Open Access
Add to BibSonomy
Abstract
Driven by the increasing demand for faster high-performance computing (HPC) networks and higher data center fabric transmission bandwidth, to favorite the needs of machine learning, data training, and computing, the adoption of co-packaged optics (CPO) and near-packaged optics (NPO) is one of the innovations to mitigate the slowing down of Moore’s law. Because of the high temperature generated by the next generation of high-speed chips like switch ASICs, CPUs, and GPUs, coupling fibers to photonic integrated circuit (PIC) with traditional epoxy-based fiber arrays is becoming more challenging and problematic. Therefore, an epoxy-free bonding method using femtosecond laser welding borosilicate glass 3.3 and optical fibers is proposed and demonstrated. Then, a low loss and polarization independent fiber to fiber coupling was demonstrated to show the reliability of bonding. In the experiment, a V groove is used for aligning and positioning two fibers. After welding, the minimum coupling loss and polarization dependent loss is 0.347 dB and below 0.1 dB respectively. The average shear force limit of the welded samples with 0.5 mm welding length is measured to be as high as ∼0.719 N. This technology could be used for epoxy-free based edge coupling the high density multi-fibers with PIC and has potential of scalable manufacturability through automation.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Co-packaged optics (CPO) and near-packaged optics (NPO) is an advanced heterogeneous integration of optics and silicon on a single packaged substrate, making the signal routing between photonic integrated circuits (PICs) through optical fibers instead of copper traces on the PCB is required. The input and output interfaces of PICs usually have very small mode field size, thus there is a large loss caused by waveguide modal distributions in the direct coupling of commercial optical fibers, like single-mode fiber (SMF). Two solutions can overcome the issue [1]: (i) Vertical coupling, the end face of the fiber is attached to the surface of a designed coupling grating which can modify the k-vector direction of the incident light beam [2,3]. (ii) Edge coupling, the fiber coupling to a PIC through a lateral side that has a mode converter [4,5]. The first strategy provides a usually polarization dependent coupling with narrower wavelength width and relatively low efficiency. While the latter coupling strategy provides a wide wavelength coupling range, high coupling efficiency, and polarization independent characteristics [6,7]. The refractive index profile of the spot size converter of edge coupling PIC can be modified by varying the oxygen partial pressure during deposition, to obtain any value from 1.47 to 2.3 [8]. This technique makes edge coupling PICs widely used. In the actual standardized manufacturing process, edge coupling can be achieved by using a traditional epoxy based fiber array to actively align fibers and position [4] precisely. Each fiber has three quality contact points, one with the glass lid, and two with the V-groove. And the interspaces among V-groove, fibers, and glass lid is filled with epoxy to ensure the fiber doesn’t move around that’s essential to get the fiber core aligned consistently with PIC even in the high-temperature environment. However, there is another problem to be solved. In the above two V groove assisted edge coupling strategies, the epoxy used to package the fiber to the PIC is easy to age and cannot withstand high-temperature, which makes it difficult to guarantee the robustness of the coupling between the PIC and fiber.
Epoxy-free fixing and packaging can solve the issue. Maniewski et al. locally sinter sub-micron silica powders using a CO2 laser, enabling rigid bonding of optical fiber to glass substrates [9]. Nauriyal et al. achieved fiber to PIC coupling by fusing a cleaved SMF to a PIC using a CO2 laser [10]. The fuse induced loss of the method is 1.3 dB (Fresnel loss and mode mismatch loss are not included), which could be due to the strong thermal effect of CO2 lasers resulting in the damage of the fiber. Conversely, the ultrafast laser with extremely short pulse duration and high peak power, whose thermal accumulation is very weak in the low repetition rate case [11,12]. Even in the high repetition rate case, the heat affected zone can be reduced to several micrometers, even hundreds of nanometers [13,14]. Femtosecond (fs) laser can induce local melting and resolidification of material interface and can be successfully applied to the welding of glass [15,16], silicon [17], ceramics [18], metals [19], and other materials, with several MPa to tens of MPa shear joint strength. Besides, fs laser can through transparent materials to process the inside and bottom of the materials. Due to the local plasmonization in the focal volume caused by the ultra-high peak power of the fs laser, which limits the impact of the differential thermal expansion experienced by the two materials, thus it can be used to weld different materials together. This has been demonstrated with fs laser welded dissimilar materials [15,19,20]. Because of the above reasons, fs laser welding has great potential to be used in the epoxy-free bonding process.
In this work, an fs laser processing method to weld commercial SMF to borosilicate glass (BSG) 3.3 lid for realizing optical fiber bonding. The BSG 3.3 with a low linear expansion coefficient of ∼3.3 × 10−6 °C-1, is commonly used in standardized PIC coupling processes. The spatial beam slit shaping of the fs laser is used for elongating the welding area. The article describes processing details and subsequent characterization of the welding strength. The spectral parameters of the welded samples including the insertion losses and polarization-dependent losses are measured for studying the influence of welding on optical transmission. Then a low loss, high quality, and robust fiber to fiber coupling are realized. These excellent characteristics indicate that the technology can be potentially used in the coupling and packaging of PICs.
2. Welding and characterization
The laser welding equipment used in the experiment consists of an fs laser (Pharos, Light-conversion), optical path system, three-dimensional precision moving system, computer control system, and imaging system. The fs laser with a central wavelength of 1028 nm, a pulse width of ∼290 fs, a repetition rate of 50 kHz, and a laser spot diameter of ∼5 mm was employed. The repetition rate can be lowered by a pulse picker. The laser beam passed through an adjustable slit whose direction was set parallel to the axial direction of the optical fiber, and then was focused by a 50× objective (Nikon, NA = 0.6). In the previous reports, the method of the slit-shaping fs laser beam has been used to develop high-quality fiber Bragg grating and symmetrical waveguides [21,22]. The modified region of transparent material by fs laser without beam shaping, with direct focus, is usually in a droplet shape. The slit beam shaping (SBS) method can transversely broaden the modulation region of fs laser to improve the circularity of waveguide cross-section, and the writing efficiency of point-by-point etching FBG. Here, the SBS method is used to increase the cross-sectional welding area without increasing it in a longitudinal affected area.
Figure 1 shows the schematic of the fs laser welding of optical fiber to a BSG lid. The transparent quartz V groove was fixed on a three-dimensional precision moving stage. Uncoated fibers were placed in the V grooves. Then a clean and smooth-surfaced BSG lid was covered on the fibers. Two fibers at least are necessary to keep the BSG lid balanced. Ultrafast laser high-quality welding has identified a requirement with optical contact, i.e., less than a quarter of the laser wavelength for the gap between two transparent glass. In general, in this case, the welding strength is higher than that in the case of a gap greater than 1/4 wavelength [20]. Here, such close contact is obtained by the weight of the BSG lid, and the van der Waals force between the fiber and BSG lid, without extra pressure. Finally, the laser focusing position was set at the interface between BSG lid and optical fiber, and then the moving stage moved along the fiber axis at a speed of 1 mm/s, as shown in Fig. 1. The repetition rate of the laser was set to 10 kHz using a pulse picker to avoid damage to the optical shutter caused by heat accumulation. The laser shutter remained closed during the acceleration and deceleration of the moving stage. It takes only 1.5 s to weld a 0.5 mm length of a single fiber, of which 1 s is the time spent in the acceleration and deceleration of the moving stage.
Fig. 1. Schematic setup for the femtosecond laser welding between the BSG lid and optical fiber. Insert: Schematic of the optical contact; Microscope image of the focused spot after and before the slit beam shaping.
The schematic of the cross-section of the fiber contact with the BSG lid setup is shown in the inset of Fig. 1. Point A is the contact point. The transverse length of optical contact (2AB) can be estimated by a geometrical relation:
where AC is the optical contact distance, which is 257.5 nm. R is the radius of the SMF, which is 62.5 µm. The transverse length satisfying the optical contact was calculated to be 11.4 µm. As shown in the insert of Fig. 1, when the slit width was set to ∼370 µm, an elongated focused spot whose ablated width is ∼11 µm with the laser power out of the slit being ∼170 mW. By contrast, the 170 mW fs laser whose focused spot size without slit beam shaping was only ∼4 µm, which was less than the optical contact length between BSG and fiber. Therefore, the fs laser with SBS for welding can avoid multi-dimensional scanning and consume more time.Looking from the BSG lid side of the sample to the optical fiber side, the areas including the unbonded and bonded is shown in Fig. 2(a). After the welding process was done, the fiber and BSG lid were separated to characterize the welding area by scanning electron microscopy (SEM). The SEM images of the laser affected zone on the BSG lid and fiber are shown in Fig. 2. As shown in Fig. 2(b), in the enlarged image of Fig. 2(a), the ablated width of the focused laser on the BSG lid is ∼11 µm, and the welding width is ∼7 µm. And the laser-induced welding width on fiber is ∼6.6 µm, which is similar to that on the BSG lid. The periodic surface structures in the laser affected area could be attributed to the influence of both Fresnel diffraction induced by slit and directional surface plasmon polariton scattering [23].
Fig. 2. (a) Microscopic image of a BSG lid bonded to a fiber. (b) SEM image of a BSG lid after welding. (c) Enlarged SEM image of (b). (d) SEM image of an SMF after welding. (e) Enlarged SEM image of (d).
3. Shear force limitation
The shear force limitation of the welded samples was measured using a tensile testing sensor and a single-axis moving stage. The BSG lid was fixed on the stage and extended optical fiber was fixed at the tensile testing sensor, using epoxy glue. The tensile force was increased by moving the stage until the welded sample was damaged. A sample with a 10 mm welding length were welded using 170mW laser, whose shear force limitation (F) is 1.19 N, as shown in the data of Fig. 3(a). When the axial tensile force of 1.19 N was applied to the sample, the fiber was broken to two sections, while the fiber on the welding area was not fall off. The reason could be the laser-induced periodic surface cracks on the fiber inducing the fiber to break apart by tensile force, thus the contribution of welding area to tensile limit could be weakened. Besides, as shown in Fig. 3(b), the average shear force limitations of 10 samples with welding lengths of 2 mm using 150 mW, 170 mW, 190mW, and 210 mW laser power are 0.708 N, 0.76 N, 0.665 N, and 0.612 N, respectively. The highest average shear force limitation was found for welded samples using 170 mW laser power. When the fs laser parameter deviation from optimal parameter values may lead to strong ablation and cracking, failing bonding [24]. Therefore, we chose to use a relatively optimized parameter of 170 mW fs laser to weld for further investigation. 50 welded samples were welded using 170mW laser, were measured for the shear force limitation, the results are shown in Fig. 3(c). The average shear force limitations of 10 samples with welding lengths of 0.5 mm, 2 mm, 5 mm, 10 mm, and 20 mm are 0.719 N, 0.76 N, 0.802 N, 0.87 N, and 0.928 N, respectively. The breaking stress increases with increasing welding length. Even the samples with 0.5 mm welding length have a strong bonding, which makes the technology can be used for real-life applications.
Fig. 3. (a) Shear force limitation test data of a sample with 10 mm welding length. (b) Shear force limit statistics of different welding fs laser power with 2 mm welding length. (c) Shear force limit statistics of 50 samples with different welding lengths. Error bar: the standard deviation of ten measurements.
The cracks formation within the fs laser welding area act as predetermined breaking points. Therefore, the stress state of the cracks should be investigated, especially the crack closest to where tension is applied. To investigate the stress evolution of the laser-induced crack under an applied tensile force in the shear force test with the welding length change. We established a simulation model using commercial finite element analysis software (COMSOL 5.6), and the results are shown in Figs. 4(a)-(c). The simulated fiber model is established including the fixed edges (fiber and BSG lid bonding area) with unconstrained cracks in the fiber surface. Figure 4(b) illustrates the length, depth, and distance distribution of the cracks model, corresponding to 1,3 and 6 µm, respectively. The welding length is defined as the distance from the first crack to the last. The mesh division is constructed using free triangular which are finer around the cracks. The fiber is employed of standard silica, including a density of 2700 kg/m3, Young’s modulus of 73 GPa, and Poisson’s ratio of 0.17. When the axial tensile force of 1 N was applied to the model, the simulated 2D stress contour of the fiber is shown in Fig. 4(c). In the figure, the red point is the tip of the first crack which is easy to break in the tensile force test, whose stress distribution was recorded with different welding lengths, as plotted in Fig. 4(d). The result shows that the stress of the recorded point can be reduced by increasing the welding length, but the effect is weakened with the increase of welding length. This is consistent with the result in Fig. 3(b), as the welding length increases, the contribution of the welding area to the shear force limitation decreases.
Fig. 4. (a) Geometric model of the welded fiber with cracks. (b) Enlarged image of the dotted box. (c) Two-dimensional stress contours while the axial tensile force of 1 N applied to the model. (d) The stress of recorded point with different welding lengths.
The long-term lifetimes of the samples made by the proposed method were investigated. 10 samples with 5 mm welding length were placed on a controllable heater. After a 12 h 250°C environmental test, the fibers did not fall off the BSG lids, which means the welded samples can stand at high temperatures. After high-temperature exposure, the average shear force limitation was 0.569 N with a standard deviation of ∼0.08. While the average shear force limitation of the other 10 samples with 5 mm welding length without high-temperature exposure was 0.802 N. The average shear force limitation was decreased after high-temperature exposure which could be caused by the difference in thermal expansion coefficient between quartz (5.5 × 10−7 °C-1) SMF and BSG (3.3 × 10−6 °C-1), resulting in the weakening strength of bonding area. In addition, the standard 85 °C/85%RH Damp Heat test was performed on other samples. After 1,000 hours 85 °C and 85% humidity environmental test in the reliability laboratory of TFC Optical Communication Co., Ltd. (“TFC”), the fibers did not fall off the BSG lids, demonstrating the long-term lifetimes of femtosecond laser bonding.
4. Optical properties measurement
To investigate the fs laser-induced effect of SMF core, the insertion loss (IL) and polarization dependent loss (PDL) of samples were measured using a broadband light source (BBS), an optical fiber polarizer, a polarization controller, and an optical spectrum analyzer (OSA). A sample with a 30 mm welding length, whose transmission spectra over a wavelength range of 1285-1615 nm, is shown in Fig. 5(a), exhibiting a low IL and PDL of below 0.14 dB. Subsequently, 5 unwelded samples as a comparison and 15 welded samples were measured for the IL and PDL, the results are shown in Fig. 5(b). The IL at 1550 nm of 5 samples with welding lengths of 0 mm, 10 mm, 20 mm, and 30 mm are 0.057 dB, 0.082 dB, 0.042 dB, and 0.012 dB, respectively. These samples have almost no loss, and the small loss difference (< 0.1dB) could be caused by the fluctuation of the light source and the fiber fusion. The average PDL of them are 0.071 dB, 0.049 dB, 0.067 dB, and 0.061 dB, respectively. It is evident from these results that the proposed fs laser welding method with SBS would not affect the transmission and polarization of the SMFs’ core. In the previous reports, the welding width can be enlarged by increasing the pulse energy or pulse count, while also enlarging the depth of the laser-induced zone, which may result in the increase of loss and birefringence of the optical fiber. Here, the SBS shaping method offers an increased welding width without changing the transmission and polarization of SMF. The results show that this technique has the potential to be applied in the coupling between the optical fiber and PIC.
Fig. 5. (a) IL and PDL of a sample with 30 mm welding length. (b) IL and PDL statistics of fifteen samples with 10 mm, 20 mm, and 30 mm welding length. Error bar: light source power fluctuation.
The fiber to fiber coupling was investigated using the fs laser welding, as a simulated condition for fiber to PIC coupling with edge coupling. Two sections of cleaved SMF were placed in a single V groove with a close distance and were covered by a BSG lid. Then two sections of SMF were welded on the BSG lid using the proposed method to realize fiber to fiber coupling, as shown in Fig. 6(a). Looking from the optical fiber side of the sample to the BSG lid side, the gap between the two fiber end facets is ∼2.9 µm, as shown in Fig. 6(b). Because the cleaved fiber section has a small angle tilt, looking from the BSG lid side of the sample to the optical fiber side, the gap between the two fiber end facets is ∼1.9 µm, as shown in Fig. 6(c). The average gap length of the sample is 2.4 µm. The coupling loss of the sample includes Fresnel reflection loss, divergence loss, and misalignment loss. The Fresnel reflection loss of about two fiber-air surfaces is calculated to be ∼0.3 dB. The fringes in Fig. 6(c) are the result of interference between the upper surface of the fiber and the lower surface of the BSG lid. The position where the two optical fibers contact the BSG lid is a dark fringe, which means that the optical path difference there is 0, indicating that the optical fibers contacted the BSG lid. And the corresponding interference fringes of the two fibers are almost aligned, which indicates that the mismatch loss can be ignored.
Fig. 6. (a) Microscopy image from fiber top of the two aligned SMFs which were welded on a BSG lid. (b) Enlarged image of (a). (c) The view from the BSG lid top.
The transmission and reflection spectra over a wavelength range of 1285-1615 nm of the sample with a 2.4 µm gap are shown in Fig. 7(a). The average IL of the sample is 0.447 dB, and the PDL is below 0.15 dB. The green curve in the reflection spectrum is from the interference between the two end facet of the SMF. Several fiber to fiber coupling samples with different gaps were fabricated to investigate the average IL and PDL, the statistics are shown in Fig. 7(b). In all 6 samples, the average PDL is below 0.11 dB, while the average IL is below 0.8 dB. Due to the increase in gap length increasing divergence loss. The ILs in the samples with 6.84 µm and 9.2 µm average gap length are higher than those with 2.08-4.3 µm. Therefore, the gap length should be kept close in the coupling process of this method to obtain a lower loss.
Fig. 7. (a) IL, PDL, and reflection spectrum of a sample with ∼2.9 µm welding length. (b) IL and PDL statistics of several samples with different gap lengths.
5. Conclusions
In conclusion, a promising and simple strategy for bonding optical fiber to a BSG lid using fs laser welding is proposed and demonstrated. The fs laser is shaped by a ∼370 µm width slit to obtain an elongated focused spot whose ablated width is ∼11 µm, which increased the welding width. The average shear force limitation of the welded samples with 0.5 mm welding length is measured to be as high as ∼0.719 N. The welding process takes only 1.5s of 0.5 mm welding length. Besides, the welding method does not change the transmission and polarization of commercial optical fiber. Fiber to fiber coupling using the fs laser welding method with V groove assisted, whose average insertion loss and polarization-dependent loss can be as low as 0.347 dB and below 0.1 dB respectively. The proposed fs laser welding method offers the potential to realize an epoxy-free, high efficiency, and low-cost optical coupling between fiber to PIC. We anticipate this new technology can be applied in high dense multi-fibers coupling with PIC.
Funding
National Natural Science Foundation of China (61925502, 62135007).
Disclosures
The authors declare no conflicts of interest.
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. R. Marchetti, C. Lacava, L. Carroll, K. Gradkowski, and P. Minzioni, “Coupling strategies for silicon photonics integrated chips Invited,” Photonics Res. 7(2), 201–239 (2019). [CrossRef]
2. D. Kwong, A. Hosseini, J. Covey, Y. Zhang, X. Xu, H. Subbaraman, and R. T. Chen, “On-chip silicon optical phased array for two-dimensional beam steering,” Opt. Lett. 39(4), 941–944 (2014). [CrossRef]
3. A. Mekis, S. Gloeckner, G. Masini, A. Narasimha, T. Pinguet, S. Sahni, and P. De Dobbelaere, “A Grating-Coupler-Enabled CMOS Photonics Platform,” IEEE J. Select. Topics Quantum Electron. 17(3), 597–608 (2011). [CrossRef]
4. J. V. Galan, P. Sanchis, G. Sanchez, and J. Marti, “Polarization insensitive low-loss coupling technique between SOI waveguides and high mode field diameter single-mode fibers,” Opt. Express 15(11), 7058–7065 (2007). [CrossRef]
5. Y. Lai, Y. Yu, S. Fu, J. Xu, P. P. Shum, and X. Zhang, “Efficient spot size converter for higher-order mode fiber-chip coupling,” Opt. Lett. 42(18), 3702–3705 (2017). [CrossRef]
6. B. B. Bakir, A. V. de Gyves, R. Orobtchouk, P. Lyan, C. Porzier, A. Roman, and J. M. Fedeli, “Low-Loss (< 1 dB) and Polarization-Insensitive Edge Fiber Couplers Fabricated on 200-mm Silicon-on-Insulator Wafers,” IEEE Photonics Technol. Lett. 22(11), 739–741 (2010). [CrossRef]
7. R. S. Tummidi and M. Webster, “Multilayer Silicon Nitride-based Coupler Integrated into a Silicon Photonics Platform with <1 dB Coupling Loss to a Standard SMF over O, S, C and L optical bands,” in Optical Fiber Communication Conference (OFC) (Optica Publishing Group, 2020), paper Th2A.10.
8. R. Machorro, E. C. Samano, G. Soto, F. Villa, and L. Cota-Araiza, “Modification of refractive index in silicon oxynitride films during deposition,” Mater. Lett. 45(1), 47–50 (2000). [CrossRef]
9. P. Maniewski, M. Fokine, and F. Laurell, “All-silica optical fiber bonding,” Opt. Express 30(11), 19805–19811 (2022). [CrossRef]
10. J. Nauriyal, M. Song, R. Yu, and J. Cardenas, “Fiber-to-chip fusion splicing for low-loss photonic packaging,” Optica 6(5), 549–552 (2019). [CrossRef]
11. R. R. Gattass and E. Mazur, “Femtosecond laser micromachining in transparent materials,” Nat. Photonics 2(4), 219–225 (2008). [CrossRef]
12. H. Chen, L. Deng, J. Duan, and X. Zeng, “Picosecond laser welding of glasses with a large gap by a rapid oscillating scan,” Opt. Lett. 44(10), 2570–2573 (2019). [CrossRef]
13. D. Tan, B. Zhang, and J. Qiu, “Ultrafast Laser Direct Writing in Glass: Thermal Accumulation Engineering and Applications,” Laser Photonics Rev. 15, 2000455 (2021). [CrossRef]
14. J. Luo, S. Liu, P. Chen, S. Lu, Q. Zhang, Y. Chen, B. Du, J. Tang, J. He, C. Liao, and Y. Wang, “Fiber optic hydrogen sensor based on a Fabry-Perot interferometer with a fiber Bragg grating and a nanofilm dagger,” Lab Chip 21(9), 1752–1758 (2021). [CrossRef]
15. T. Tamaki, W. Watanabe, and K. Itoh, “Laser micro-welding of transparent materials by a localized heat accumulation effect using a femtosecond fiber laser at 1558 nm,” Opt. Express 14(22), 10460–10468 (2006). [CrossRef]
16. K. Cvecek, S. Dehmel, I. Miyamoto, and M. Schmidt, “A review on glass welding by ultra-short laser pulses,” Int. J. Extrem. Manuf. 1(4), 042001 (2019). [CrossRef]
17. W. Watanabe, S. Onda, T. Tamaki, and K. Itoh, “Direct joining of glass substrates by 1 kHz femtosecond laser pulses,” Appl. Phys. B 87(1), 85–89 (2007). [CrossRef]
18. E. H. Penilla, L. F. Devia-Cruz, A. T. Wieg, P. Martinez-Torres, N. Cuando-Espitia, P. Sellappan, Y. Kodera, G. Aguilar, and J. E. Garay, “Ultrafast laser welding of ceramics,” Science 365(6455), 803–808 (2019). [CrossRef]
19. G. Zhang and G. Cheng, “Direct welding of glass and metal by 1 kHz femtosecond laser pulses,” Appl. Opt. 54(30), 8957–8961 (2015). [CrossRef]
20. G. Zhang, R. Stoian, W. Zhao, and G. Cheng, “Femtosecond laser Bessel beam welding of transparent to non-transparent materials with large focal-position tolerant zone,” Opt. Express 26(2), 917–926 (2018). [CrossRef]
21. X. Xu, J. He, J. He, B. Xu, R. Chen, K. Yang, C. Liao, Y. Yang, and Y. Wang, “Slit Beam Shaping for Femtosecond Laser Point-by- Point Inscription of High-Quality Fiber Bragg Gratings,” J. Lightwave Technol. 39(15), 5142–5148 (2021). [CrossRef]
22. P. S. Salter, A. Jesacher, J. B. Spring, B. J. Metcalf, N. Thomas-Peter, R. D. Simmonds, N. K. Langford, I. A. Walmsley, and M. J. Booth, “Adaptive slit beam shaping for direct laser written waveguides,” Opt. Lett. 37(4), 470–472 (2012). [CrossRef]
23. T. Wang, L. Jiang, X. Li, J. Hu, Q. Wang, S. Ye, H. Zhang, and Y. Lu, “Controllable anisotropic wetting characteristics on silicon patterned by slit-based spatial focusing of femtosecond laser,” Opt. Express 24(22), 25732–25741 (2016). [CrossRef]
24. S. Butkus, E. Gaižauskas, D. Paipulas, Ž Viburys, D. Kaškelyė, M. Barkauskas, A. Alesenkov, and V. Sirutkaitis, “Rapid microfabrication of transparent materials using filamented femtosecond laser pulses,” Appl. Phys. A 114(1), 81–90 (2014). [CrossRef]
