Objective Er ³⁺/Yb ³⁺ co-doped fiber laser has attracted significant attention. It is widely applied in communication, military, medical, and scientific research fields. Er ³⁺/Yb ³⁺ co-doped fiber laser operates at 1.5 μm, which is located in the low-loss transmission window of silica fiber and considered as an eye-safe wavelength band. In Er ³⁺/Yb ³⁺ co-doped fiber, Yb ³⁺ ions, which act as an excellent sensitizer, have wide absorption bands of 800--1000 nm and high absorption intensity. When Yb ³⁺ and Er ³⁺ are sufficiently close, resonance energy transfer occurs between them, which improve the luminescence efficiency at 1.5 μm. The introduction of Yb ³⁺ also increases the distance between Er ³⁺ and prevents Er ³⁺ from forming clusters. However, the amplified spontaneous emission (ASE) at 1 μm emitted by Yb ³⁺ is inclined to give rise to the bottleneck effect of Er ³⁺/Yb ³⁺ co-doped fiber lasers. The output power of 1.5-μm laser and the slope efficiency will be strongly diminished by the ASE of Yb ³⁺. In this study, a series of technical solutions to alleviate ASE is proposed to achieve high efficiency Er ³⁺/Yb ³⁺ co-doped fiber lasers. Most of the proposed methods focus on changing the structure of fiber lasers. However, optimizing the composition of the fiber core and fabricating high-performance Er ³⁺/Yb ³⁺ co-doped fiber seems to be a more fundamental approach to suppress ASE. Thus, the Er ³⁺/Yb ³⁺ co-doped fiber is prepared by modified chemical vapor deposition (MCVD) process combined with solution doping technology, and the Er ³⁺/Yb ³⁺co-doped fiber laser system is constructed. Finally, the output power and ASE spectra at different pump powers are investigated.

Methods The MCVD and solution doping technologies are used to prepare fibers by introducing a large amount of phosphorus into the core. The refractive index profile (RIP) of the fiber is also analyzed by P104. Moreover, PK2500 is utilized to test the background loss and absorption coefficient of the fiber. The ASE test structure, as shown in Fig. 4, is established to observe a 1-μm ASE spectrum, which varies with pump power. The linear relationship between the output power of the 1550-nm laser and the pump power is also observed. The laser test configuration, as shown in Fig. 6, is established to further verify the fiber laser performance. By optimizing the fiber, the slope efficiency and output power are 35.5% and 2.5 W, respectively.

Results and Discussions Fig. 2 shows the RIP of the optical fiber preform. The numerical aperture (NA) attains 0.216, and the RIP center depression is caused by the volatilization of the core phosphorus element during the collapsing process. Fig. 3 shows related optical fiber performance parameters. The background loss is about 42.15 dB/km at 1190 nm. The cladding absorption coefficient at 940 nm and the core absorption coefficient at 1535 nm are 3.58 dB/mand 34.5 dB/m, respectively. The backward 1-μm ASE spectrum changes with the pump source power, as shown in Fig. 5. The inset shows that the 1.5-μm laser power increases in synchronization. When the Yb ³⁺ pumping rate is faster than the energy transfer rate of Yb ³⁺→Er ³⁺, the ²F_5/2 energy level of Yb ³⁺ will accumulate a large number of the excited state particles, and the 1-μm signal will gradually be amplified in the laser system due to the unsaturated gain. Thus, it reduces the system pump conversion efficiency and produces parasitic lasers or even giant pulses, which can burn the fiber output end face. For this fiber, along with the 1-μm ASE power continuing to increase, the peak power is below -45 dBm from beginning to end. The 1-μm ASE does not accumulate rapidly with the increase of pump light power. The reason may be attributed to the enhancement of Yb ³⁺→Er ³⁺energy transfer efficiency by phosphorus doping. A two-level laser test platform, as shown in Fig. 6, is developed to further verify the laser performance of the homemade Er ³⁺/Yb ³⁺ co-doped fiber. Furthermore, the laser output power under each pump power of the second stage is tested, and a linear fit is made to the obtained data. Fig. 7 shows the test result with a slope efficiency of 35.5% and coefficient of association of 0.99805. When the output power reaches 2.5 W, there is still no over-rolling phenomenon, suggesting that the 1-μm ASE has no significant adverse impact on the 1.5-μm laser output.

Conclusions The Er ³⁺/Yb ³⁺co-doped double-clad fiber is successfully fabricated by MCVD and solution doping. The diameter of the core and cladding are 10 μm and 130 μm, respectively. The cladding absorption coefficient reaches 3.4 dB/m at 940 nm. The core absorption coefficient is 35 dB/m at 1535 nm. The laser test results show that the maximum slope efficiency of 1550 nm laser is 35.5%, and the output power is greater than 2.5 W. This indicates that phosphorus doping has a significant effect on the optical performance of Er ³⁺/Yb ³⁺ co-doped fiber. The fiber rapid development lays a solid foundation for further research on a 1.5-μm high-power fiber laser.