Not just fast - Ultrafast

Fiber laser technology has made a strong impact on industrial applications in the past several years, displacing solid-state lasers in laser marking and materials processing applications.

Th 0701ilsnotjustfast01

Emerging micromachining applications call for industrially rated femtosecond fiber lasers

Fiber laser technology has made a strong impact on industrial applications in the past several years, displacing solid-state lasers in laser marking and materials processing applications. Ultrafast fiber lasers, which provide pulses that are in the femtosecond range in duration, are poised to make a similar impact, with new applications enabled by the unique capabilities of femtosecond precision material modification and ablation. In particular, new fiber laser based technology enables precision laser processing of transparent materials used in microelectronics, display, and telecom applications.

Th 0701ilsnotjustfast01
FIGURE 1. 50μm circles made with varying pulse energy and repetition rate show a comparison between high pulse energy/low repetition rate and low pulse energy/high repetition rate.
Click here to enlarge image

The technology used for producing ultrafast lasers has traditionally been solid-state based, with Ti:Sapphire crystals the favored gain medium. Over the past 15 years, ultrafast rare-earth-doped optical fiber lasers have been developed as an alternative, designed to provide a more compact and industrially qualified laser source with greater robustness and reliability than other ultrafast lasers. A side benefit of this technology base is the ease with which higher repetition rates (from 100s of kHz to MHz), with moderate pulse energies, are obtained as compared to solid-state based technology.

Materials processing

The remarkably clean ablation properties exhibited by ultrashort laser pulses stem from extremely high peak power and short pulse duration, which ionizes and vaporizes materials quickly before thermal effects such as heat diffusion can occur.1 Because of this, the energy in the irradiated region of the target is not lost to the surrounding area during the short time that the laser pulse is depositing energy in the material. This localizes the heating and ionization almost exclusively to the irradiated area, and thereby lowers the threshold for ablation. Moreover, the heated and vaporized material is expelled quickly before it can heat the neighboring areas of the target material. This results in a reduction of the heat affected zone and the shock affected zone, as contrasted with nanosecond laser sources, where ablation is based on thermal processes. Less material is melted and re-fused, resulting in less recast, and less debris is scattered in the vicinity of the ablation zone.

To meet demands for industrial 24/7 laser micromachining with ultrashort laser pulses, IMRA America Inc. developed an Ytterbium-based fiber chirped-pulse amplification laser capable of operating at repetition rates between 100 kHz and 5 MHz. This ultrafast laser balances the need for higher repetition rate with that of sufficient pulse energy to work at or near ablation threshold, while meeting industrial standards for ambient temperature, shock, and vibration. The pulse energy is on the order of 1-10 µJ, sufficient for many processes.

Th 234533
FIGURE 2. Glass welding setup and results.
Click here to enlarge image

Processing speed and micromachining quality are improved by using higher-repetition-rate, µJ-level pulses. Figure 12 illustrates the creation of 50-µm diameter circles on glass. Following the red arrows from left to right, continuous improvements in results are seen when moving from 100 µJ at 1 kHz to 4 µJ at 100 kHz. Significant damage surrounding the processed area on the glass surface and poor edge quality can be seen in the results with high pulse energy and low repetition rate (far left photo). Reducing pulse energy to 4 µJ eliminates the surface damage, while still producing a rough cut edge. Slowing down the translation speed while maintaining the pulse energy creates a smoother cut edge, but with longer total processing time, which increases from 1.6 s to 50 s. Operating at 100 kHz and 4 µJ improves the cut quality and reduces the process time to 0.5 s (far right photo).

Refractive index modification

Production of low-loss waveguides for communications and sensor devices in glasses can be easily achieved using a femtosecond fiber laser operating at low pulse energies and megahertz repetition rates. The refractive index modification necessary to make waveguides can be performed directly with the 1µm output in some materials. In others, better results occur when second harmonic output is used, obtained by frequency doubling using standard conversion crystals. Waveguides were produced with second harmonic output (522 nm) at 1 MHz in fused silica.3 The size of the area that is modified can be adjusted by writing the waveguides at various speeds of moving the glass on motorized translation stages. Comparisons have been made showing molecular structure differences between waveguides written with a repetition rate of 1 MHz and those written with 1 kHz. These comparisons have shown that higher repetition rate produces fewer defects and higher index change.4

Using femtosecond-pulse micromachining, glass surfaces can be bonded at the unexposed interface of two pieces, as opposed to conventional laser welding, where the region to be bonded must be exposed. Glass welds have been demonstrated using low pulse energy and high repetition rate from a femtosecond fiber laser.5 This weld process is relatively quick and simple, in part due to nonlinear absorption of energy coupled to the glass, which allows for welding without the addition of specially designed linearly absorbing media (see Figure 2). Such welds could be useful in the development of microfluidic devices, and can potentially be made in other dielectric materials.

Th 0701ilsnotjustfast03
FIGURE 3. Cross-sectional images of scribe lines in a silicon wafer by fundamental wavelength using 10 µJ pulse energy; polarization parallel (a) perpendicular (b) to the scanning direction (c) scribe depth at various scan speeds.
Click here to enlarge image

Scribing of wafers used in the production of electronics devices is a necessary step. As an alternative to diamond scribing, laser scribing can provide superior results in terms of clean and stress-free cuts. Laser scribing at high rep rate is necessary to achieve fast processing times required in manufacturing, and as the wafer or substrate becomes thinner, higher precision is required. Thinner silicon wafers are needed for applications where System in Package (SiP) is being used, such as mobile electronics and IC cards. Thinner wafers are more fragile and difficult to dice using conventional diamond blade technology unless slower speeds are used to produce sufficient cut quality. Ultrashort pulse laser scribing may provide the solution to overcome difficulties with blades, including chipping and cracking. As with other laser materials processing applications, polarization state of the light can play a role in the process results, as shown in Fig. 3.6 With polarization parallel to the scanning direction, the scribe depth can be as deep as 50 μm, twice as deep as with perpendicular polarization with the same pulse energy and scanning speed (20 mm/s). The width of surface modification was ~12 μm.

Conclusion

Increased speed and precision make ultrafast, ultrashort pulsed fiber lasers ideally suited for the emerging and growing market requiring greater precision for materials processing based upon ultrashort-pulse material ablation. Commercial solutions for μJ-level femtosecond fiber lasers extend the reach of fiber lasers in the industrial laser marketplace.

Michelle L. Stock, Ph.D., (mstock@imra.com) is marketing manager for IMRA America Inc., Ann Arbor, MI.

References

  1. Ultrafast Lasers: Technology & Applications, Marcel Dekker Inc., New York, 2003, Ch. 7.
  2. A. Arai, J. Bovatsek, F. Yoshino, G. Sucha, M. Stock, and Y. Uehara, Laser Applications International (2006).
  3. 3. L. Shah, A.Y. Arai, S.M. Eaton and P.R. Herman, Opt. Express, 13, 1999 (2005).
  4. W. Reichmann, D. Krol, L. Shah, F. Yoshino, A. Arai, S. Eaton, and P.R. Herman, CME2 OSA CLEO (2006).
  5. J. Bovatsek, A. Arai, and C.B. Schaffer, CThEE6 OSA CLEO (2006).
  6. A, Arai, J. Bovatsek, Z. Liu, F. Yoshino, G. Cho, L. Shah, M. Fermann, and Y. Uehara, Proceedings of SPIE Vol. 6343 (2006).

More in Micromachining