Open Access
Add to BibSonomy
Abstract
In this study, (1 + 1) × 1 side-pump couplers made of tellurite fibers were fabricated and investigated. The whole optical design of the coupler was established on the basis of ray tracing models and validated by experimental results. By optimizing the preparation conditions and structural parameters, the tested component achieved a coupling efficiency of 67.52% and an insertion loss of 0.52 dB. To the best of our knowledge, this is the first time a tellurite-fiber-based side-pump coupler was developed. The fused coupler presented will simplify many mid-infrared fiber lasers or amplifier architectures.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Given their advantages such as high conversion efficiency and specific absorption properties, mid-infrared (MIR) fiber lasers have received a growing interest in many applications [1–3]. Recently, rare-earth-doped fluoride glass fiber is a favored gain medium for MIR fiber lasers due to its desirable features. In 2009, Tokita et al. demonstrated up to 24 W output powers at 3 µm using double-clad fluoride fibers [4], while in 2022, M. Lemieux-Tanguay et al. achieved an output power of 15 W at 3.55 µm [5]. However, fluoride fibers are still limited in several specific applications due to their disadvantages such as poor chemical durability and thermal stability. Tellurite glass is a potential alternative to fluoride fibers, as it offers good physical properties, thermal stability, and chemical durability, and strong upconversion emission has been observed in rare-earth-doped tellurite glass fibers [6,7]. In 2018, cascaded Raman fiber lasers based on tellurite fibers with wavelength extensions over 3 µm and output powers in the tens of watts were demonstrated [8]. However, tellurite fiber lasers are not yet generally available outside research laboratories for several factors, including the scarcity of certain crucial tellurite fiber-based pump components.
Pump components are classified as end-pump combiners or side-pump couplers based on their pumping methods, both of which aim at coupling the pump light into the rare-earth-doped signal fiber to achieve a strong upconversion emission. In contrast, the side-pump coupler has attractive characteristics that the end-pump combiner cannot achieve, including uninterrupted signal fiber core during the fabrication process, unaltered signal fiber mode field diameter (MFD), reduced signal insertion loss, and reduced heating of the pump-end fiber Bragg grating within the laser cavity [9,10]. Moreover, the pump absorption can be evenly distributed at multi-points along the side of the signal fiber to simplify the thermal management and maintain high backward pump isolation, which can effectively protect the laser diode and component in both the co- and counter-propagating configurations [11].
The fabrication process of silica-fiber-based couplers is now well mastered, and different types of side-pumping technologies are available, including v-groove [12], embedded mirror [13], adhered micro prism [14], diffraction grating [15], angle polishing [16], tapered capillary tube [9], GT-wave technology [17], and the tapered-fused method. Among these technologies, the tapered-fused method fuses one or more tapered pump fibers parallel to the horizontal side of the signal fiber, which yields several benefits such as high coupling efficiency and the ability to handle pump powers up to the kW level. In 2012, this method was used to fabricate a silica fiber-based (4 + 1) × 1 side-pump coupler with a total pump transmission efficiency of 90.2% and the capability to handle 440 W of pump power [18]. Recently, a silica fiber-based (2 + 1) × 1 side-pump coupler was fabricated with optimized parameters, resulting in a coupling efficiency of ∼97% under a 1.87-kW pump source (NA = 0.19) injection [19]. Given that silica glass becomes opaque beyond 2 µm, a fiber coupler in soft-glass fiber (fluoride, tellurite, and chalcogenide fiber are examples) was required, which is less mechanically robust and more difficult to fabricate than silica fiber-based couplers. In addition, difficulties arises from the fact that the pigtailed silica fiber used to provide the pump diode output must be coupled with the soft-glass fiber. The glass transition temperature and thermal expansion coefficients of these two types of materials are considerably different, complicating the splicing or side coupling fusion between the two fibers. Thus, a reliable heating source with a very low temperature for fusing the tapered pump and signal fibers is required for the fabrication of a soft-glass-based side-pump coupler. These problems have slowed the research on soft-glass fiber couplers for MIR fiber lasers.
However, some notable advancements have been achieved. In 2021, using an asymmetric fusion method can reportedly decrease the loss at the effective splice point between the silica and soft-glass fibers to only 0.1 dB [20]. The stability and robustness of this splicing point are high enough for pump source injection. In 2018, a fluoride-based side-pump coupler with an 83% coupling efficiency for the fiber laser output of 15 W at 2.8 m was reported, but it depends on angle polishing, which is a highly challenging method [21]. In 2019, the 2.8-µm laser output was further increased to 33 W with the contribution of this side-pump coupler [22]. The short coupling length of the angle-polished taper is considered a limitation in increasing the efficiencies. In 2020, a fuseless side-pump coupler without any polishing and splices was proposed and obtained an efficiency as high as 93% at a 96-W pump source injection [23]. The two fibers of this coupler were adhered to by acetone and should be immersed in a low-refractive-index polymer before acetone evaporates, which is highly dependent on the chance of manual operation.
In this study, we fabricated a (1 + 1) × 1 tellurite-fiber-based side-pump coupler on the basis of the tapered-fused method and investigated the coupling efficiency through detailed simulations and experiments. The coupling efficiency is affected by the taper length and the taper ratio of the pump fiber [18]. Therefore, we built a 3D model of the coupler on the basis of the analysis according to geometrical optics theory and performed ray-tracing simulations to optimize the different combinations of parameter values to improve the coupling efficiency. To the best of our knowledge, we are the first to successfully prepare a tellurite-fiber-based side-pump coupler with a 67.52% coupling efficiency at an 80-mW pumping power by using graphite filament as the heat source. We set the preparation parameters according to the simulation results and evaluated the coupling efficiency of the fabricated couplers with a 1310-nm pump light. The experimental results are consistent with the simulation results. This coupler can be used in a tellurite-fiber-based laser setup, opening up a new track for MIR fiber side pump couplers and providing a meaningful reference for their preparation.
2. Simulation analysis
2.1 Schematic and simulation model
On the basis of the theoretical model of T. Theeg et al. [18], Fig. 1(a) depicts the construction of the (1 + 1) × 1 side-pump coupler in which a tapered pump fiber (PF) is melted onto the external surface of the signal fiber (SF). To enhance the coupling, a coreless fiber was used as the pump fiber, with a cladding diameter of 125 µm. The diameters of the signal fiber core and the cladding were 10 (NA 0.22) and 125 µm, respectively. With constant taper length, the propagation angle of the pump light in a tapered pump fiber increases by the taper ratio (TR), which is defined as the ratio between the initial fiber diameter (Do) and the diameter of the taper waist (Dwaist), leading to an increase in the pump power transfer to the signal fiber cladding. The coupling efficiency is significantly impacted by the degree of contact between the fused fibers, which is controlled by the fusion length and fusion depth. As presented in Fig. 1(a), the fusion depth is defined as FD = (2z)/(DPF + DSF), where DPF and DSF are the respective cladding diameters of the pump fiber and the signal fiber at a certain taper position, and z represents the distance of the fused pump fiber and signal fiber [18]. Pump light rays gradually couple into the signal fiber along the fusion length, which consists of the taper length (TL) and the waist length, and most of the power successfully transmits at the cladding (PTC) of the signal fiber. However, some pump power leaks from the coupler, which are classified as power leakage at the cladding (PLC), power leakage at the taper of the pump fiber (PLT), power leakage at the waist of the pump fiber (PLW), and power reflected back from the forward side of the signal fiber due to Fresnel reflection, which may result in leakage from the backside of the signal fiber (PLB) and propagates back to the pump fiber. These pump light rays, which typically have a relatively narrow propagation angle, mainly leak into the air at the waist of the pump fiber as PLW if the internal total reflection of the pump fiber is met; otherwise, the rays are released into the air as PLC and PLT. The propagation angle of the pump light in respect of the vertical decreases by 2β (β = arctan((Do – Dwaist)/TL) after each reflection in the tapered pump fiber. In order to ensure the propagation angle to satisfy the total internal reflection angle, the taper ratio (TR) and the taper length (TL) should be altered so that light is efficiently coupled into the signal fiber.
Fig. 1. (a) Schematic side view of (1 + 1) × 1 side-pumped coupler including important transmission paths and (b) Simulation model applied to ray tracing.
The 3D tellurite-fiber-based side-pump coupler model for the ray tracing simulations is presented (c.f. Figure 1(b)). References [18,24] provide detailed information on ray tracing in tapered cylindrical fibers. The laser pigtail with a 125 µm cladding diameter was spliced to the pump fiber. The length of the waist remained unchanged at 5 mm. For all the cladding structures, the refractive index of glass at 1310 nm was used. In the simulations, a simplified linear form rather than a parabolic shape was used to represent the geometrical shape of the longitudinal taper. The FD was set to 1.99 as the model allows for some glass fusion overlap between the pump fiber and the signal fiber. The rays were detected collectively at the signal fiber and the laser pigtail outputs, and the coupling efficiency was determined.
2.2 Simulation results
As mentioned, the TL and TR of the pump fiber have the greatest influence on coupling efficiency. The coupling efficiency, which is calculated by the power transmitted at the cladding (PTC) of the signal fiber compared with the output power of the laser pigtail with a NA of 0.14, is depicted in Fig. 2. The power percentage of the PLW and PLB related to the TR and the TL is also depicted. Figure 2(a) shows that the coupling efficiencies increase with the TR at a constant TL of 20 mm. For example, a TR of 3 leads to a theoretical maximum pump coupling efficiency of 57.9%, while a TR of 7 will result in a coupling efficiency of 79.58%. Furthermore, the PLW decreases as TR increases, and the percentage drops from 71.2% to 2.3%, which can be explained by the reduction of the pump fiber’s diameter that prevents most of the light rays from being confined in the taper. Therefore, most of the light propagates into the signal fiber as TR increases, which also leads to an increase in PLB. In addition, the significant reduction of the fiber mechanical stability corresponding to a TR of 10 due to the gradual reduction of the diameter from 125 µm to 12.5 µm should be considered in the subsequent preparation, and the percentage of PTC remains the same after the TR exceeds 6.
Fig. 2. (a) Transmission paths power percentage related to the taper ratio at a taper length of 20 mm and (b) related to taper length at a taper ratio of 6.
Figure 2(b) illustrates that for a constant of 6, increasing the length of the taper length results in a small improvement in coupling efficiency. For instance, the coupling efficiency obtained from the simulation was 74.6% when the TL was 5 mm, and 79.5% when the TL was 40 mm, indicating an improvement of approximately 4.9%. The pump light rays are being bounced off the lateral surface of the converging taper portion more and more often, which can explain the enhanced coupling behavior at long TLs. In contrast, PLW and PLB were not sensitive to the change in taper length and showed a slight fluctuation, remaining basically at 9% and 5.5%, respectively.
Therefore, the simulations expectedly show an increase in coupling efficiency with the increasing TR or TL, which also reveals the change in lossy power (PLW and PLB). Other power losses were incurred, that is, approximately 10%, including the previously analyzed PLT and PLC and the non-negligible light reflected to the pigtail end of the pump laser. However, increasing the TL and the TR is challenging, and the degree of fusion between fibers can also have an impact on the coupling efficiency, none of which can be easily controlled during the manufacturing process. Hence, an appropriate heating source should be identified to obtain high-quality tapered tellurite fibers to prepare side-pump couplers.
3. Experimental setup and discussion
3.1 Fabrication process of the coupler
Home-made tellurite glass rods made in the lab through an extrusion method were fed into the drawing tower to produce a fiber with a core diameter of 10 µm (NA = 0.22) and a cladding diameter of 125 µm as the signal fiber, and a polished prefabricated rod was drawn into the coreless fiber with a 125 µm diameter as the pump fiber. The core of the signal fiber consists of 70TeO2-15ZnO-5La2O3-10WO3, and the clad of the signal fiber and the pump fiber consist of 70TeO2-15ZnO-6.5La2O3-8.5WO3.
The signal fiber was coated with an ∼75 µm thick polyethersulfone coating and then wound on a fiber reel for preservation, as shown in Fig. 3(a). Figure 3(b) demonstrated the cross-sectional view of the fiber under the microscope with 1000X. The transition temperatures (Tg) and the crystallization onset temperature (Tx) of the cladding glasses of the signal fiber were ∼395 °C and 542 °C, respectively, exhibiting high thermal stability (T = Tx – Tg) of ∼147°C, which could effectively avoid the crystallization caused by high heating temperature during taper drawing and fusing. The use of a cut-back method revealed that the background fiber loss at 1310 nm was ∼1.3 dB/m, primarily including the electronic absorption, Raleigh, Mie, and wavelength-independent scattering caused by impurities and the OH- bond, introduced during the glass melting and fiber fabrication process [25].
Fig. 3. (a) Well-preserved homemade tellurite fiber (Used as a signal fiber). (b) Cross-sectional view of signal fiber taken under a microscope with 1000X.
Figure 4 represents the (1 + 1) × 1 side-pump coupler construction approach. A miniature graphite wire was employed as a heat source in the fabrication process, and the temperature can be controlled by varying its energizing power. During the fabrication process of the coupler, a pump fiber was first tapered with a specially designed taper length and waist diameter. The low loss was attained by adiabatically tapering the pump fiber, which was controlled by two system parameters: the speed of the pulling stage and the power of the heat source. The coreless glass fiber was placed on the fiber platform of an optical glass fiber processor (XQ7190 from OSCOM TECHNOLOGY), heated, and drawn into a symmetrical tapered fiber, and a coreless tapered pump fiber was obtained by cutting the waist area of the fiber. Then, the polymer coating of the signal fiber was stripped off. Thereafter, the pump fiber and the signal fiber were passed through the inner groove of the inverted Ω-type graphite wire and kept laterally by two fiber holders, allowing the tapered pump fiber was tightly bonded to the outside surface of the signal fiber without strain and distortion. A tiny quantity of ethanol was used to increase the adhesion between the two fibers through surface tension. Subsequently, the previously pump fiber and signal fiber treated were melted by rescanning the graphite wire for heating. Long-time high-temperature fusion results in high-power coupling efficiency but causes considerable signal fiber insertion loss, so the scanning speed of the heating element and the heating power of the graphite filament should be precisely controlled.
One of the main advantages of the side-pump coupler is that the signal fiber core is not changed, resulting in extraordinarily low insertion loss as compared to the tapered fused bundle end-pump combiner. However, whether the signal fiber is deformed during the heating step and affects the signal insertion loss of the coupler is worth considering. The 1310-nm diode with an output pigtail of SMF-28 is used to measure the insertion loss of the homemade coupler, and the input side of the signal fiber is coupled with SMF-28 using a UV-cured epoxy, as shown in Fig. 5(a). Detector #1 is used to detect the output power of the signal light after transmission in the coupler, and the insertion loss is calculated as IL = -10lg(Pout1/Pin). As shown in Fig. 5(b), the forward and backward output powers of the coupler were respectively measured by Detectors 2 and 3 under a 1310 nm pump light injection, and the coupling efficiency and the power share of PLB were obtained by the corresponding calculation. The calculation equations are CE = Pout2/Pin and PLB = Pout3/Pin, respectively. Also using SMF-28, the diode pigtail was coupled with the pump fiber by a UV-cured epoxy under the microscope. The loss at the curing junction was approximately 0.67 dB. To avoid introducing unnecessary curing loss when testing the coupling efficiency, the pump fiber was truncated at a distance of ∼5 cm from the junction, and the output of the test light passing through the junction was taken as the Pin.
Fig. 5. Experimental setup for measuring (a) insertion loss and (b) coupling efficiency and percentage of PLB.
3.2 Experimental results and discussion
3.2.1 Insertion loss of the coupler
Long-time heating under high temperatures can lead to signal fiber deformation and thus introduce signal insertion loss. Therefore, the power to energize the graphite wire was retained to maintain the same heating temperature, and only the heating time was changed. The heating time of the coupling region was varied in the case of TR = 6 and TL = 20 mm to investigate the effect of different heating times on the insertion loss to obtain a low-loss pump coupler. The waist area length was maintained at 5 mm. We prepared five sets of side-pump couplers according to different heating times (15, 30, 45, 60, and 75 s).
Figures 6(a)–(d) show the partial microscope pictures obtained at various areas of the manufactured side-pumping coupler when the heating time was controlled at 30 s. With proper heating time, the surface of the fiber remained smooth and flat, with no deformation, and each section was tightly fused together. Figure 6(d) shows the side view of the coupling area when the pump fiber is drawn to the finest 20 µm, and its cross-sectional view is shown in Fig. 6(e). The core-cladding structure of the signal fiber was not destroyed. Furthermore, the fusion between two fibers was indeed achieved. The fusion depth obtained by 30 s heating at the same heating temperature remains at a shallow level, which is consistent with the FD = 1.99 set in the simulation (see section 2.1).
Fig. 6. Different sections of the side-pump coupler when the pump fiber is (a) 100 µm, (b) 65 µm, (c) 35 µm, and (d) 20 µm in the down-taper region through a microscope 500X; (e) (microscope 1000X) cross-sectional view of the coupler at pump fiber of 20 µm.
Figure 7 shows the variation of the insertion loss (blue) and the corresponding coupling efficiency (red) for different heating times. The insertion loss increased from 0.11 dB at 15 s to 2.5 dB at 75 s with the increase of the heating time at 15-s intervals. After 45 s, the loss increased dramatically, which is related to the further heating of the molten fiber near the bottom of the graphite wire groove due to the downward bending and depression by gravity. Severe deformation and bending will not only increase the insertion loss further but also affect the coupling efficiency; as shown in the figure, the coupling efficiency increased from less than 30% at 15 s to 62.98% at 30 s and then gradually decreased until 40.61% at 75 s. The heating time of 15 s was insufficient to melt the two fibers together, and the actual fusion length was less than 20 mm, so the coupling efficiency at 15 s was the lowest. However, the heating time of 30 s was suitable for the two fibers to melt each other. The melted fibers are difficult to separate and could obtain the best coupling efficiency. Although extending the heating time can improve the coupling efficiency by obtaining a deeper fusion depth, it also changes the fiber morphology and introduces more losses, resulting in reduced final output power, as evidenced by the performance of several groups of couplers after 45 s. In summary, we can obtain couplers with an insertion loss as low as 0.52 dB and a coupling efficiency as high as 62.98% at a heating time of 30 s, which will be maintained for all couplers mentioned subsequently in this article.
Fig. 7. Insertion loss (blue) and coupling efficiency (red) of couplers prepared with different heating times.
3.2.2 Taper ratio of the pump fiber
As the taper length of the fiber increases, its taper angle decreases, making it more difficult to control the fiber. Therefore, the taper length was kept constant at 20 mm to increase the difficulty of the preparation, and the effect of different waist diameters on the coupling efficiency and PLB was investigated. After tapering the pump fiber, the variation of the diameter exhibited a symmetric distribution. The diameters of the left and right parts of the fiber uniformly decreased toward the taper length, remaining 10 mm long at the thinnest diameter. Truncating the fiber along the symmetry axis yielded a tapered pump fiber with a length of 5 mm in the waist region. As shown in Fig. 8(a), 12-, 20-, 30-, 40-, and 50-µm tapered fibers can be prepared by varying only the size of the waist diameter to prepare couplers of different TRs, corresponding to 10, 6, 4, 3, and 2.5, respectively. Figure 8(b) presents the tapered fibers with different finest diameters in the waist region after truncation. The fiber diameter remained linear and constant at the waist region.
Fig. 8. Diameter variation of the pump fiber (a) in the longitudinal position and (b) in the microscope corresponding to different TRs with different waist diameters.
Figure 9 illustrates the coupling efficiency (CE, red) and percentage of PLB (blue) of the coupler for different TRs the dashed lines are the simulation results mentioned earlier, and the dots are the experimental results. The coupling efficiency was the lowest at 36.97% for TR = 2.5 and the highest at 67.52% for TR = 10, and β was the largest at TR = 10 so that the propagation angle of light could be reduced quickly to couple with the signal fiber. The coupling efficiency increased with the TR and gradually saturated, showing an excellent agreement with the simulation trend. The decreasing diameter of the waist allowed more light to escape from the pump fiber and increased the probability of coupling with the signal fiber while increasing the possibility of the PLT. The actual maximum coupling efficiency is nearly 10% less than that in the simulation, but considering that the fiber itself had a loss of approximately 0.45 dB, the experiment result is nearly the same as the simulation result. The measured PLB is also very near the simulation results, where the measured and simulated values are 5.15% and 7.42% for TR = 10, respectively.
Fig. 9. Coupling efficiency (red) and percentage of PLB (blue) for different taper ratios (TR). The dashed line and dots are the simulation and experimental results respectively.
3.2.3 Taper length of the pump fiber
The pump fiber and signal fiber were fused horizontally not only at the down-taper transition but also along the waist, thus the total coupling length was the down-taper length plus the waist length. Four different tapered transition lengths were prepared, namely, 10, 20, 30, and 40 mm. To reduce the influence of the waist length, it was kept as short and constant as possible at 5 mm, thus corresponding to total coupling lengths of 15, 25, 35, and 45 mm, respectively. As shown in Fig. 10, the tapered fiber diameter kept changing linearly, both from ∼122 µm to 20 µm, and the TR at this time was approximately 6. By optimizing the heating power, the heating time, the scanning rate, and other preparation parameters in advance, we could obtain high-quality tapered fibers and cut the fibers under the CCD that came with the optical glass fiber processor, precisely controlling the length of the waist area, as shown in the figure. To demonstrate the taper variation, we purposely prepared tapered fibers with a taper length of only 1 mm, as shown in the inset.
Fig. 10. Diameter scanning of different taper lengths in the longitudinal position. Inserted diagram: taper area under the microscope when TL = 1 mm (does not discuss the coupler performance under this paper).
Figure 11 shows the coupling efficiency of the four couplers. The coupling efficiency increased with the taper length, from 61.35% at TL = 10 mm to 65.26% at TL = 40 mm. A long taper will have a long coupling length and a small β, which increases the number of light propagation back and forth the pump fiber and the signal fiber, reducing the PLW and the PLT, and further improving the coupling efficiency, although the effect is insignificant, which may be related to the fact that the backward isolation decreases as the coupling length increases. During the measurement process, the increase in TL did not affect the PLB, which was maintained at approximately 3.5%, 2% less than the simulation result, probably because the light reflected from the signal fiber in the forward direction must pass through the coupling area again and is lost in the form of PLC.
In the measurement, the coupler with an output of 54 mW and a coupling efficiency of ∼65% was placed in different environments, including on a groove of the polymer and metal, as well as in the air. The maximum temperatures in the coupling area measured by an infrared camera were 37.5 °C, 27.4 °C, and 33.5 °C, respectively. Heat accumulation is mainly caused by light leakages such as PLT and PLC, so the selection of the environment is very important for the control of the temperature in the coupling area, and active water cooling and pure copper with high thermal conductivity are often used to reduce the temperature damage to the device. Thermal management is also an important factor in coupler performance as the input power continuously increases. Thus, the temperature must be monitored to select active cooling to improve stability.
4. Conclusions
We first prepared a (1 + 1) × 1 side pump coupler on the basis of tellurite fiber and investigated the coupling efficiency of the coupler by simulation and experiment. Couplers prepared using a novel and simple graphite filament heating method allow a certain fusion depth between tellurite fibers, which is conducive to improving the robustness and reliability of the components. Through our investigations, we found that methods such as optimizing the heating time of the heating element, increasing the taper ratio of the pump fiber, and selecting a suitable taper length can help improve the coupling efficiency. A power handling of nearly 100 mW was achieved with a coupling efficiency of 67.52%, and the experimental results agree well with the ray-tracing calculations. The signal insertion loss is only 0.52 dB. The development of pump couplers that can be applied to MIR fiber lasers is an important endeavor in all-fiber technology. In future work, the structure of the signal fiber and active cooling techniques will be further optimized to improve and stabilize the performance of pump couplers.
Funding
National Natural Science Foundation of China (62090062, 62090063, 62090064, 62090065); Key R&D Program of Ningbo City (2022Z208); K. C. Wong Magna Fund in Ningbo University.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. S. D. Jackson, “Towards high-power mid-infrared emission from a fibre laser,” Nat. Photonics 6(7), 423–431 (2012). [CrossRef]
2. A. B. Seddon, Z. Tang, D. Furniss, S. Sujecki, and T. M. Benson, “Progress in rare-earth-doped mid-infrared fiber lasers,” Opt. Express 18(25), 26704–26719 (2010). [CrossRef]
3. M. Ebrahim-Zadeh and K. Vodopyanov, “Mid-infrared coherent sources and applications: introduction,” J. Opt. Soc. Am. B 33(11), MIC1 (2016). [CrossRef]
4. S. Tokita, M. Murakami, S. Shimizu, M. Hashida, and S. Sakabe, “Liquid-cooled 24 W mid-infrared Er:ZBLAN fiber laser,” Opt. Lett. 34(20), 3062–3064 (2009). [CrossRef]
5. M. Lemieux-Tanguay, V. Fortin, T. Boilard, P. Paradis, F. Maes, L. Talbot, R. Vallee, and M. Bernier, “15 W monolithic fiber laser at 3.55 microm,” Opt. Lett. 47(2), 289–292 (2022). [CrossRef]
6. W. Wang, L. Li, D. Chen, and Q. Zhang, “Numerical analysis of 2.7 µm lasing in Er3+-doped tellurite fiber lasers,” Scientific Reports 6 (2016).
7. X. Fan, K. Li, X. Li, P. Kuan, X. Wang, and L. Hu, “Spectroscopic properties of 2.7 µm emission in Er 3 + doped telluride glasses and fibers,” J. Alloys Compd. 615, 475–481 (2014). [CrossRef]
8. T. Yao, L. Huang, P. Zhou, B. Lei, J. Leng, and J. Chen, “Power scaling on tellurite glass Raman fibre lasers for mid-infrared applications,” High Power Laser Sci. Eng. 6, e24 (2018). [CrossRef]
9. C. Jauregui, S. Böhme, G. Wenetiadis, J. Limpert, and A. Tünnermann, “Side-pump combiner for all-fiber monolithic fiber lasers and amplifiers,” J. Opt. Soc. Am. B 27(5), 1011 (2010). [CrossRef]
10. Y. O. Aydin, F. Maes, V. Fortin, S. T. Bah, R. Vallee, and M. Bernier, “Endcapping of high-power 3 microm fiber lasers,” Opt. Express 27(15), 20659–20669 (2019). [CrossRef]
11. Q. Tan, T. Ge, X. Zhang, and Z. Wang, “Cascaded combiners for a high power CW fiber laser,” Laser Phys. 26(2), 025102 (2016). [CrossRef]
12. J. P. Koplow, L. Goldberg, and D. A. V. Kliner, “Compact 1-W Yb-doped double-cladding fiber amplifier using V-groove side-pumping,” IEEE Photonics Technol. Lett. 10(6), 793–795 (1998). [CrossRef]
13. J. P. Koplow, S. W. Moore, and D. A. V. Kliner, “A new method for side pumping of double-clad fiber sources,” IEEE J. Quantum Electron. 39(4), 529–540 (2003). [CrossRef]
14. T. Weber, W. Lüthy, and H. P. Weber, “Side-pumped fiber laser,” Appl. Phys. B 63(2), 131–134 (1996). [CrossRef]
15. F. Zhang, W. ChunCan, G. Rui, T. Zhi, N. TiGang, and J. ShuiSheng, “Novel grating couplers for diode-bars multi-point side-pumping double-clad fiber,” Opt. Commun. 279(2), 346–355 (2007). [CrossRef]
16. Q. Xiao, P. Yan, S. Yin, J. Hao, and M. Gong, “100 W ytterbium-doped monolithic fiber laser with fused angle-polished side-pumping configuration,” Laser Phys. Lett. 8(2), 125–129 (2011). [CrossRef]
17. H. Zhan, Y. Wang, K. Peng, Z. Wang, L. Ni, X. Wang, J. Wang, F. Jing, and A. Lin, “2 kW (2 + 1) GT-wave fiber amplifier,” Laser Phys. Lett. 13(4), 045103 (2016). [CrossRef]
18. T. Theeg, H. Sayinc, J. Neumann, L. Overmeyer, and D. Kracht, “Pump and signal combiner for bi-directional pumping of all-fiber lasers and amplifiers,” Opt. Express 20(27), 28125–28141 (2012). [CrossRef]
19. Z. Li, F. Min, T. Xin, X. Zhao, H. Li, Z. Chen, Z. Wang, and J. Chen, “Investigation of the pump coupling efficiency of a side-pumping combiner based on tapered-fused method,” Opt. Express 29(12), 17784–17794 (2021). [CrossRef]
20. L. Yang, B. Yan, R. Zhao, D. Wu, T. Xu, P. Yang, Q. Nie, and S. Dai, “Ultra-low fusion splicing loss between silica and ZBLAN fiber for all-fiber structured high-power mid-infrared supercontinuum generation,” Infrared Phys. Technol. 113, 103576 (2021). [CrossRef]
21. C. A. Schäfer, H. Uehara, D. Konishi, S. Hattori, H. Matsukuma, M. Murakami, S. Shimizu, and S. Tokita, “Fluoride-fiber-based side-pump coupler for high-power fiber lasers at 2.8 mum,” Opt. Lett. 43(10), 2340–2343 (2018). [CrossRef]
22. H. Uehara, D. Konishi, K. Goya, R. Sahara, M. Murakami, and S. Tokita, “Power scalable 30-W mid-infrared fluoride fiber amplifier,” Opt. Lett. 44(19), 4777–4780 (2019). [CrossRef]
23. S. Magnan-Saucier, S. Duval, C. Matte-Breton, Y. O. Aydin, V. Fortin, S. LaRochelle, M. Bernier, and R. Vallée, “Fuseless side-pump combiner for efficient fluoride-based double-clad fiber pumping,” Opt. Lett. 45(20), 5828 (2020). [CrossRef]
24. C. T. Chang and D. C. Auth, “Radiation characteristics of a tapered cylindrical optical fiber,” J. Opt. Soc. Am. 68(9), 1191 (1978). [CrossRef]
25. X. Jiang, J. Lousteau, B. Richards, and A. Jha, “Investigation on germanium oxide-based glasses for infrared optical fibre development,” Opt. Mater. 31(11), 1701–1706 (2009). [CrossRef]
