Open Access
Add to BibSonomy
Abstract
A low-numerical aperture (NA) confined-doped long-tapered (LCT) Yb-doped fiber is proposed and fabricated by modified chemical vapor deposition combined with solution doping technique. The LCT fiber owns the core NA of ∼0.05 and the gain dopant doping diameter ratio of ∼77%, with a core/cladding diameter of 25/400 µm at both ends and 37.5/600 µm in the middle. The laser performance is demonstrated by a bidirectional pumping all-fiber amplifier, of which a 4.18-kW single-mode (M2 factor ∼1.3) laser output is achieved with a slope efficiency of ∼82.8%. Compared with the conventional fiber, the co-pumped and counter-pumped transverse mode instability thresholds and beam quality of the LCT fiber are remarkably enhanced. Throughout the continuous operation, the LCT fiber amplifier presents high power stability with fluctuation of < 1%. These results indicate that LCT fiber has great potential in power scaling remaining excellent beam quality.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
High power fiber lasers and amplifiers, based on their advantages of high conversion efficiency, good beam quality, efficient heat dissipation, and compactness [1–3], have been broadly employed to industrial manufacturing, military defense, medical science and other fields [4,5]. Hitherto, with the development of the active Yb-doped fibers, high-power laser diodes (LDs) and passive devices, the output powers of fiber lasers and amplifiers have been unprecedentedly enhanced, achieving over 20 kW [6]. However, the remarkable evolution of continuous-wave broadband fiber lasers is hindered by nonlinear effects (NLEs), such as stimulated Raman scattering (SRS) [7–10] and transverse mode instability (TMI) [11–20], which becomes the dominant limitations of the further power scaling. To some extent, the SRS and TMI effects will be mitigated by the optimization of the configurations and parameters of the laser systems, such as adopting the proper bending diameter [21], altering the pump wavelength [22], and optimizing the pump power distribution [23]. However, the most essential and effective strategies for the suppression of these two effects are still to design and fabricate active fiber with high laser performance. Some common strategies, including decreasing the numerical aperture (NA) of the large mode area (LMA) fiber [24–26], confining Yb3+ ions doping in the fiber core [27–30] and varying core size in the longitudinal dimension [31–33], are theoretically and experimentally demonstrated to be capable of the effective restraint of the SRS and TMI effects.
Low NA (<0.06), by cutting down the NA of the fiber core, can reduce the normalized frequency V parameter under circumstance of the core diameter and center wavelength remaining unchanged. Thus, by properly designing the low NA fiber, there is only fundamental mode (FM) without higher order modes (HOMs), achieving intrinsic single-mode operation in the fiber systems. In recent years, low-NA fibers with efforts to implement robust single-mode operation have made significant breakthroughs. An ultra-low NA fiber with ∼0.038 NA and the core diameter of 35 µm was fabricated by optimized modified chemical vapor deposition (MCVD) process in conjunction with solution doping technique (SDT) [34], which the laser efficiency was ∼81% and the M2 factor was less than 1.1. One year later, an extremely low NA (<0.03) fiber with a mode field diameter of >35 µm and an effective area of >1000 µm2 was successfully fabricated [35]. In a fiber laser configuration, the efficiency of this fiber was greater than 85% (M2 = 1.04) without any sign of photodarkening. Another year passed, a free-space amplifier with the record output power of 4.4 kW was demonstrated based on a low-NA fiber with a core diameter of 23 µm and NA of 0.042 [36], which was permitted for an unprecedented modal stability. Recently, researchers replaced the inherent silica cladding with a Ge-doped cladding [37], obtaining a LMA fiber with low core NA of 0.04 and high absorption of 27 dB/m at 976 nm. An all-fiber tandem-pumped amplifier employed only ∼14 m of this fiber was constructed, which finally achieved a near-Gaussian beam (M2 ∼ 1.27) at 836 W with a high slope efficiency of ∼83%.
Confined doping, also called gain filtering, gain tailoring, or preferential gain, means that the gain dopant is selectively doped and distributed in the fiber core. For a conventional fiber, the fiber core is fully doped, and all the supported modes in the core could obtain the gain. While in a confined-doped fiber, transverse mode discrimination is established and only the modes that occupy the doped regions of the core could extract the gain. Consequently, FM becomes the dominant mode in transverse mode competition, while HOMs are suppressed. In the early stage of the observation of the TMI effect, confined doping was predicted and demonstrated to be a feasible way to improve the TMI threshold [38,39]. In 2019, our research group fabricated a confined-doped Yb/Ce co-doped aluminosilicate fiber with the Yb-ions doping diameter ratio of ∼70%, based on MCVD process [40], and the TMI threshold of confined-doped fiber was demonstrated to be about 1.74 times that of conventional fiber. Subsequently, this confined-doped fiber was adopted in a counter-pumped all-fiber oscillator, which eventually achieved 3 kW fiber laser output [41]. In the same year, researchers theoretically calculated the TMI threshold of four fiber structures [42], including confined-doped fiber, etc. The simulation results showed that the TMI threshold of confined-doped fiber was ∼1230W with a core diameter of 30 µm which was the highest among four fiber. In 2021, Wu et al. fabricated a confined-doped fiber with the core/inner cladding diameter of 40/250 µm and the relative doping ratio of 0.75 [43]. Based on a tandem-pumped amplifier, 6.2 kW signal laser was realized with an optical-to-optical efficiency of∼82.22% and it was a pity that the beam quality at 5.37 kW was not good, with the M2 factor of ∼2.
A long-tapered fiber with a gradually varying core size in the axial direction displays its preponderance in high power fiber laser systems. The small core section can effectively control the number of modes in the fiber and increase the TMI threshold, while the large core section can effectively reduce the laser power density and increase the SRS threshold. Compared with conventional fiber, long-tapered fiber has unique advantage in balancing the mitigation of both TMI and SRS effects. As early as in 2008, researchers reported a novel tapered double clad fiber with the core/1st clad/2nd clad diameters of 27/834/890 µm and the core/clad NA of 0.11/0.21 [44]. Tested in several fiber laser systems, this tapered fiber was demonstrated robust single mode operation. In 2020, a spindle-shaped Yb-doped fiber, namely tapered core and tapered inner cladding, was designed and fabricated with a core/cladding diameter of 20/400 µm at both ends and 30/600 µm in the middle [45]. A near-single-mode (M2 factor ∼1.3) laser oscillator of 3 kW was achieved with an optical-to-optical conversion efficiency of 78.4% and further power scaling was limited by SRS. After one year, the same group optimized the dimension of spindle-shaped fiber [46], which became the core/cladding diameter of 27/410 µm at both ends and 39.5/600 µm in the middle. Then, an output power of over 5 kW with the M2 factor of ∼ 1.9 and optical-to- optical efficiency of 66.6% was achieved in a bidirectional pumping amplifier. In the same year, our group fabricated a constant-cladding and tapered-core Yb-doped fiber with a constant cladding diameter (∼400 µm) and a varying core diameter (∼24 µm) at both ends and (∼31 µm) in the middle. The laser performance of this fiber was verified by other group, which a maximum power of 3.42 kW was obtained with an optical-to-optical efficiency of ∼55.2% and the M2 factor of 1.88 [47].
In this manuscript, we proposed and demonstrated a novel low-NA confined-doped long-tapered (LCT) ytterbium-doped fiber for the first time, fabricated by MCVD process in conjunction with SDT. The LCT fiber simultaneously united the advantages of low core NA, Yb-ions confined doping and long taper in longitudinal direction, which theoretically possessed excellent laser performance. In terms of fiber parameters, the core NA was ∼0.05 and the gain dopant doping diameter ratio was ∼77%, with a core/cladding diameter of 25/400 µm at both ends and 37.5/600 µm in the middle. In terms of laser performances, the co-pumped and counter-pumped TMI thresholds of the LCT fiber were over 2.1 times and over 2.2 times as much as that of the conventional fiber. Ultimately, a maximum 4.18 kW output power utilizing LCT fiber was achieved with a slope efficiency of 82.8%, based on a 976-LD bidirectional pumping fiber amplifier. In the whole experiment, signal laser remained single-mode operation, with the M2 factor of ∼1.3 at 4.18 kW. These results demonstrated the LCT fiber owned great potential to achieve high power output with excellent beam quality.
2. Fiber fabrication and characterization
The fabrication of the low-NA confined-doped long-tapered (LCT) fiber by MCVD combined with SDT could be divided into the following steps: deposition and pre-sintering of soot layers, solution soaking, dehydration, vitrification and collapsing into preforms, milling into an octagonal shape, and drawing into fiber. The key points for the LCT fiber fabrication were that, in the one hand, the manufactured preform needed to present the features of the low NA and Yb3+ ions confined doping in the fiber core, on the other hand, the preform required to be drawn into a biconical fiber by proper drawing speed and tapered period in the drawing tower.
Solution doping process played a significant role in fabricating the LCT fiber preform, which finally determined the fiber characterization and laser performances. The Yb3+ ions were leading ions without doubt, and as the co-doped ions, the Al3+, Ce3+ and F- played their indispensable roles. For taking into account both low NA and confined doping, the fiber core was divided into active and passive doping regions, where the overall NA was set to be 0.05. Hereon, active region included Yb3+, Al3+, Ce3+ and F-, and passive region contained Al3+ and F-. In other words, Yb3+ and Ce3+ were confined to a smaller area of the fiber core, while Al3+ and F- were distributed everywhere in the fiber core. Furthermore, each compound for the refractive index increments in the active and passive regions of the fiber core can be expressed as:
Fig. 1. Refractive index profile of the low-NA confined-doped long-tapered preform. (Inset: schematic diagrams of active doping region and passive doping region.)
The manufactured LCT fiber preform was milled into octogen shape and subsequently placed on the top of the drawing tower. By operating the proper drawing procedure, the preform was drawn into double-clad fiber at a periodically varying drawing speed. Ultimately, the high-quality biconical fiber was obtained. The structure diagram in the axial direction was exhibited and the microscope images of the cross section were also presented in Fig. 2(a). The core and inner cladding diameter were 25/400 µm at both ends and 37.5/600 µm in the middle, respectively. Simultaneously, the total length of the LCT fiber was about 27 m and, the length of each small-sized section, each transitional section and the large-sized section was 1.5 m, 8 m and 8 m, respectively. According to the integration of different mode diameters in longitudinal direction [45], effective mode diameter was calculated to be about 27 µm. Additionally, the concentration distributions of Yb3+, Al3+, Ce3+, and F- ions across the core region of the LCT fiber was measured by electron probe microanalysis (EPMA), as described in Fig. 2(b). Here, the results were tested by the quantitative measurement which could reflect the overall distribution tendencies of the doping elements. Yb3+ and Ce3+ ions were restrictively distributed in active region of the LCT fiber core, while Al3+ and F- ions existed in the whole core. The dotted yellow line represented the diameter of the whole core, 25 µm, and the dotted green line denoted the of Yb3+ ions doping diameter, 19.3 µm, which indicated that the ratios of Yb3+ ions doping diameter and area were ∼77% and ∼60%, respectively. Additionally, the cladding pump absorption coefficient of the LCT fiber at 976 nm was measured to be ∼0.9 dB/m, and thus, the total pump absorption was ∼24.3 dB.
Fig. 2. (a) Schematic diagram of the axial structure of the low-NA confined-doped long-tapered fiber (not to scale). The red and nattier blue areas represent the core and inner cladding, respectively. The insets are the microscope images of the cross sections of the both ends. (b) The elemental distribution of the low-NA confined-doped long-tapered fiber measured by electron probe microanalysis. The dotted yellow and green lines represent the whole core diameter and Yb3+ ions doping diameter. The inset is the dotting diagram in the fiber core.
For purpose of a proper bending condition to adapt the low NA of the LCT fiber, bending losses at different bend radiuses are calculated using the finite element method with perfectly matched layer. In order to achieve the effective single-mode operation, the bending loss of the FM is less than 0.1 dB/m and the bending loss of the HOMs is more than 1 dB/m [49,50]. Due to the biconical structure of the LCT fiber, only the bending losses of the LP01 and LP11 modes are numerically calculated in small-sized section, while those of the LP01, LP11, LP21, and LP02 modes are calculated in large-sized section. The calculated results of the bend losses and the effective mode areas (EMA) of the small-sized and large-sized sections as functions of the bend radiuses are shown in Figs. 3(a) and (b), respectively. The calculated results manifest that the supported modes of the large-sized section are more than those of the small-sized section and it is more difficult for the large-sized section to ensure the single-mode operation. Additionally, the results of EMA indicate that the large-sized section could support larger mode field area. It means that the purposes of the small-sized and large-sized sections are to control the modes number and to provide large mode area, respectively. Thus, the selection of the bend radius of the LCT fiber is herein mainly determined by the small-sized section. It is obvious that, as the increment of the bend radius, the bend losses of the transverse modes gradually decrease, while the EMAs gradually increase. In order to achieve the single-mode operation theoretically, the proper bend radius of the small-sized section is calculated and proposed, between 9 cm and 15 cm. Therefore, combined with simulation results, we consider that the proper bend radius region of the small-sized section of the LCT fiber is between 9 cm and 15 cm and the corresponding EMA is between 403 µm2 and 412 µm2. Hereon, for the small-sized section, the bend radius of 12 cm is selected and corresponding LP01 loss, LP11 loss, and EMA are 0.001 dB/m, 5.6 dB/m, and 409.3 µm2, respectively, so as to achieve the better laser performances in the laser test.
Fig. 3. (a) The bend losses and the EMAs of the small-sized section as the functions of the different bend radiuses. Simulation parameters: core/inner cladding diameter of 25/400 µm, NA of 0.05 and center wavelength of 1080 nm. (b) The bend losses and the EMAs of the large-sized section as the functions of the different bend radiuses. Simulation parameters: core/inner cladding diameter of 37.5/600 µm, NA of 0.05 and center wavelength of 1080 nm.
For a distinct comparison, a conventional fully-doped fiber was also fabricated and characterized, which the core and inner cladding diameter were 25 µm and 400 µm, respectively. Here, the refractive index profile of the conventional fiber preform was also measured, which also indicated the excellent uniformity along the axis of the preform. As a whole, the core NA of the conventional fiber was ∼0.065, as depicted in Fig. 4(a). The cladding pump absorption coefficient of the conventional fiber was measured to be ∼1.6 dB/m. The conventional fiber with the length of 15 m was adopted, and thus, the total pump absorption was ∼24 dB. To obtain a proper bend radius, the finite element method with perfectly matched layer was also employed to calculate the bending losses of the conventional fiber at different bend radiuses, as illustrated in Fig. 4(b). Herein, the bend losses of four transverse modes, such as LP01, LP11, LP21, and LP02, were calculated. According to the calculation results, the proper bend radius of the conventional fiber was selected to be 5 cm. Meanwhile, the corresponding bend losses of LP01, LP11, LP21, and LP02 modes were 0.0012 dB/m, 4.54 dB/m, 220 dB/m, and 604 dB/m, respectively, and the corresponding EMA was 352 µm2.
Fig. 4. . (a) Refractive index profile of the conventional fiber preform. (b) The bend losses and the EMAs of the conventional fiber as the functions of the different bend radiuses. Simulated parameter: core/inner cladding diameter of 25/400 µm, NA of 0.065 and center wavelength of 1080 nm.
3. Experimental setup
An all-fiberized bidirectional-pumped master oscillator power amplifier (MOPA) was constructed, as depicted in Fig. 5. In the first place, the master oscillator (MO) formed from a linear-cavity oscillator which was composed of a pair of gratings and an active fiber was pumped by three 915-nm laser diodes (LDs) from JDSU with each maximum pump power of 130 W. The high reflectivity fiber Bragg grating (HR-FBG) and the output coupler fiber Bragg grating (OC-FBG) manufactured by Teraxion provided the reflectivity of 99.950% and 11.252%, the core NA of ∼0.07, the center wavelength of ∼1080 nm, the core and inner cladding diameter of 14 µm and 250 µm, and the 3 dB bandwidth of 2.962 nm and 1.022 nm, respectively. To generate the excellent seed laser, the active fiber was employed with core NA of ∼0.07 and the core/cladding diameter of 14/250 µm, respectively. After stripped by cladding light stripper (CLS), the seed light with signal power of ∼170 W was obtained, which provided the center wavelength of ∼1080 nm and the beam quality M2 factor of ∼1.2. A specially manufactured mode field adaptor (MFA) possessed the core/cladding diameter of 14/250 µm on one end and 20/250 µm on the other, which was applied to match the signal fiber of the forward combiner.
Fig. 5. . Schematic diagram of the experimental setup. LD, laser diode; PC, pump combiner; HR-FBG, high reflectivity fiber Bragg grating; OC-FBG, output coupler fiber Bragg grating; CLS, cladding light stripper; MFA, mode field adaptor; F- and B-PSC, forward- and backward-pump and signal combiner; PD, photodetector; QBH, quartz block holder; CM, collimating mirror; BS, beam splitter; DM, dichroic mirror; NDF, neutral density filter; BSQ, beam squared.
After that, the seed light from the MO and the pump light from the forward/backward 976-nm LDs with the unstable wavelength were poured into the power amplifier (PA). The signal fiber core/inner cladding diameters of the forward pump and signal combiner (F-PSC) were 20/250 µm, respectively, of which the output fiber core/inner cladding diameters were 20/400 µm, respectively. Additionally, the signal fiber core/inner cladding diameters of the backward pump and signal combiner (B-PSC) were 30/250 µm, respectively, of which the output fiber core/inner cladding diameters were 30/400 µm, respectively. Herein, the pigtails of the commercial F-PSC and B-PSC were passive fibers manufactured by Nufern and their core NAs were both about 0.06. The above-mentioned LCT fiber with the length of 27 m was coiled in a racetrack-typed water-cooled plate with minimum and maximum bending diameter of 24 cm and 32 cm, respectively. Due to the low core NA, the above bending diameter could provide less than 0.1 dB/m of the LP01 bending loss and more than 1 dB/m of the LP11 bending loss, which ensured the property of the single-mode operation. The amplified laser propagated through CLS to strip the cladding light, which was received by a InGaAs photodetector (PD) to collect the time-domain signal and monitor the TMI process. Finally, the signal laser was exported through the quartz block holder (QBH).
As for the measurement of the laser performance, the diverging signal laser output from QBH was collimated by a collimating mirror with the focal length of 150 mm, and then, the collimated laser was split by a high-reflectivity beam splitter (BS) with the reflectivity of >99%. A dichroic mirror (DM) is utilized to further eliminate the residual pump light, and subsequently, the purer laser was attenuated by a set of neutral density filter (NDF). The beam quality evolution was recorded by the Beam Squared (BSQ) manufactured by Spiricon and the optical spectra of the output laser was also measured with an optical spectrum analyzer (OSA, AQ6370D) produced by Yokogawa.
4. Results and discussion
4.1 Laser performances in co-pumped MOPA
For a start, the laser performances of the conventional fiber and the LCT fiber were verified in co-pumped MOPA. The performance of the conventional fiber was depicted in Fig. 6. The output power versus pump power of the conventional fiber were recorded and the results were shown in Fig. 6(a), which manifested that the maximum output power of co-pumped MOPA was 1225 W with the slope efficiency of ∼81.3%. From the time- and frequency-domain signals as illustrated in Fig. 6(b), they were stable at the output power of 1174 W, which however, were obvious fluctuations at the output power of 1225 W. The fluctuations in time- and frequency-domain signals indicated the onsets of TMI. Thus, we considered that the TMI threshold of the conventional fiber in co-pumped MOPA was about 1225 W and meanwhile, the further power scaling was limited by TMI effect.
Fig. 6. Laser performance of the conventional fiber in co-pumped MOPA, (a) output power and O-O efficiency dependence on the pump power; (b) time- and frequency-domain signals at the output power of 1174 W and 1225 W.
The laser performance of the LCT fiber in co-pumped MOPA were presented in Fig. 7. The output power from the fiber amplifier continuously increased to 2561 W without power rollover and the slope efficiency was linearly fitted to ∼81%, as shown in Fig. 7(a). When operating at output power of 2561 W, as shown in Fig. 7(b), the Raman Stokes light was observed and the intensity was ∼29 dB below the signal laser, which suggested that the power scaling would be limited by SRS effect. Furthermore, the time- and frequency-domain spectrums were also observed by oscilloscope at the output power of 2561 W, which illustrated that it was absent of the TMI effect from Fig. 7(c).
Fig. 7. Laser performance of the low-NA confined-doped long-tapered fiber in co-pumped MOPA, (a) output power and O-O efficiency dependence on the pump power; (b) optical spectrum at 2561 W; (c) time- and frequency-domain signals at the output power of 2561 W.
4.2 Laser performances in counter-pumped MOPA
Then, the laser performances of the conventional fiber and the LCT fiber were also contrasted in counter-pumped MOPA. Compared with co-pumped MOPA, counter-pumped led to more outperformance. The output power versus pump power of the conventional fiber was shown in Fig. 8(a), and then, the slope efficiency of the output power was linearly fitted to ∼83.7%. The time- and frequency-domain signals at the output power of 1642 W and 1698 W were analyzed, as exhibited in Fig. 8(b). The results revealed that the time-domain signal was stable at the output power of 1642 W. However, with further increment of pump power, a mass of fluctuations occurred in time and frequency domains at 1698 W. Thus, the TMI threshold of the conventional fiber in counter-pumped MOPA was about 1698 W, and further power scaling was limited by TMI effect. Besides, no Raman light component was observed at 1698 W, as shown in Fig. 8(c). Corresponding to the appearance of TMI, the beam quality suddenly deteriorated and the M2 factor increased from 1.482 to 1.605, as shown in Fig. 8(d).
Fig. 8. Laser performance of the conventional fiber in counter-pumped MOPA, (a) output power and O-O efficiency dependence on the pump power; (b) time- and frequency-domain signals at the output power of 1642 W and 1698 W; (c) optical spectrum at the output power of 1698 W; (d) the M2 factor dependence on the output power, inset: beam profiles.
The laser performance of the LCT fiber in counter-pumped MOPA was depicted in Fig. 9. The output power versus pump power was shown in Fig. 9(a) and the experimental results manifested that the slope efficiency of the signal laser was ∼85.1%. The output laser power increased to 3785 W with the pump power, where no rollover power or optical efficiency decline was observed in the power scaling. When the main amplifier was operated at the output power of 3785 W, the optical spectrum, the time- and frequency-domain signals, and the beam quality were completed recorded, as shown in Figs. 9(b), (c), and (d), respectively. There was the noteless Raman light component in the optical spectrum at 3785 W, which illustrated a slight sign of the SRS effect. The time- and frequency-domain spectrums remained stable all the time. The beam quality of the output laser at 3785 W was depicted in Fig. 6(d) and the inset was the beam profile at the beam waist. The measured M2 factors in the x and y directions were 1.349 and 1.291, which also verified the absence of the TMI effect.
Fig. 9. Laser performance of the low-NA confined-doped long-tapered fiber in counter-pumped MOPA, (a) output power and O-O efficiency dependence on the pump power; (b) optical spectrum at the output power of 3785 W; (c) time- and frequency-domain signals at the output power of 3785 W; (d) the beam quality at the output power of 3785 W, inset: beam profile.
From the aforementioned experimental results of the co-pumped and counter-pumped main amplifier, the laser performances of the LCT fiber were superior than that of the conventional fiber. Concretely speaking, in terms of the co-pumped MOPA, the TMI threshold was remarkably enhanced, from 1225 W to over 2561 W, by replacing the conventional fiber with the LCT fiber. In terms of the counter-pumped MOPA, the immense increment of the TMI threshold was demonstrated, from 1698 W to over 3785 W. Simultaneously, the beam quality M2 factor of the conventional fiber deteriorated from ∼1.48 at seed light to ∼1.60 at 1698 W, while the M2 factor of the LCT fiber was only 1.32 at the output power of 3785 W.
The underlying physical mechanism was mainly attributed to two factors. On the one hand, the outstanding structure characteristics of the LCT fiber could effectively mitigate the TMI effect. On the other hand, compared with the conventional fiber, a better distribution of the heat load of the LCT fiber was achieved by the longer absorption length, which led to higher TMI threshold in co-pumped and counter-pumped amplifiers.
4.3 Laser performance of the LCT fiber in bidirectional pump
The output laser performance of the LCT fiber with bidirectional pump was measured and recorded, as exhibited in Fig. 10(a). By continuously increasing the 976-nm pump power to the injected counter-pumped power of 4348 W and the injected co-pumped power of 750 W, the output power from the fiber amplifier enhanced from 92 W to 4188 W, which the slope efficiency of the output power was linearly fitted to ∼82.8%. No power rollover was observed. The optical-to-optical efficiency manifested unstable feature during the power scaling process, in accordance with the feature of the unstable-wavelength pump source. The signal laser centered at the wavelength of ∼1080 nm, and the full width at half maximum increased from ∼1.5 nm of the seed light to ∼3.3 nm of 4188 W, as depicted in Fig. 10(b). There was obvious Raman Stokes light in the optical spectrum, and the intensity was ∼18 dB below that of signal laser. For the safety of the laser system, further power scaling was not pursued to avoid the exponential increase of the Raman Stokes light. When operating at the output power of 4188 W, the time- and frequency-domain signals remained stable without any fluctuation, as described in Fig. 10(c), which indicated that there was no TMI. The M2 factors and beam profile (inset) versus output powers illustrated in Fig. 10(d), and the detailed test report of the beam quality at output power of 4188 W was also exhibited in the inset. The measured M2 factors in the x and y directions were 1.310 and 1.286, which also demonstrated the absence of the TMI effect.
Fig. 10. Laser performance of the low-NA confined-doped long-tapered fiber in bidirectional pumping MOPA, (a) output power and O-O efficiency dependence on the pump power; (b) optical spectrum at several output powers; (c) time- and frequency-domain signals at the output power of 4188 W; (d) the M2 factor dependence on the output power, inset: beam quality at the output power of 4188 W.
The output power stability of the LCT fiber amplifier was monitored at ∼3.67 kW for an hour and the power was recorded with an interval of 0.133 second. As illustrated in Fig. 11, during the long-term operation, the output laser power presented a power fluctuation within a range from 3658 W to 3689 W, < 1%. The result of continuous operation indicated a high-power stability of the LCT fiber amplifier without the sign of photodarkening.
When operating at the bidirectional pumping MOPA, the laser performance of the LCT fiber was still outstanding, maintaining a high TMI threshold with excellent beam quality. Certainly, the longer fiber was also the significant factor in enhancing the TMI threshold. For achieving an adequate pump absorption, the longer LCT fiber length was utilized, resulting in a lower average heat load and the improvement of the TMI threshold. On the other hand, due to the low pump absorption coefficient and unstable wavelength pump, the LCT fiber with 27 meter is employed for meeting the adequate pump absorption. However, the longish fiber leads to the early emergence of SRS effect, influencing on the power scaling of main amplifier. Therefore, further optimization will concentrate on the improvement of the pump absorption and effective mode area of this fiber to mitigate SRS. Additionally, it is necessary to utilize the stable wavelength pump sources, which will benefit for the optimization of the fiber length.
5. Conclusion
In summary, we report on the fabrication of the low-NA (0.05) confined-doped long-tapered fiber by MCVD process combined with SDT. The Yb-ions doping diameter ratio is ∼77% and the core/cladding diameter in the middle section is 37.5/600 µm, tapering to 25/400 µm at both ends. The co-pumped and counter-pumped TMI thresholds of the LCT fiber are over 2.1 times and over 2.2 times as much as that of the conventional fiber, and simultaneously, this fiber manifests more superior beam quality. In the bidirectional pumping MOPA configuration, the laser output of 4.18 kW is obtained with the slope efficiency of 82.8%. Under the output power of 4.18 kW, the intensity of the Raman Stocks light is ∼18 dB below that of signal laser and the M2 factor is 1.3, maintaining single-mode output. The LCT fiber amplifier is continuously operated at the 3.67 kW for an hour with the power fluctuation of less 1%, manifesting a significant power stability. The above results demonstrate that the promising prospect of employing low-NA confined-doped long-tapered Yb-doped fiber to achieve high power output with high brightness.
Funding
National Natural Science Foundation of China (61975061, 61735007).
Acknowledgments
The authors wish to thank Changjin Laser Ltd. for providing fibers. The authors also thank the Analytical and Testing Centre at the HUST for performing elemental characterization of fiber samples.
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. D. J. Richardson, J. Nilsson, and W. A. Clarkson, “High power fiber lasers: current status and future perspectives [Invited],” J. Opt. Soc. Am. B 27(11), B63 (2010). [CrossRef]
2. C. Jauregui, J. Limpert, and A. Tünnermann, “High-power fibre lasers,” Nat. Photonics 7(11), 861–867 (2013). [CrossRef]
3. M. N. Zervas and C. A. Codemard, “High Power Fiber Lasers: A Review,” IEEE J. Sel. Top. Quantum Electron. 20(5), 219–241 (2014). [CrossRef]
4. O. Kalisky, “The status of high-power lasers and their applications in the battlefield,” Opt. Eng. 49(9), 091003 (2010). [CrossRef]
5. K. Zhao, H. Wang, J. Yan, and H. Shen, “Welding Fused Silica Glass by Picosecond Pulsed Laser With Low Numerical Aperture,” J. Micro Nano-Manufacturing 7(4), 1–8 (2019). [CrossRef]
6. B. Shiner, “The Impact of Fiber Laser Technology on the World Wide Material Processing Market,” in CLEO (2013), paper AF2J.1.
7. C. Jauregui, T. Eidam, D. N. Schimpf, J. Limpert, and A. Tünnermann, “Raman threshold for CW double-clad fiber amplifiers,” Opt. Express 17(10), 8476–8490 (2009). [CrossRef]
8. R. Tao, H. Xiao, H. Zhang, J. Leng, X. Wang, P. Zhou, and X. Xu, “Dynamic characteristics of stimulated Raman scattering in high power fiber amplifiers in the presence of mode instabilities,” Opt. Express 26(19), 25098 (2018). [CrossRef]
9. Q. Hu, X. Zhao, X. Tian, M. Wang, and Z. Wang, “Raman suppression in high-power fiber laser oscillator by long period fiber grating,” Results Phys. 26, 104460 (2021). [CrossRef]
10. Q. Hu, X. Zhao, X. Tian, H. Li, and M. Wang, “Raman suppression in 5 kW fiber amplifier using long period fiber grating fabricated by CO 2 laser,” Opt. Laser Technol. 145, 107484 (2022). [CrossRef]
11. T. Eidam, C. Wirth, C. Jauregui, F. Stutzki, F. Jansen, H.-J. Otto, O. Schmidt, T. Schreiber, J. Limpert, and A. Tünnermann, “Experimental observations of the threshold-like onset of mode instabilities in high power fiber amplifiers,” Opt. Express 19(14), 13218 (2011). [CrossRef]
12. F. Stutzki, H.-J. Otto, F. Jansen, C. Gaida, C. Jauregui, J. Limpert, and A. Tünnermann, “High-speed modal decomposition of mode instabilities in high-power fiber lasers,” Opt. Lett. 36(23), 4572 (2011). [CrossRef]
13. C. Jauregui, T. Eidam, H.-J. Otto, F. Stutzki, F. Jansen, J. Limpert, and A. Tünnermann, “Temperature-induced index gratings and their impact on mode instabilities in high-power fiber laser systems,” Opt. Express 20(1), 440 (2012). [CrossRef]
14. H.-J. Otto, F. Stutzki, F. Jansen, T. Eidam, C. Jauregui, J. Limpert, and A. Tünnermann, “Temporal dynamics of mode instabilities in high-power fiber lasers and amplifiers,” Opt. Express 20(14), 15710 (2012). [CrossRef]
15. H. J. Otto, C. Jauregui, T. Eidam, F. Stutzki, F. Jansen, J. Limpert, and A. Tünnermann, “Dynamics and origin of mode instabilities in high power fiber laser amplifiers,” Conference on Lasers and Electro-Optics (2012), paper CF2N.7.
16. C. Jauregui, T. Eidam, H.-J. Otto, F. Jansen, F. Stutzki, J. Limpert, and A. Tünnermann, “On the thermal origin of mode instabilities in high power fiber lasers,” Proc. SPIE 8237, 82370D (2012). [CrossRef]
17. L. Dong, “Stimulated thermal Rayleigh scattering in argon,” Opt. Express 9(6), 2365–2370 (2013). [CrossRef]
18. C. Jauregui-Misas, C. Stihler, A. Tünnermann, and J. Limpert, “Origin and evolution of phase-shifts in high-power fiber laser systems: detailed insights into TMI,” Proc. SPIE 10897, 1089704 (2019). [CrossRef]
19. M. N. Zervas, “Transverse-modal-instability gain in high power fiber amplifiers: Effect of the perturbation relative phase,” APL Photonics 4(2), 022802 (2019). [CrossRef]
20. C. Jauregui, C. Stihler, and J. Limpert, “Transverse mode instability,” Adv. Opt. Photonics 12(2), 429–484 (2020). [CrossRef]
21. R. Tao, R. Su, P. Ma, X. Wang, and P. Zhou, “Suppressing mode instabilities by optimizing the fiber coiling methods,” Laser Phys. Lett. 14(2), 025101 (2017). [CrossRef]
22. B. Yang, P. Wang, H. Zhang, X. Xi, C. Shi, X. Wang, and X. Xu, “6 kW single mode monolithic fiber laser enabled by effective mitigation of the transverse mode instability,” Opt. Express 29(17), 26366–26374 (2021). [CrossRef]
23. R. Tao, P. Ma, X. Wang, P. Zhou, and Z. Liu, “Theoretical study of pump power distribution on modal instabilities in high power fiber amplifiers,” Laser Phys. Lett. 14(2), 025002 (2017). [CrossRef]
24. V. Khitrov, J. D. Minelly, R. Tumminelli, V. Petit, and E. S. Pooler, “3 kW single-mode direct diode-pumped fiber laser,” Proc. SPIE 8961, 89610V (2014). [CrossRef]
25. W. Xu, Z. Lin, M. Wang, S. Feng, L. Zhang, Q. Zhou, D. Chen, L. Zhang, S. Wang, C. Yu, and L. Hu, “50 µm core diameter Yb3+/Al3+ /F− codoped silica fiber with M2 < 1.1 beam quality,” Opt. Lett. 41(3), 504–507 (2016). [CrossRef]
26. F. Zhang, Y. Xing, L. Liao, Y. Wang, H. Xu, Y. Chen, N. Dai, L. Yang, H. Li, and J. Li, “Thermo-optic behaviour of intrinsically single-mode low-numerical aperture (NA) ytterbium-doped fibre,” Laser Phys. 29(7), 075106 (2019). [CrossRef]
27. M. Gong, Y. Yuan, C. Li, P. Yan, H. Zhang, and S. Liao, “Numerical modeling of transverse mode competition in strongly pumped multimode fiber lasers and amplifiers,” Opt. Express 15(6), 3236 (2007). [CrossRef]
28. J. R. Marciante, “Gain filtering for single-spatial-mode operation of large-mode-area fiber amplifiers,” IEEE J. Sel. Top. Quantum Electron. 15(1), 30–36 (2009). [CrossRef]
29. C. Ye, J. Koponen, T. Kokki, J. Montiel, I Ponsoda, A. Tervonen, and S. Honkanen, “Confined-doped ytterbium fibers for beam quality improvement: fabrication and performance,” Proc. SPIE 8237, 823737 (2012). [CrossRef]
30. L. Liao, F. Zhang, X. He, Y. Chen, Y. Wang, H. Li, J. Peng, L. Yang, N. Dai, and J. Li, “Confined-doped fiber for effective mode control fabricated by MCVD process,” Appl. Opt. 57(12), 3244 (2018). [CrossRef]
31. B. Yang, H. Zhang, C. Shi, X. Wang, Z. Pan, Z. Wang, P. Zhou, and X. Xu, “High power monolithic tapered ytterbium- doped fiber laser oscillator,” Opt. Express 27(5), 7585–7592 (2019). [CrossRef]
32. Y. Ye, X. Xi, C. Shi, B. Yang, X. Wang, H. Zhang, P. Zhou, and X. Xu, “Comparative study on transverse mode instability of fiber amplifiers based on long tapered fiber and conventional uniform fiber,” Laser Phys. Lett. 16(8), 085109 (2019). [CrossRef]
33. X. Lin, Z. Zhang, Y. Chu, Y. Wang, Y. Xing, G. Chen, J. Peng, H. Li, N. Dai, and J. Li, “Fabrication and laser performance of cladding uniform core tapered fiber,” Opt. Fiber Technol. 64(April), 102561 (2021). [CrossRef]
34. D. Jain, Y. Jung, P. Barua, S. Alam, and J. K. Sahu, “Demonstration of ultra-low NA rare-earth doped step index fiber for applications in high power fiber lasers,” Opt. Express 23(6), 7407 (2015). [CrossRef]
35. Vincent Petit, Richard P. Tumminelli, John D. Minelly, and Victor Khitrov, “Extremely low NA Yb doped preforms (<0.03) fabricated by MCVD,” Proc. SPIE 9728, 97282R (2016). [CrossRef]
36. F. Beier, F. Möller, J. Nold, B. Sattler, S. Kuhn, C. Hupel, S. Hein, N. Haarlammert, T. Schreiber, R. Eberhardt, and A. Tünnermann, “TMI investigations of very low NA Yb-doped fibers and scaling to extreme stable 4.4 kW single-mode output,” Laser Congress (ASSL, LAC) (2017), paper ATu3A.2.
37. K.-J. Lim, S. Kai-Wen Seah, J. Yong’En Ye, W. W. Lim, C.-P. Seah, Y.-B. Tan, S. Tan, H. Lim, R. Sidharthan, A. R. Prasadh, C.-J. Chang, S. Yoo, and S.-L. Chua, “High absorption large-mode area step-index fiber for tandem-pumped high-brightness high-power lasers,” Photonics Res. 8(10), 1599 (2020). [CrossRef]
38. K. R. Hansen, T. T. Alkeskjold, J. Broeng, and J. Laegsgaard, “Thermally induced mode coupling in rare-earth doped fiber amplifiers,” Opt. Lett. 37(12), 2382 (2012). [CrossRef]
39. S. Naderi, I. Dajani, T. Madden, and C. Robin, “Investigations of modal instabilities in fiber amplifiers through detailed numerical simulations,” Opt. Express 21(13), 16111 (2013). [CrossRef]
40. F. Zhang, Y. Wang, X. Lin, Y. Cheng, Z. Zhang, Y. Liu, L. Liao, Y. Xing, L. Yang, N. Dai, H. Li, and J. Li, “Gain-tailored Yb/Ce codoped aluminosilicate fiber for laser stability improvement at high output power,” Opt. Express 27(15), 20824 (2019). [CrossRef]
41. Z. Zhang, F. Zhang, X. Lin, S. Wang, C. Cao, Y. Xing, L. Liao, and J. Li, “Home-made confined-doped fiber with 3-kW all-fiber laser oscillating output,” Acta Phys. Sin. 69(23), 1–6 (2020). [CrossRef]
42. J. Leidner and J. Marciante, “Three fiber designs for mitigating thermal mode instability in high-power fiber amplifiers,” Opt. Express 28(19), 28502–28517 (2020). [CrossRef]
43. H. Wu, R. Li, H. Xiao, L. Huang, H. Yang, Z. Pan, J. Leng, and P. Zhou, “High-power tandem-pumped fiber amplifier with beam quality maintenance enabled by the confined-doped fiber,” Opt. Express 29(20), 31337–31347 (2021). [CrossRef]
44. V. Filippov, Yu. Chamorovskii, J. Kerttula, K. Golant, M. Pessa, and O. G. Okhotnikov, “Double clad tapered fiber for high power applications,” Opt. Express 16(3), 1929–1944 (2008). [CrossRef]
45. L. Zeng, X. Xi, Y. Ye, H. Zhang, X. Wang, Z. Pan, Z. Wang, and X. Xu, “Near-single-mode 3 kW monolithic fiber oscillator based on a longitudinally spindle-shaped Yb-doped fiber,” Opt. Lett. 45(20), 5792–5795 (2020). [CrossRef]
46. L. Zeng, Z. Pan, X. Xi, H. Yang, Y. Ye, L. Huang, H. Zhang, X. Wang, Z. Wang, P. Zhou, X. Xu, and J. Chen, “5 kW monolithic fiber amplifier employing homemade spindle-shaped ytterbium-doped fiber,” Opt. Lett. 46(6), 1393–1396 (2021). [CrossRef]
47. Y. Ye, X. Lin, X. Xi, C. Shi, B. Yang, H. Zhang, and X. Wang, “Novel constant-cladding tapered-core ytterbium-doped fiber for high-power fiber laser oscillator,” High Power Laser Sci. Eng. 9, e1–7 (2021). [CrossRef]
48. V. Petit, T. Okazaki, E. H. Sekiya, R. Bacus, K. Saito, and A. J. Ikushima, “Characterization of Yb3 + clusters in silica glass preforms,” Opt. Mater. (Amst). 31(2), 300–305 (2008). [CrossRef]
49. K. Saitoh, Y. Tsuchida, and M. Koshiba, “Design of effectively single-mode leakage channel fibers with large mode area and low bending loss,” IEICE Electron. Express 6(7), 412–417 (2009). [CrossRef]
50. X. Wang, S. Lou, W. Lu, X. Sheng, T. Zhao, and P. Hua, “Bend Resistant Large Mode Area Fiber with Multi-Trench in the Core,” IEEE J. Sel. Top. Quantum Electron. 22(2), 117–124 (2016). [CrossRef]
