The measurement of the fabricated microchannels were carried out using a transilluminated optical microscope (Olympus BX53). Before observation, the samples were placed in the clean room for one day to make sure the water in the microchannel had been evaporated. As shown in Fig. 3, the fabricated channels appeared dark in Fig. 3 under the optical microscope because the channels were filled with air, which has a much lower refractive index than that of glass. It can be observed that the higher the incident pulse energy, the longer and the thicker the drilled microchannel for both the focusing schemes. Meanwhile, the aspect ratios of the drilled channels were different for the CF and SSTF schemes, i.e., the aspect ratio was lower with the CF scheme than that obtained with SSTF scheme, as compared in Fig. 3 (a-f) and (g-l). It should be point out that water played an important role in microchannels drilling because when the rear surface of the sample was in contact with distilled water, the inflow water would disperse ablated material and debris, the effects of redeposition and blocking of ablated material are greatly reduced, which benefit the elongation of microchannels [20].
Fig. 3. Optical micrographs of microchannels drilled from the rear surface of a 10 mm thick silica glass with CF scheme (a-f) and SSTF scheme (g-l) at different laser powers using a 0.06 NA lens. Writing speed: 5 µm/s; laser pulse energy: 6 µJ in (a, g), 10 µJ in (b, h), 20 µJ in (c, i), 40 µJ in (d, j), 60 µJ in (e, k) and 80 µJ in (f, l). Scale bar: 200 µm.
Secondly, the cross sectional profiles of the drilled channels were different. When the laser energy was near the ablation threshold of silica glass, the profiles of the channels were nearly round-shaped for both schemes, as shown in Fig. 3 (a) and (g). However, when the laser pulse energy becomes significantly higher than the ablation threshold, the cross sectional shape of the drilled channel extends along the direction of the spatial chirp in SSTF. This can be explained as follow. It is known that at the geometric focus, the SSTF spot is round shaped given that a symmetric beam profile is chosen for the femtosecond laser pulses before they pass through the two gratings. However, after the geometric focus, the SSTF spot will rapidly expand along the spatial chirp direction as the different frequency components have different incident positions in the back aperture of the focal lens. Since the focal spot is moving upwards during the drilling, the opening of the channel will still be ablated even the rear surface is slightly out of focus due to the peak intensity far above the ablation threshold intensity. As shown in Fig. 3 (f) and (l), the channel obtained with the SSTF scheme is ∼4.5 times longer than that obtained with the CF scheme. The length of the fabricated microchannel has a limitation because the debris will be generated at the irradiated region. As the length of the microchannel increased, more and more debris will gather together in the area of the femtosecond laser pulses interact with the glass [13,14]. As a result, the scattering and absorption of the laser pulses will be induced, which interrupt the microchannel fabrication.
The elliptical sectional shape associated with the SSTF scheme can be mitigated by further reducing the NA of the focal lens, as in this way, the fan-out of the frequency components after the geometric focus can be slowed down. Therefore, we chose a telescope system consisting of a convex lens (f = 300 mm) and a concave lens (f = −100 mm) to reduce the beam width to 3 mm. The effective NA was reduced to 0.02, whereas the pulse energy was varied from 80 µJ to 150 µJ. The writing speed was maintained to be 5 µm/s.
Figure 4 show the results obtained with the CF and SSTF schemes at 0.02 NA. With the CF scheme, the results in Fig. 4(a-d) show only very shallow holes resulting from the ablation. In contrast, with the SSTF scheme, the widths of the drilled channels were measured to be 35 µm, 48 µm, 52 µm, and 59 µm, and the lengths of the channels were measured to be 112 µm, 1420 µm, 1432 µm and 1287 µm in Fig. 4(e-h), respectively. The removal rate was 18.08 µm3 per pulse with the pulse energy of 100 µJ. It can be observed that the cross sectional profiles of the channels drilled with SSTF are almost round-shaped. It is noteworthy that the aspect ratio of the channels drilled with SSTF scheme are much larger than that with the CF scheme for the same laser parameters. In addition, as compared with the results in Fig. 3, the aspect ratio of channels drilled with the CF scheme decreases significantly at lower effective NA despite the same power density used in the two experiments. The results clearly show that drilling with SSTF scheme can generate microchannels of much higher aspect ratios than the CF scheme, particularly for low NA focal lenses. It is found that with the pulse energy of 100 µJ, we achieved the highest aspect ratio of the fabricated channels, as shown in Fig. 4(g). The aspect ratio of channels become lower with a higher pulse energy, for example, 150 µJ. There are two possible reasons for this phenomenon. On one hand, the diameter of the drilled hole becomes larger with high pulse energies. Since the water flow into the channel primarily by the capillary effect, the large hole would be harmful for the water assistance for drilling [21]. On the other hand, strong plasma would be generated with the high energy pulses, even with SSTF scheme. The long plasma formation will deteriorate the precise energy deposition and consequently impede material ablation.
Fig. 4. Optical micrographs of microchannels drilled from the rear surface of a 10 mm thick silica glass with CF scheme (a-f) and SSTF scheme (g-l) at different laser powers at 0.02 NA. Writing speed: 5 µm/s, laser pulse energy: 80 µJ in (a, e), 100 µJ in (b, f), 120 µJ in (c, g) and 150 µJ in (d, h). Scale bar: 200 µm.
Furthermore, we also studied femtosecond laser drilling in thick silica glass with different writing speeds. In this case, we keep all other parameters the same as that in Fig. 4 except the writing speed. Figure 5 shows the results obtained using SSTF scheme and CF scheme at NA = 0.02 with the writing speed of 10 µm/s. It can be seen that the aspect ratio was much higher with the SSTF scheme than that obtained with CF scheme, as shown in Fig. 5 (a-f) and (g-l). Comparing with the result with the writing speed of 5 µm/s in Fig. 4, the microchannels fabricated with the writing speed of 10 µm/s have similar aspect ratio with SSTF scheme. When the writing speed was further increased to 20 µm/s, the microchannel can hardly be fabricated with either SSTF or CF scheme due to the insufficient irradiation time of femtosecond laser pulses. It should be point out that the femtosecond laser source employed in our experiment has a low repetition rate of 1 kHz. The fabrication efficiency can be effectively enhanced by replacing the femtosecond laser with a higher-repetition-rate femtosecond laser.
Fig. 5. Optical micrographs of microchannels drilled from the rear surface of a 10 mm thick silica glass with CF scheme (a-f) and SSTF scheme (g-l) at different laser powers at 0.02 NA. Writing speed: 10 µm/s, laser pulse energy: 80 µJ in (a, e), 100 µJ in (b, f), 120 µJ in (c, g) and 150 µJ in (d, h). Scale bar: 200 µm.
3.2 Laser drilling of 3D microchannels in thick silica glass
At last, we fabricated curved and helical microchannels from the rear surface of the 10 mm thick silica glass at a constant writing speed of 10 µm/s. The experimental parameters were the same as that in Fig. 5(f) as the channel in Fig. 5(f) exhibits the highest aspect ratio. The micrographs of the two curved (i.e., of different radii of curvature) channels are shown in Fig. 6(a) and (b), and the helical channel is shown in Fig. 6(c). The coil radius of the helical microchannel is 150 µm. It can be seen in the zoom-in views in Fig. 6(d-f) that there are no signatures of microcracks formed in the microchannels, which is critical for microfluidic application. By taking advantage of the capability of 3D fabrication, we can apply SSTF to drill 3D microchannels of arbitrary geometries inside transparent materials.
Fig. 6. Micrographs and schematic drawing (Inserts) of curved microchannels (a, b) and a helical microchannel (c) fabricated with the SSTF scheme. (d-f) show zoom-in views of the microchannels. Writing speed: 10 µm/s, laser pulse energy: 100 µJ. Scale bar: 50 µm.
4. Conclusion
To conclude, we have demonstrated water-assisted laser drilling of high-aspect-ratio 3D microchannels in thick silica glass using an unconventional SSTF scheme. The maximum aspect ratio of microchannels achieved in our experiment is ∼30, which is impossible to achieve with the conventional focusing scheme. We expect that our technique can be used for 3D drilling of high-aspect-ratio microchannels not only in glass but also in other transparent materials such as polymers and crystals.
Funding
National Natural Science Foundation of China (NSFC) (11674340, 11734009, 11822410, 11874375, 61590934, 61761136006).
References
1. K. Sugioka and Y. Cheng, “Femtosecond laser three-dimensional micro- and nanofabrication,” Appl. Phys. Rev. 1(4), 041303 (2014). [CrossRef]
2. M. Malinauskas, A. Žukauskas, S. Hasegawa, Y. Hayasaki, V. Mizeikis, R. Buividas, and S. Juodkazis, “Ultrafast laser processing of materials: from science to industry,” Light: Sci. Appl. 5(8), e16133 (2016). [CrossRef]
3. A. Manz and H. Becker, “Microsystem technology in chemistry and life sciences,” Springer Science & Business Media (2003).
4. J. Voldman, M. L. Gray, and M. A. Schmidt, “Microfabrication in biology and medicine,” Annu. Rev. Biomed. Eng. 1(1), 401–425 (1999). [CrossRef]
5. Y. Cheng, “Internal laser writing of high-aspect-ratio microfluidic structures in silicate glasses for Lab-on-a-Chip applications,” Micromachines 8(2), 59 (2017). [CrossRef]
6. A. Marcinkevičius, S. Juodkazis, M. Watanabe, M. Miwa, S. Matsuo, H. Misawa, and J. Nishii, “Femtosecond laser-assisted three-dimensional microfabrication in silica,” Opt. Lett. 26(5), 277–279 (2001). [CrossRef]
7. Y. Bellouard, A. Said, M. Dugan, and P. Bado, “Fabrication of high-aspect ratio, micro-fluidic channels and tunnels using femtosecond laser pulses and chemical etching,” Opt. Express 12(10), 2120–2129 (2004). [CrossRef]
8. C. Hnatovsky, R. S. Taylor, E. Simova, P. P. Rajeev, D. M. Rayner, V. R. Bhardwaj, and P. B. Corkum, “Fabrication of microchannels in glass using focused femtosecond laser radiation and selective chemical etching,” Appl. Phys. A 84(1-2), 47–61 (2006). [CrossRef]
9. S. Kiyama, S. Matsuo, S. Hashimoto, and Y. Morihira, “Examination of Etching Agent and Etching Mechanism on Femotosecond Laser Microfabrication of Channels Inside Vitreous Silica Substrates,” J. Phys. Chem. C 113(27), 11560–11566 (2009). [CrossRef]
10. A. Ródenas, M. Gu, G. Corrielli, P. Paiè, S. John, A. K. Kar, and R. Osellame, “Three-dimensional femtosecond laser nanolithography of crystals,” Nat. Photonics 13(2), 105–109 (2019). [CrossRef]
11. Y. Li, K. Itoh, W. Watanable, K. Yamada, D. Kuroda, J. Nishii, and Y. Jiang, “Three-dimensional hole drilling of silica glass from the rear surface with femtosecond laser pulses,” Opt. Lett. 26(23), 1912–1914 (2001). [CrossRef]
12. R. An, Y. Li, Y. Dou, Y. Fang, H. Yang, and Q. Gong, “Laser micro-hole drilling of soda-lime glass with femtosecond pulses,” Chin. Phys. Lett. 21(12), 2465–2468 (2004). [CrossRef]
13. Y. Li and S. Qu, “Femtosecond laser-induced breakdown in distilled water for fabricating the helical microchannels array,” Opt. Lett. 36(21), 4236–4238 (2011). [CrossRef]
14. D. J. Hwang, T. Y. Choi, and C. P. Grigoropoulos, “Liquid assisted femtosecond laser drilling of straight and three-dimensional microchannels in glass,” Appl. Phys. A 79(3), 605–612 (2004). [CrossRef]
15. F. Théberge, N. Aközbek, W. Liu, A. Becker, and S. L. Chin, “Tunable ultrashort laser pulses generated through filamentation in gases,” Phys. Rev. Lett. 97(2), 023904 (2006). [CrossRef]
16. B. Zeng, W. Chu, H. Gao, W. Liu, G. Li, H. Zhang, J. Yao, J. Ni, S. L. Chin, Y. Cheng, and Z. Xu, “Enhancement of peak intensity in a filament core with spatiotemporally focused femtosecond laser pulses,” Phys. Rev. A 84(6), 063819 (2011). [CrossRef]
17. C. G. Durfee, M. Greco, E. Block, D. Vitek, and J. A. Squier, “Intuitive analysis of space-time focusing with double-ABCD calculation,” Opt. Express 20(13), 14244–14259 (2012). [CrossRef]
18. C. Jing, Z. Wang, and Y. Cheng, “Characteristics and applications of spatiotemporally focused femtosecond laser pulses,” Appl. Sci. 6(12), 428 (2016). [CrossRef]
19. D. N. Vitek, D. E. Adams, A. Johnson, P. S. Tsai, S. Backus, C. G. Durfee, D. Kleinfeld, and J. A. Squier, “Temporally focused femtosecond laser pulses for low numerical aperture micromachining through optically transparent materials,” Opt. Express 18(17), 18086–18094 (2010). [CrossRef]
20. R. An, Y. Li, Y. Dou, H. Yang, and Q. Gong, “Simultaneous multi-microhole drilling of soda-lime glass by water-assisted ablation with femtosecond laser pulses,” Opt. Express 13(6), 1855–1859 (2005). [CrossRef]
21. X. Zhao and Y. C. Shin, “Femtosecond laser drilling of high-aspect ratio microchannels in glass,” Appl. Phys. A 104(2), 713–719 (2011). [CrossRef]
