Fiber lasers have provided many advantages over other lasers since their inception. In terms of structure, the optical fiber has a large surface-area-to-volume ratio; therefore, there is usually no need to add a large heat dissipation device, making the fiber laser more compact. This provides significant advantages over traditional solid-state lasers. In addition, fiber lasers are less sensitive to microscopic contaminants and exhibit excellent optical transmission characteristics. In terms of optical transmission, as excellent optical waveguide devices, optical fibers can effectively constrain light beams without the need for cumbersome alignment procedures, thus greatly simplifying actual operations. Practical applications often have certain requirements for the polarization of light, and the laser output is usually natural or linearly polarized light. Therefore, adjustment of the polarization state of the laser output has attracted much attention. Among the currently reported methods for controlling the laser polarization state, adjustment based on linear cavity structures is prone to standing waves and hole-burning phenomena, thereby affecting the gain, whereas special optical devices (such as liquid-crystal light valves, metasurface materials, and semi-circular materials) are used outside the cavity. Wave plates regulate the polarization state of the output. Although their flexibility is high, the beam quality is low. In addition, the devices in the adjustment optical path are connected through a spatial optical path, which requires high collimation of the system. Therefore, this study reports a new all-fiber ring-cavity laser output device that introduces a side-hole fiber (SHF) into the cavity to obtain laser output in different polarization states by twisting the SHF. The proposed structure is simple and easy to operate and provides new ideas for polarization-based sensing and measurement technology.

In this study, the ellipticity and azimuth angle were used to characterize the polarization state of the laser. First, the polarization direction of the polarizer in the ring cavity, X-axis of the SHF, and polarization direction of the analyzer, were designed to coincide. After twisting the SHF, the analyzer was rotated, and the minimum and maximum values of the output laser energy were observed using a power meter. The ratio of these values represents the size of the ellipticity. Concurrently, the angle of rotation of the analyzer was determined as the output laser energy was being observed. The azimuthal angle was obtained. A laser polarization-adjustment device was also used to measure the torsion angle, which can be obtained by monitoring the change in the azimuth angle. The SHF and fiber grating (FBG) were connected in series and then embedded into the cavity of the ring laser. The changes in the output laser wavelength and energy were monitored by a spectrometer for the simultaneous measurement of torsion angle and strain.

When the twist angle θ was 2°, 53°, 91°, 160° and 180°, the output of laser was linearly polarized light. When θ was 0° and 90°, the ellipticity tan γ did not achieve the minimum value as expected. This is mainly because when the light emitted from the SHF is transmitted to the analyzer through a long ordinary optical fiber, local microbending causes the polarization state to change. In addition, measurement errors have an impact on the results. When θ was 53° and 160°, the output laser was also linearly polarized light, because φ is located near kπ (k is an integer) at this time, so tan γ is close to 0. When θ was 91°?124° and 124°?160°, the maximum variation range of tan γ was 0.02?0.48 (Fig. 4). When θ was 124°?160°, the fitting degree R² between tan γ and θ exceeded 0.98 (Table 1). When θ was 5°?97°, the azimuth angle decreased with increasing θ; when θ was 97°?180°, it increased with θ (Fig. 5). When θ was 5°?97°, the fitting degree of the azimuth angle and θ exceeded 0.98 (Table 2). The laser polarization adjustment effect was improved by changing the angle between the polarization direction of the incident light and the X-axis of the SHF when the twist angle was 0° (Fig. 6). A polarization adjustment device was used to measure the torsion angle. In the range of 5° to 97°, the sensing sensitivity of the torsion angle was -1.0757 (Fig. 7). After the SHF and FBG were connected in series and embedded in the cavity of the ring laser, the output laser intensity was 0.0489 mW; the optical signal-to-noise ratio was approximately 64 dB; and the corresponding 3 dB bandwidth was approximately 0.015 nm (Fig. 8). Within the range of 0°?220°, the output laser energy twist angle first increased and then decreased, and the output laser wavelength remained unchanged with the twist angle; within the range of 0?1533×10^-6, the output laser energy changed periodically with strain, and the output laser wavelength changed linearly with strain (Figs. 9 and 10).

Based on the experimental results, when θ is 124°?160°, the maximum variation range of the output laser tan γ is0.02?0.48, and when θ is 5°?97°, the output laser azimuth angle can both be effectively adjusted by twisting the SHF. A polarization adjustment device was used for torsion angle sensing in the range of 5°?97 °. Embedding SHF and FBG in the ring laser cavity in series, provides the advantages of high signal-to-noise ratio and high measurement accuracy compared to traditional sensors that measure the torsion angle and strain. From the experimental results, the sensitivity coefficients of SHF and FBG to torsion angle changes and strain changes can also be obtained. After monitoring the output laser wavelength change Δλ and energy change ΔP through a spectrometer, these two parameters can be measured simultaneously by substituting the torsion angle and strain matrix.