Significance A fiber gas laser (FGL) based on gas filled hollow-core fibers (HCFs) is a laser source that combines the advantages of traditional gas and fiber lasers. According to the operating mechanism, FGLs can be divided into two categories: one is based on the stimulated Raman scattering (SRS) of gas molecules, and the other is based on population inversion realized by intrinsic absorption of gas molecules between the vibrational-rotational energy levels. The threshold power for population inversion is much lower than that for the SRS effect, making it easier to realize continuous-wave (CW) laser emission. The laser wavelengths corresponding to the vibrational-rotational energy level transition of most gas molecules are in the mid-infrared waveband; thus, the FGLs based on population inversion provide a novel method for the mid-infrared fiber lasers, which have wide applicability in military, biomedicine, and atmospheric communication fields. In HCFs, most of the mode energy is concentrated in the hollow-core. The laser mode edge overlaps with a small amount of glass material in the cladding. As the field intensity in the glass material is at least one order of magnitude smaller than the peak field intensity in the core region, the theoretical damage threshold of HCFs is much higher than that of solid-core fibers, making HCFs ideal to operate at much higher power level. Meanwhile, the hollow-core structure can be filled with various gain gas media to achieve plenty of laser wavelengths, especially beyond 4 μm, which is very difficult for traditional rare-earth-doped fiber lasers. Because of the intrinsic properties of gas molecular energy levels, laser output with narrow linewidth (several-hundred MHz) can also be obtained without additional linewidth control technology for FGLs, which has great advantages in maintaining laser linewidth at high power compared with lasers using solid-core fibers. Hence, FGLs provide a universal solution for the technical bottlenecks encountered by traditional mid-infrared fiber lasers in power enhancement and wavelength expansion.

Progress The advent of HCFs greatly promotes gas laser development, as it provides an ideal interaction environment between light and gas molecules. Since the first FGL based on population inversion was reported in 2011, it has obtained great attention because of its potential advantages in generating effective mid-infrared laser emission. In recent years, with the fast development of anti-resonant HCFs with low transmission loss in the mid-infrared waveband, FGLs operating at the mid-infrared waveband have been intensively studied recently. FGLs based on C₂H₂-, CO-, CO₂-, N₂O-, I₂-, HBr-, and HCN-filled HCFs have been reported, and most laser wavelengths are within the range of 3--5 μm, except for the I₂ laser, which operates at the 1.3-μm band.

In 2017, Xu et al. achieved the highest laser power using C₂H₂-filled anti-resonant HCFs pumped by a narrow linewidth 1.5 μm diode laser amplified by an erbium-doped fiber amplifier (Fig. 7). The highest continuous output power at 3.1 μm is 1.12 W at 0.6 mbar pressure, and the slope efficiency is as high as approximately 33%. The HCF can effectively confine both the gases and the pump light within the core area over a distance much longer than the length of traditional gas cells, greatly reducing the pump threshold and improving the conversion efficiency. In 2017, Dadashzadeh et al. studied the output laser beam quality of FGLs based on C₂H₂-filled Kagome HCFs (Fig. 8). The experimental results show that the mid-infrared FGL has good beam quality as traditional fiber lasers, the best M² factor measured is less than 1.4, and the best value is approximately 1.15, showing the beam quality near the diffraction limit. In 2019, we achieved the 4.3 μm CW FGLs based on CO₂-filled anti-resonant HCFs (Fig. 10), which is also the first CW fiber laser with output wavelength larger than 4 μm. The pump source is a self-developed thulium-doped fiber amplifier seeded by a tunable narrow linewidth 2-μm diode laser. And the pump source is employed to pump a low transmission loss anti-resonant HCF with a length of 5 m, which is filled with low-pressure (several mbar) CO₂. At the optimal pressure of 500 Pa, the laser threshold and the maximum output power are approximately 100 and 80 mW, respectively, with a laser slope efficiency of approximately 9.3%. In 2019, Aghbolagh et al. reported that a 45-cm long Kagome HCF filled with N₂O gas was pumped with a 1.517-μm-band OPO to produce a 4.6-μm band laser with maximum output energy of 75 nJ under a pressure of 80 Torr (1 Torr≈133 Pa). However, the laser slope efficiency is only 3% because of high transmission loss of HCFs.

Conclusions and Prospect While fiber laser and gas laser technologies have reached a high level of maturity because of intense research over the last 50 years, the FGL is just in its infancy, and there are still many basic physical issues and key technologies that need in-depth investigation, such as theoretical FGL models especially at high-power, basic gas parameters' measurement, further reduction of HCFs' transmission loss, and efficient and high-power coupling technology between HCFs and solid-core fibers.

An all-fiber structure is one of the major development directions of FGLs in the future as it is an ideal choice in practical applications. However, presently, the pump light is usually coupled into HCFs by the spatial optical path coupling method. The spatial coupling structure is unstable and easily influenced by the external environment, leading to decreased coupling efficiency. To realize all-fiber FGLs, we need to resolve the following key issues: low-loss coupling between HCFs and solid-core fibers and fabrication of high-stability low-pressure all-fiber gas cells.

Another important direction for the development of mid-infrared FGLs is to achieve high-power output. Compared with the traditional solid-core fiber, most of the mode field energy in HCF is concentrated in the hollow-core region. The overlap area between the glass material and core area is very small. Therefore, theoretically, the damage threshold is much higher than that of the solid-core fiber, which is very potential for its high-power output. However, the highest power reported is only at the watt level. In the future, we will resolve several key issues to achieve higher power output, mainly including a high-power theoretical model, suitable narrow linewidth high-power pump source, and low-loss coupling of the high-power pump laser.

Obtaining more abundant laser wavelengths is also an important direction in the future. Compared with solid-core rare-earth-doped fiber, gas gain media are more convenient to be replaced in GFLs, and there are more choices. Suppose the HCFs' transmission bands are designed properly, with suitable gases and pump sources. In that case, we can obtain lots of laser wavelengths, especially in the mid-infrared band, which are not easy to achieve with traditional fiber lasers. In the future, if we use soft glass to manufacture HCFs, it is expected that far-infrared FGLs can be realized. On the other hand, FGLs also have certain advantages in realizing laser output in the visible and ultraviolet bands. Especially in the ultraviolet band, the photon darkening effect of HCFs is much weaker than that of solid-core fibers. The choice of gain media in the visible and ultraviolet bands is very rich, including common inert gases, various chlorides, and metal vapors. The pumping method can use optical pumping and electric excitation.