Intracavity Coherent Addition of Lasers

Abstract

The ability to obtain high output power with good beam quality from lasers has been, and still is, a major challenge, drawing considerable scientific attention. The two key parameters in a given size laser system, power and beam quality, are inversely related, whereby high output power results in low beam quality. Excellent output beam quality (i.e., low divergence) is obtained by operating a laser with the fundamental transverse TEM00 mode (transverse mode of a laser, with a Gaussian energy distribution). This is generally achieved by inserting an aperture of small diameter inside the resonator. Unfortunately, with such an aperture, the output power is reduced since because only a small volume of the gain medium is exploited. Increasing the diameter of the aperture, results in transverse multimode operation with inferior output beam quality but considerably higher output power. Heat dissipation problems and thermal effects, which are usually dominant in high-power laser systems, often degrade the beam quality even further. Thus, in an ordinary laser configuration of given size, there exists an inherent trade-off between the output power and the output beam quality, limiting the total optical brightness of the output beam. The trade-off between output power and beam quality is not unique only to free-space conventional lasers. With semiconductor diode lasers, where the dimensions are typically small, the single-mode power from a single laser is rather low. Higher output powers are usually obtained by resorting to arrays of single lasers (diode bars and stacks). However, because the lasers are not coherent with each other the output beam quality is low, resulting in a combined output beam of low brightness. With the recent development of double-clad fiber lasers, relatively high CW (continuous wave) output powers with good beam quality are now readily achieved. The possibility to distribute the gain over very long single-mode waveguide fibers, with very good optical efficiency and low thermal distortions, already allow for single-mode output powers in the Kilowatt range. However, also with this type of lasers, single-mode output power is eventually limited by the small size of the core, whereby increasing the core size results in multimode operation with degraded output beam quality. Various approaches have been investigated over the years to obtain lasers with high output powers and good beam quality. Among these are specialized laser resonator configurations, such as Talbot or Fourier transform resonators, Vernier-Michelson resonators, evanescent waves coupling, the introduction of amplitude, and phase diffractive component elements into the resonator, or insertion of interferometric combiners inside a laser resonator. One of the most promising approaches for achieving good beam quality and high output power is to coherently add several low power and good beam quality laser distributions. In this approach several identical lasers, each with low power and excellent beam quality, are operated in a phase locked manner that enables their coherent addition. The output beam is a coherent superposition of all the individual beams; thus its power is the sum of all the powers of the individual beams, while the beam quality is as good as that of the individual beam. Because it is rather straightforward to obtain good beam quality with a low-power laser, it seems a very natural approach to try and coherently combine many such lasers. However, careful study of past investigations shows that phase locking of a large array of lasers is difficult in practice, and has been a constant challenge in the field of laser design. In this chapter, we review and present our approach for achieving efficient intracavity coherent addition of several laser distributions with novel interferometric combiners. Section 6.2 describes our basic configuration for intra-cavity coherent addition, demonstrating coherent addition of 2, 4, 16 and 25 Gaussian laser distributions. It also includes an explanation on longitudinal mode considerations and upscaling limitations. Section 6.3 presents experimental results revealing how our approach can be exploited for coherent addition of several single high-order distributions and even multimode distributions. Section 6.4 describes our techniques for improving the output beam quality of multimode laser resonators, demonstrating increased output brightness. Finally, some concluding remarks are presented in Section 6.5.