The application space for fiber lasers is increasing with their deployment in 2003 and through 2004. Recent publications state that fiber lasers will continue to increase and show improvements in continuous wave (CW) output powers, well into the kilowatts range. Further improvements will continue, in parallel, to this increase (such as maintaining the fiber laser's excellent, natural single-mode beam quality), which will have a positive, clear application impact. The area not so clearly covered is performance of fiber lasers within pulsed or high-peak power applications.
Extractable peak energies from Nd:YAG or Nd:YVO4 (Vanadate) lasers are naturally limited by the crystal and by pumping method. The two types of pumping—flashlamp and diode pumping—result in two very different performance matrixes. While diode pumping shows distinct performance differences between Nd:YAG and Vanadate, the overlap in the results between Nd:YAG and Vanadate from flashlamp pumping adds just one additional application window—high energy and long pulse. Typically, therefore, there is no application choice or question as to which product is the right one. The unique performance differences in diode pumping of Nd:YAG and Vanadate result in quite different application questions and sometimes confusion.
Figure 1. The performance of Nd:YAG versus Vanadate lasers.
Diode-pumped Nd:YAG yields very high peak powers at short pulses and low repetition rate. Values of 1–2 mJ or 50 to 100 kW at repetition rates less than 10 kHz are common. In diode-pumped Nd:YAG lasers energy or peak power starts to drop significantly after 10 kHz. This means, in general terms, applications or processes requiring energies of 1–2 mJ or peak powers of 50–100 kW are limited to lower processing speeds. On the other hand, diode-pumped Vanadate provides for higher processing speeds, but with a performance penalty. Vanadate typically provides 0.3-0.4 mJ or 25–50 kW at repetition rates of 10–20 kHz for higher processing speeds—a clear overlap at a lower energy or peak power with diode-pumped Nd:YAG. The value in Vanadate is that it can continue operation to, and above, 100 kHz, but the energy drops to 0.1 mJ or around 3–5 kW.
The only possible option line to increase speeds today would be to consider Vanadate and hope that the application could be done at lower energies or peak power. If the process or application has always been carried out with a diode-pumped Nd:YAG, there may have been little choice but to live with the current process speeds or process efficiencies. This is where fiber lasers offer an alternative as well as the potential to eliminate Vanadate completely at repetition rates starting around 10–20 kHz or at low processing speeds.
To date, pulsed fiber lasers are limited to peak output of about 3–5 kW, or pulsewidths of around 200 ns at a repetition rate of 20 kHz, up to 40 ns pulsewidth at a repetition rate of 100 kHz. These specifications have greatly limited the number of available applications. The limitations of fiber lasers are mainly due to gain medium self-saturation (competition within the laser for stored energy), the damage thresholds of the glass fiber designs and nonlinear limits—nature's natural limits. With the exception of the damage threshold of glass, these limitations can be designed around during development and manufacture. Different glass fiber designs have different damage thresholds, primarily due to the way the energy is distributed over a larger area of glass while still preserving the beam quality expected. These designs will continue to evolve over time.
Redesigning the traditional active fiber in fiber lasers is, therefore, a prerequisite for competitive application performance. For example, SPI (Southampton, UK) has developed fiber design options for its commercial products to increase this application space. Today these fiber lasers are limited to about 25 kW of peak power or about 1 mJ at short pulsewidths of around 40 ns. They have similar performance to diode-pumped Vanadate even up to repetition rates of 100 k Hz at these low pulsewidths. Currently, these specifications provide an alternative to Vanadate-based applications.
In the near future, these same fibers are expected to achieve about 50 kW or about 1 mJ at a pulsewidth of 20 ns and with a repetition rate of 10 kHz, which is now within the competitive space of even the diode-pumped Nd:YAG.
SPI has been investigating the extent that the upper limit of the repetition rate can be increased while preserving pulse energy and low pulsewidth. Based on experimental data, the company believes that fiber laser technology will easily be capable of repetition rates of 100 kHz and pulse energy of 1 mJ or 50 kW peak power, consequently, changing the processing line speeds of a significant number of applications being done today with both Nd:YAG and Vanadate lasers.
Figure 2. Fiber lasers offer an alternative for energy and repetition rate performance.
Solar cell scribing is well suited for pulsed fiber lasers. In this application, depending on the type of processing being done, contact grids to cutting and the processing speed needed, either diode-pumped Nd:YAG or Vanadate lasers are typically considered. In this application, fiber lasers offer equal performance in beam quality, which is required for the high ablation rates with small track widths. But fiber lasers go on to offer maintained energies at high repetition rates; therefore, eliminating the question of which product best fits. At a time when fossil fuels are being depleted and the need for alternative energies are at their highest, fiber lasers are offering a one product fits all the processing needs of this application space.
Fiber laser technology has come of age and should now be a prime candidate for consideration as an option when evaluating a diode-pumped Nd:YAG (or a diode-pumped Vanadate) for an application where the requirement is to significantly increase processing speeds while maintaining pulse energy.
Stuart Woods is business development director at Southampton Photonics Inc., Southampton, England. E-mail: firstname.lastname@example.org. Referenced experimental data from Y. Jeong et al, ORC, CLEO, 2003.