Ultra-fast lasers can realize selective welding of glasses through nonlinear effect and have the advantages of having high welding accuracy, having small heat-affected zone, having high joint strength and not requiring intermediate absorption layer, which provides a new idea for chip packaging. However, this method is limited to the condition of optical contact, which is difficult to use in engineering applications. To solve this problem, an innovative seal welding method by picosecond laser is presented in this paper. In this method, a picosecond laser performs firstly rapid oscillating scan local welding of the two natural overlap glasses with a large contact gap to form a closed area consisting of spot welds. The glass contact gap in the closed area can be reduced to about 1.5 μm through the solidification shrinkage effect of spot welds. Then a good seal weld without plasma ablation, micro-hole, and micro-crack defects can be achieved by picosecond laser in this closed area to realize seal welding of glasses with a large contact gap. The sealing test results show that the seal welding samples can keep good sealing performance for more than one week.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Glass is an excellent material for chip packaging in the field of solar cells, implanted microelectronics, organic light-emitting diode (OLED), micro-electromechanical systems (MEMS), microsensors and so on, and has very extensive potential applications and market prospects in the automotive, aerospace, electronic semiconductor, and biomedical science. This is because glass material can be regarded as a “neutral” substance for organisms due to its good biocompatibility for fluids and tissues, and cannot degrade spontaneously or be corroded by body fluid like other material. Therefore, there is a long service life due to no immune rejection when the glass is implanted in the human body. Besides, glass material has a light transmittance and does not interfere with electromagnetic waves, which is conducive to the transmission of light and electromagnetic waves through the glass envelope .
At present, the main packaging methods of glass materials include anode bonding , melt bonding, adhesive method  and laser welding [4–7]. However, these methods have their own shortcomings, such as: Anode bonding needs high temperature and high pressure, which is easy to damage the device; Melt bonding requires heating the glass material until it melts, so it is not suitable for most chip packaging; Adhesive method is easy to be aging and volatile, which will directly affect the reliability, stability, and lifetime of the chip; Laser welding requires opaque substrate or filling light-absorbing medium, which seriously affects the light transmittance of the device.
The technology of ultra-fast laser welding of glasses  provided a new idea for solving the problems existing in the above packaging methods. In this technology, an ultra-fast laser was adopted to produce nonlinear effect with glass material at only the focal point, so that the glass material at only the focal point could absorb laser energy and melt [9,10], and then condense to form a sealing weld . Since this weld is realized by melting the material itself, so it has the advantages of high connection strength, good welding accuracy, small heat-affected zone and selective welding [12–18]. However, it requires a harsh condition of optical contact with a glass gap less than 1/4 wavelength , or even within 100 nm , which obviously cannot be used in engineering application. Although some methods [19–21] were proposed to increase the welding gap available into almost 3 μm, it was still smaller than the contact gap between two glasses in natural stacks status . Therefore, there is still a tough problem to realize the engineering application of ultrafast laser seal welding of glasses. In 2019, Chen et al.  proposed a rapid oscillating scan method for welding of glasses with a large gap using picosecond laser. This method cannot only create enough molten material to fill the gap by expanding the area where the laser interacts with the glass, but also release the internal thermal pressure during the welding process to achieve a weld with good quality and high joint strength. However, due to the limitation of the welding path, this method can only form local welds, and cannot directly conduct seal welding on the glass.
In this paper, an innovative seal welding method for glass with a large gap was explored to solve these aforementioned problems, and a seal welding of two natural overlap glasses with a large contact gap was successfully achieved by skillfully combining local spot welding and seal welding. This technology had already been applied for patents by our team, and the related patent publication numbers are CN 110039177A and CN108609841A.
2. Experiment setup
2.1. Experimental material
In this study, a commercial soda-lime glass was used and the physical parameters are listed in Table 1. Its 3D dimensions are 25 mm × 75 mm × 1.0 mm, the main components are SiO2 (73%), Na2O (14%), CaO (9%) and impurity composition (4%).
2.2. Experimental equipment
The laser welding equipment used in the experiment consisted of a solid-state picosecond laser, micro-welding optical path system, precision moving system, and computer control system, as shown in Fig. 1. The picosecond laser manufactured by Edgewave company emits a 10 ps pulses beam with quality factor M2≤1.3 at a wavelength of 1064 nm, and average output power of 0~80 W at a repetition rate of 1 MHz. The focusing spot diameter was about 19.5 μm through the optical path system composed of beam expanding mirror, reflector, galvanometer and f-θ lens. With the cooperation of scanning galvanometer and X-Y-Z 3D precision moving system, a rapid welding of any complex figure within the range of 70mm × 70mm could be realized by the computer control system, and the highest welding speed could reach 9600 mm/s. The whole laser welding equipment had the characteristics of convenient operation, accurate control and high degree of automation.
2.3. Experiment and test method
2.3.1. Correction of the laser focus position
The glass sample to be welded was naturally stacked on X-Y precision moving work table, without any fixture to put pressure on it. In order to focus the picosecond laser beam at the contact gap between two glasses after passing through the upper glass, it is necessary to eliminate the influence of the refraction effect of the upper glass.
The correction method was placing a piece of glass under the galvanometer scanner firstly and adjusting the Z-axis to focus the laser beam on the glass surface, then stacking the second piece of glass naturally on the first piece of glass. The contact gap between two glass sheets measured by a microscope was about 10 μm, which mainly depends on the unevenness of the glass. Due to the refraction effect of the upper glass, the actual focus position will move down. The relative displacement distance can be calculated through an approximate formula as follows:
2.3.2. Seal welding method
The method of picosecond laser seal welding of glasses is shown in Fig. 2. Firstly, the rapid oscillating scan method proposed in reference  was adopted to perform spot welding with a radius of 0.3 mm at every interval g, as shown in Fig. 2(a), to investigate the influence law of picosecond laser welding parameters, spot welding spacing and quantity on the glass contact gap, and obtain the best value of welding parameter, spot welding spacing, and quantity. Then, the optimal welding parameters and spot welding spacing were adopted to investigate the impact rule of spot welding enclosure mode on glass surface contact gap, as shown in Fig. 2(b), and obtain the appropriate spot welding enclosure mode. Finally, picosecond laser seal welding was carried out by line welding form in the enclosed area to study the performance of the sealing weld, as shown in Fig. 2(c).
The welding parameters used in the experimental research were: repetition frequency f: 1MHz; single pulse energy J: 8-12 μJ, step length 2 μJ; scanning times m: 50-150 times, step length 50 times; scanning speed v: 1000-2000 mm/s, step length 500 mm/s; welding spot spacing g was 1 mm, 5 mm, and 10 mm respectively. Ring weld form was adopted for seal welding, and the ring spacing s was 0.01 mm, ring width w was 0.3 mm and radius R was 5 mm, as shown in Fig. 2(c).
2.3.3. The test method
220.127.116.11. Method of measuring the contact gap
Although the contact gap between two naturally stacked glasses can be observed and measured through an optical microscope or scanning electron microscope (SEM) from the side, it is difficult to measure the contact gap in the middle of the glass, especially when the contact gap is at the micro-nanometer level. If the common mechanical method was used to cut glass for measurement, it would cause the change of the contact gap due to the mechanical force and the release of stress at the spot weld, resulting in errors, or even damaging the spot weld. Fortunately, when the glass contact gap is small enough, interference fringe will occur due to mutual interference of reflected light on the upper and lower glass surfaces (see Fig. 3(a)), and the optical path difference can be expressed by the following equation:
where, θ is the wedge interference inclination angle, n0 is the air refractive index, d is the contact gap between glasses, and is the half-wave loss caused by the light from sparse to dense medium. Therefore, Eq. (2) can be expressed as follows:
Considering that is small and n0 is 1, Eq. (3) can be simplified to:
Since the value of k is not known in the actual calculation, it is necessary to measure the contact gap dx corresponding to the x-level interference fringe at the sample edge with the aid of a microscope (see Fig. 3(b)), and the following equation can be obtained
18.104.22.168. Method of sealing performance test
The sealing performance of the sample was tested by water immersion . The specific operation is to place the sample in water and directly observe it with naked eyes whether any water entering the enclosed area. The area where water enters is transparent, otherwise is mirror plane.
3. Experimental results
3.1. The effect of picosecond welding parameters on the contact gap
Figure 4 shows the influence of different picosecond welding parameters, spot welding spacing and quantity on the glass contact gap based on the interference fringe calculation method. As the interference fringes are relatively narrow, almost only bright red fringes can be observed (Fig. 3(a)), therefore the wavelength was chosen according to the central wavelength of the red spectrum, namely the wavelength λ = 700nm. The experimental results show that the contact gap between two pieces of glass will decrease with the increase of the spot welding quantity under any picosecond laser welding parameters. When the spot welding quantity increases to about 16, the contact gap will stabilize around a certain value and will not decrease with the increase of the spot welding quantity. That is to say, the change of picosecond laser welding parameters has basically the same influence trend on the glass contact gap. However, the effect of different picosecond welding parameters on the glass contact gap is different. The contact gap between two pieces of glass decreases with the increase of pulse energy J and scanning times m, and increases with the increase of scanning speed v and spot welding spacing g, as shown in Figs. 4(a)–4(d). When g = 5 mm and 10 mm, the spot welding quantity is reduced due to the limitation of glass size (25 mm × 75 mm), as shown in Fig. 4(d). The optimum welding parameters are J = 12 μJ,v = 1000 mm/s,m = 150,g = 1 mm and the spot welding quantity is 16. Under these parameters, the contact gap between two pieces of glass can be reduced into about 1.5 μm, as shown in Fig. 4.
Figure 5 shows the measured results of the glass contact gap varying with the spot welding quantity under the optimum welding parameters. The red dot is the gap value measured by SEM after cutting and polishing, and the black dot is the calculated value. The comparison results show that the variation trend of the contact gap between the measured and calculated results with the spot welding quantity is basically the same, which verifies that the calculation method is suitable and reliable. The measured values are slightly larger because of the error caused by the effect of external force and internal stress release on the original glass gap in the sample cutting and polishing process.
3.2. Effect of spot welding enclosure mode on the seal welding quality
Figure 6 shows the influence law of different spot welding enclosure mode on the quality of seal welding under the conditions of optimum welding parameters and spot welding spacing. The experimental parameters of seal welding are: J = 12 μJ, f = 1 MHz, v = 1000 mm/s, m = 150, R = 5 mm, w = 0.3 mm, s = 0.01 mm. In the case of I-type, as shown in Fig. 6(a), the contact gap only around the spot weld can be narrowed to about 1.5 μm, and the interference fringes gradually disappear as the distance from the spot weld increases. In this case, the seal weld near the spot weld can obtain better quality, otherwise, the weld quality is poor, and even cannot form a weld. The SEM microstructure of the weld away from spot weld indicates that there is a lot of granular loose substance, white powder and less glass melt inside the weld, as shown in Fig. 6(b). As the spot welding enclosure mode becomes L-type and Π-type, the interference fringes inside the area surrounded by spot welds gradually become sparse and the interference fringe area increases, as shown in Figs. 6(c) and 6(e). Under this circumstance, the quality of sealing weld can be significantly improved, not only the granular loose material gradually decreases and the glass melt increases, but also the formation of the weld becomes better, and the white powder gradually disappears, as shown in Figs. 6(d) and 6(f). When the spot welds form a closed mode, the interference fringes become the most uniform inside the closed area, as shown in Fig. 6(g). The seal welding ring carried out inside this area and the interference fringes in it are almost a set of concentric circles, which shows that the gap near the seal welding is good consistency, and the weld quality is uniform. The SEM images also show that the weld is basically melted, and there is no granular loose substance and white powder, so the quality of the weld is further improved, as shown in Fig. 6(h).
3.3. Sealing test results
Under the same conditions of welding parameters, two kinds of seal welding shapes were carried out in the closed area formed by spot welds: one was rectangular and the other was circular. The side length of the rectangular and the diameter of the ring were all 1 cm, the ring width w was 0.3 mm, and the line spacing s was 0.01 mm. The welded samples are shown in Figs. 7(a) and 7(d), respectively. The samples were put into water for sealing test and showed good sealing performance within 1 hour. No water infiltrated into the sealing zone (Figs. 7(b) and 7(e)). When it was more than 1 hour, the sample of rectangular seal ring began to infiltrate water into the sealing zone from the corner, as shown in Fig. 7(c), while the sample of the circular seal ring could maintain good sealing performance after one week in water (Fig. 7(f)).
The unevenness of the glass surface will cause the contact gap between two pieces of glass to fluctuate greatly, which makes it impossible to obtain an effective weld formation. By adopting the picosecond laser rapid oscillating scan welding method, the range of interaction between picosecond laser and glass can be enlarged to produce more glass melt which is pushed into the gap by thermal expansion and pressure to connect the upper and lower layers of glass, and in the solidification process of glass melt, an inducing tensile stress caused by the shrinkage effect will narrow the contact gap near the spot weld . Therefore, with the increase of the spot welding quantity, the total tensile stress induced by these spot welds will also increase to reduce further the glass contact gap. At the same time, due to the unevenness of the glass surface, the convex portion of the glass surface is elastically deformed, resulting in reverse elastic stress that hinders the shrinkage of the glass contact gap. Once the increment of induced tensile stress and reverse elastic stress reaches a dynamic equilibrium state, the glass contact gap reaches a dynamic stable value, and fluctuates around the value, and no longer decreases with the increase of spot welds.
Obviously, the increase of the laser pulse energy and the laser scanning number or the decrease of scanning speed will all increase the deposition of laser energy in the glass material. The more the energy is deposited, the more the glass melts, the larger the volume expansion of the molten body, the more obvious the shrinkage effect and the larger the induced shrinkage force. Therefore, the effect of reducing the glass contact gap is more significant. In addition, the smaller the spacing of spot welding is, the larger the induced tensile stress per unit length is, therefore, the more significant the shrinkage effect is, the more uniform the contact gap is, and the smaller the wedge angle θ is, as shown in Fig. 8, where some slag breakage in the weld seam of Fig. 8(b) is not caused by welding, but the material removed during grinding. However, if the spacing between spot welding is too small, micro-cracks will occur due to the interaction between heat-affected zones.
Although the contact gap near the spot weld will shrink during the re-solidification process, interference fringes with a narrower width and higher density appear around the spot weld, as shown in Fig. 3(a), which indicates that the solidification shrinkage effect makes the contact gap as wedge-shaped, and the wedge angle θ is large. As the quantity of spot weld increases to form I-type, the contact gap along the I direction decreases with the increase of spot welding quantity (the interference fringes become wider and sparser gradually, and the θ value decreases), but the contact gap is not uniform, as shown in Fig. 6(a). Moreover, the wedge angle of the contact gap perpendicular to the I direction is still large, and the contact gap gradually increases with the increase of the distance (the interference fringes gradually disappear). Therefore, in the area far from spot weld, the excessive contact gap will not only cause insufficient melt filling gap, but also produce etching effect, forming a lot of granular loose substance and white powder inside the weld, as shown in Fig. 6(b), and resulting in the failure to form an effective weld . With the increase of the extent of the area surrounded by spot welds (from type I to L and Π), the interference fringes in the area surrounded by spot welds gradually become sparse, the width of interference fringes gradually increases and the area of interference fringes increases, as shown in Figs. 6(c) and 6(e), which shows that the wedge angle of glass is becoming smaller and smaller, and the contact gap is becoming more uniform. In this case, the etching effect inside the weld gradually weakens to make the granular loose material and white powder gradually disappeared, and the melting amount gradually increases, as shown in Figs. 6(d) and 6(f). Once the spot welds surround the area, a planar shrinkage force on the glass is formed, which cannot only greatly reduce the glass contact gap inside the area to achieve a subsequent seal welding with high quality, but also ensure greatly the uniformity of the glass contact gap inside the area to obtain a seal welding with uniform quality, as shown in Fig. 6(h).
In addition, the form of seal welding has a great influence on the sealing performance. The reason why the sealing effect of rectangular seal weld is not good is that under the same parameters of picosecond laser welding, more energy is deposited at the corner of the rectangular seal weld to produce greater stress at the corner. The uneven stress of the rectangular seal weld can easily lead to glass cracking at the corner so that it cannot maintain good sealing for a long time. However the sample of circular sealing weld has a good consistency and can obtain a uniform stress distribution on the weld path, so good sealing performance can be obtained. From the section microstructure of seal weld shown in Fig. 9, it can be seen that the melt not only fills the contact gap and connects the upper and lower glass perfectly, but also the weld is well-formed without defects such as plasma ablation, micro-hole, and micro-crack.
In this paper, a method of picosecond laser seal welding of glasses with a large gap was presented. By using the solidification shrinkage effect of melt obtained by rapid oscillating scan welding method, the effects of picosecond welding parameters, spot welding spacing, and quantity, and the spot welding enclosure mode on the glass contact gap were investigated and the relevant mechanism was revealed and discussed. The experimental results show that under the optimum conditions, the glass contact gap can be reduced to about 1.5 μm. Moreover, inside the enclosed area surrounded by spot welds, a good seal weld quality without plasma ablation, micro-hole and micro-crack defects can be obtained, and the annular seal weld has good sealing performance, which proves the effectiveness and reliability of the method and has an important influence on the engineering application of laser welding technology for glass.
National Natural Science Foundation of China (51675205).
1. E. Axinte, “Glasses as engineering materials: A review,” Mater. Des. 32(4), 1717–1732 (2011). [CrossRef]
2. A. Elrefaey, J. Janczak-Rusch, and M. M. Koebel, “Direct glass-to-metal joining by simultaneous anodic bonding and soldering with activated liquid tin solder,” J. Mater. Process. Technol. 214(11), 2716–2722 (2014). [CrossRef]
3. Y. J. Pan and R. J. Yang, “A glass microfluidic chip adhesive bonding method at room temperature,” J. Micromech. Microeng. 16(12), 2666–2672 (2006). [CrossRef]
4. Q. Wu, N. Lorenz, K. M. Cannon, and D. P. Hand, “Glass Frit as a Hermetic Joining Layer in Laser Based Joining of Miniature Devices,” IEEE Trans. Compon. Packag. Tech. 33(2), 470–477 (2010). [CrossRef]
5. C. Lin, Y. Shen, C. Huang, and S. Tu, “Laser sealing of organic light-emitting diode using low melting temperature glass frit,” Opt. Quantum Electron. 49(6), 1–10 (2017). [CrossRef]
6. U. M. Mescheder, M. Alavi, K. Hiltmann, C. Lietzau, C. Nachtigall, and H. Sandmaier, “Local laser bonding for low temperature budget,” Sens. Actuators A Phys. 97, 422–427 (2002). [CrossRef]
7. A. de Pablos-Martín, S. Tismer, F. Naumann, M. Krause, M. Lorenz, M. Grundmann, and T. Höche, “Evaluation of the bond quality of laser-joined sapphire wafers using a fresnoite-glass sealant,” Microsyst. Technol. 22(1), 207–214 (2016). [CrossRef]
8. T. Tamaki, W. Watanabe, J. Nishii, and K. Itoh, “Welding of Transparent Materials Using Femtosecond Laser Pulses,” Jpn. J. Appl. Phys. 44(22), L687–L689 (2005). [CrossRef]
9. 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] [PubMed]
10. C. B. Schaffer, J. F. García, and E. Mazur, “Bulk heating of transparent materials using a high-repetition-rate femtosecond laser,” Appl. Phys., A Mater. Sci. Process. 76(3), 351–354 (2003). [CrossRef]
12. 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]
13. I. Miyamoto, A. Horn, and J. Gottmann, “Local Melting of Glass Material and Its Application to Direct Fusion Welding by Ps-laser Pulses,” J. Laser Micro Nanoeng. 2(1), 7–14 (2007). [CrossRef]
14. A. Horn, I. Mingareev, and A. Werth, “Investigations on Melting and Welding of Glass by Ultra-short Laser Radiation,” J. Laser Micro Nanoeng. 3(2), 114–118 (2008). [CrossRef]
15. I. Miyamoto, K. Cvecek, Y. Okamoto, and M. Schmidt, “Novel fusion welding technology of glass using ultrashort pulse lasers,” Phys. Procedia 5, 483–493 (2010). [CrossRef]
16. S. Richter, S. Döring, A. Tünnermann, and S. Nolte, “Bonding of glass with femtosecond laser pulses at high repetition rates,” Appl. Phys., A Mater. Sci. Process. 103(2), 257–261 (2011). [CrossRef]
17. I. Miyamoto, K. Cvecek, Y. Okamoto, and M. Schmidt, “Internal modification of glass by ultrashort laser pulse and its application to microwelding,” Appl. Phys., A Mater. Sci. Process. 114(1), 187–208 (2014). [CrossRef]
18. S. Richter, F. Zimmermann, A. Tünnermann, and S. Nolte, “Laser welding of glasses at high repetition rates – Fundamentals and prospects,” Opt. Laser Technol. 83, 59–66 (2016). [CrossRef]
20. J. Chen, R. M. Carter, R. R. Thomson, and D. P. Hand, “Avoiding the requirement for pre-existing optical contact during picosecond laser glass-to-glass welding,” Opt. Express 23(14), 18645–18657 (2015). [CrossRef] [PubMed]
21. S. Richter, F. Zimmermann, R. Eberhardt, A. Tünnermann, and S. Nolte, “Toward laser welding of glasses without optical contacting,” Appl. Phys., A Mater. Sci. Process. 121(1), 1–9 (2015). [CrossRef]