Open Access
Add to BibSonomy
Abstract
We demonstrate a single-stage all-fiber nanosecond amplifier with a total average power of greater than 1.4 kW by employing what we believe to be a novel multi-cavity passively Q-switched fiber laser as the seed laser. The multi-cavity seed laser adopts a piece of Yb-doped fiber (YDF) as saturable absorber (SA), and it includes two external cavities resonating at 1030 nm and an internal cavity working at 1064 nm, respectively. Using such a scheme, a stable dual-channel laser output with a total average power of >35 W, a pulse width of 45 ns, and an optical conversion efficiency of 72% operating at 1064 nm is achieved. By power scaling the multi-cavity seed laser, a dual-channel single-stage nanosecond amplifier is obtained with a single-port average power of exceeding 700 W and a pulse energy of about 7.3 mJ. To the best of our knowledge, this work is the highest average power and optical conversion efficiency for passively Q-switched all-fiber laser employing SA fiber, and the highest average power for a single-stage all-fiber nanosecond amplifier.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Pulsed fiber lasers have been widely utilized in the fields of scientific research, industrial processing, and medical treatment due to their superior conversion efficiency, compactness, and reliability [1–4]. Particularly, all-fiberized nanosecond pulsed amplifiers with high power and large energy are in great demand in various applications such as laser cleaning and surface treatments [5]. Benefited from the advancements of large-mode-area fibers and high-performance fiber components, the monolithic all-fiberized nanosecond pulsed amplifiers have made great progress. For instance, R. T. Su et al. [6] achieved a laser output with an average power of 913 W and a peak power of 28.6 kW, while Y. M. Cai et al. [7] achieved a 52 ns laser amplifier with an average power of 512 W, a pulse energy of 51.2 mJ and a peak power of 1 MW by using YDF with a core diameter of 300 µm. More recently, our team achieved 8 ns laser pulses delivering an average power of 1093 W, a pulse energy of 10.9 mJ and a peak power of 0.78 MW based on the distortion of long pulses [8]. However, due to the low power of nanosecond seed lasers, the monolithic all-fiber nanosecond amplifiers with high power and large energy are often composed of a nanosecond seed, multi-stage preamplifiers, and a main amplifier, which inevitably makes the system’s structure complex. Moreover, when the seed pulse is amplified by the pre-amplifiers and then further amplified by the main amplifier, the nonlinear effect noise such as stimulated Raman scattering (SRS) and four-wave mixing (FWM) originated from the pre-amplifiers can also be amplified in the main amplifier, hindering the scaling of average power and pulse energy [8–10]. Therefore, it is of great significance to investigate the amplification of nanosecond seed with high average power and weak nonlinear effects for optimizing all-fiberized amplifier systems.
In comparison to the conventional Q-switching devices such as fiber pigtailed acousto-optic modulator and semiconductor saturable absorbers, the rare-earth ion-doped fibers exhibit high damage threshold and can be used as an ideal SA for generating high-power nanosecond seed [11–13]. Accordingly, fibers doped with Sm [11], Yb [12], Bi [14], Cr [15], Ho [16] and other rare-earth ions [17,18] have been successfully employed as SAs to generate nanosecond pulses. Nevertheless, the direct generation of pulses using SA fibers often results in undesirably long pulse durations and temporal instability [11,19]. In the pursuit of generating stable and short pulses, T. Y. Tsai et al. [20] introduced the method of mode-area mismatch, thereby facilitating swift bleaching of the SA fiber. Exploiting the mode-area mismatch approach, Y. Lu et al. [21] realized a nanosecond laser delivering pulses as short as 140 ns, yielding an average power of 14 W at a repetition frequency of 100 kHz. Regrettably, the fused region of fibers with varying core diameters of mode-area mismatch technology cannot withstand the further increase of high power. Fortunately, V. V. Dvoyrin [22] proposed a pulsed fiber laser with cross-modulated laser cavities, merging passive Q-switching and gain-switching. This innovative strategy has achieved a stable pulse train, characterized by low temporal fluctuations and high average power. D. C. Jin et al. [23] achieved a dual-cavity all-fiber laser with an average power of 1.8 W, a pulse width of 45 ns, and an optical conversion efficiency of about 42% by using a piece YDF with a core/cladding diameter of 5/130 µm as SA fiber and fiber Bragg grating (FBG) with different central wavelengths. By expanding the core/cladding diameter of the active fiber to 10/125 µm, D. C. Jin et al. [13] further achieved a stable pulse train with a repetition frequency of 113.6 kHz, an average power of 21 W, a pulse width of 49 ns, and an optical conversion efficiency of about 50%. However, it is noteworthy that the optical conversion efficiency of such lasers remains to be improved, and further power scaling may trigger nonlinear effects such as SRS in small-core fibers.
In this work, we proposed and confirmed a multi-cavity passively Q-switched laser seed with an SA fiber for the first time, which encompasses an internal cavity operating at 1064 nm and two external cavities resonating at 1030 nm. The external cavities are Q-switched by the Yb-doped SA fiber employed in the internal cavity, and the internal cavity experiences gain-switched by the pulses generated in the external cavities. Consequently, this interplay results in the generation of pulses within the internal cavity, which are then amplified within the dual external cavities. This synergistic arrangement culminates in the realization of a stable dual-channel nanosecond pulse output with a total average power exceeding 35 W, a pulse width of 45 ns, a repetition frequency of 95.6 kHz, and an optical conversion efficiency of 72%. Moreover, we present an all-fiberized single-stage amplifier by using the high-power dual-channel nanosecond fiber seed laser. This amplifier delivers a total average power of 1405 W, constituting a single-port average power of greater than 700 W and a pulse energy of about 7.3 mJ.
2. Multi-cavity Q-switched nanosecond seed based on SA fiber
2.1 Setup of the multi-cavity nanosecond seed laser
The monolithic Yb-doped all-fiber multi-cavity seed laser consists of an internal cavity and two external cavities, as depicted in Fig. 1. In this scheme, the laser cavities have the fiber Bragg gratings (FBGs) within the internal cavity of 1064 nm and the external cavities of 1030 nm. The 1064 nm internal cavity and the 1030 nm external cavities are connected by a 2 × 2 output coupler (OC) with a coupling ratio of 50:50. The two arms at one end of the 2 × 2 OC are connected to form a reflection ring, and the internal cavity is placed on the reflection ring. Concretely, the 1064 nm internal cavity is composed by an FBG with high reflectivity of ≥ 99% and an output FBG with reflectivity of 90%, both of which are written on passive fiber with a core/cladding diameter of 10/125 µm. The 1030 nm external cavities are constructed by the reflection ring and high-reflectivity FBG I and FBG II (HR ≥ 99%) for accumulating high intra-cavity energy to bleach the absorption of the SA fiber. The FBG I and the reflection ring form the external cavity I, while the FBG II and the reflection ring show the external cavity II.
All active fibers in the external cavity I, external cavity II, and internal cavity are YDF with a core/cladding diameter of 10/125 µm, and a cladding absorption coefficient of 1.65 dB/m at 915 nm. The YDF in the two 1030 nm external cavities is 4 m in length and is pumped by a 27 W fiber-pigtailed 976 nm laser diode (LD) via a (2 + 1) × 1 signal/pump combiner. While the YDF in the 1064 nm internal cavity is 1 m in length without 976 nm pumping to function as an SA for passive Q-switching of the external cavities. Since the FBGs employed by the external cavities have high reflectivity, the generated 1030 nm radiation oscillates inside the cavity and supplies a core-pump for the internal cavity, producing pulses at 1064 nm. Besides, the pulses generated by internal cavity propagate into the active fiber of the two external cavities via the 2 × 2 OC, which can be further amplified. In addition, the two external cavities share an identical cavity length to ensure that the pulses of port I and port II are synchronized.
In this work, we use the following instruments to characterize the output performance of the system. A water-cooled power meter is used to measure the average power. A digital oscilloscope (KEYSIGHT MSOX3104T) with a sampling rate of 5 GSa/s and a bandwidth of 1 GHz is employed for capturing the temporal characteristics of the pulses. The photodetector (THORLABS DET08C/M) used in combination with the digital oscilloscope has a bandwidth of 5 GHz and a rise time of 70 ps, respectively. And an optical spectrum analyzer (YOKOGAWA AQ6370D) with a minimum resolution of 0.02 nm is adopted to measure the spectral properties.
2.2 Output performance of the multi-cavity nanosecond seed laser
Experimentally, the gain fiber of dual external cavities is synchronously pumped by the 976 nm LD. The average power, pulse width and repetition frequency at Port I and Port II are measured with the escalating total pump power of external cavities, as shown in Fig. 2(a) and Fig. 2(b). Upon augmenting the total pump power of the 1030 nm external cavities to a level of 1.8 W, stable nanosecond pulses can be generated, at which point the average power of Port I and Port II is about 0.4 W with a repetition frequency of 5.8 kHz. Figure 2(a) delineates the dependence of the average power and the optical conversion efficiency of the multi-cavity nanosecond seed on the total pump power. The average power is proportional to the pump power with a value of 35.4 W at pump power of 49 W, where the output power of Port I is 17.2 W and the output power of Port II is 18.2 W. Discrepancies in average power between Ports I and II can be ascribed to slight error in the coupling ratio of the 2 × 2 coupler, as well as subtle disparities in the power of the 976 nm LDs within the external cavities. Besides, the optical conversion efficiency of the system saturates rapidly with the growing pump power, stabilizing at approximately 72%. Benefiting from the amplification of the dual external cavities, the output average power and optical conversion efficiency of our multi-cavity nanosecond fiber seed are much higher than that in the mode-field-area mismatch method and the previously reported single external cavity method [13,24]. Furthermore, Fig. 2(b) shows that the pulse width and repetition frequency of port I and port II are identical under the same pump power, indicating that the two external cavities are synchronization well. As the pump power grows linearly, the pulse width is compressed from ∼ 290 ns to ∼ 45 ns, while the repetition frequency is increased from 5.8 kHz to 95.6 kHz. This phenomenon of pulse compression aligns with the findings elucidated by Herda et al. [25], which establish an inverse relationship between pulse shortening induced by the gain compression effect and the pump power under strong pumping conditions. Moreover, Ref. [23] indicates that the pulse width can be further shortened by using highly doped gain fiber or shorter-wavelength external cavities to reduce the length of the multi-cavity.
Fig. 2. The (a) output average power and optical conversion efficiency, (b) pulse width and repetition frequency, (c) pulse energy, (d) peak power of the multi-cavity nanosecond seed laser versus incident pump power.
The corresponding pulse energy and peak power of Port I and Port II are also calculated, and their evolution trend with the pump power is depicted in Fig. 2(c) and Fig. 2(d). At a pump power of 49 W, the pulse energy and peak power of port I are calculated to be 180 µJ and 4.0 kW, and that of port II are calculated to be 190 µJ and 4.1 kW, respectively. Although Fig. 2(b) shows that the pulses of the two ports are synchronized well in pulse width and repetition frequency, the pulse energy and peak power of Port I are slightly lower than that of Port II due to the minor variation in the average power levels between the two ports.
To explore the temporal performance of the multi-cavity nanosecond seed laser, the pulse profiles at maximum output power of 35.4 W (49 W pump power) are captured, as illustrated in Fig. 3. The oscilloscope trace measured at Port I on a time scale of 85 µs is shown in Fig. 3(a), where stable pulses with a repetition frequency of 95.6 kHz can be observed. The minimum pulse widths of Port I and Port II are about 45 ns, showing identical pulse profiles characterized by a Gaussian-like shape (see Fig. 3(b)). In addition, the output spectra of Port I and Port II at a pump power of 49 W are shown in Fig. 4(a). The output spectra of the two ports are similar, which has a central wavelength of 1064 nm, a 3 dB linewidth of 1.6 nm, and a weak SRS component in the range of 1100 nm to 1150 nm. Despite the use of highly reflective FBGs (HR ≥ 99%) tailored for 1030 nm, the 1030 nm radiation generated in the cavity at high pump power is also strong, thus showing a significant spillover in the 1030 nm wavelength band in the spectra. For Port I, the component of 1030 nm band accounts for about 6.5% of this port’ s output power, while the 1030 nm component in Port II accounts for about 4.6%. Due to the large emission cross section of YDF in the 1030 nm band, these spilled 1030 nm components are not conducive to the subsequent amplification of the seed laser. Therefore, an isolator (ISO) with bandpass filter is employed to eliminate the components of 1030 nm band and SRS wavelength band and safeguard the seed laser. This isolator boasts a filtering bandwidth of 14 nm, with a handling power capacity of 30 W. As shown in Fig. 4(b), the filtered spectra of Port I and Port II at a pump power of 49 W indicate that the SRS and 1030 nm components are successfully removed.
Fig. 3. The temporal performance of multi-cavity nanosecond seed laser. (a) The oscilloscope trace of the stable pulse train at 95.6 kHz repetition frequency. (b) Pulse profiles of Port I and Port II at a pump power of 49 W.
Fig. 4. (a) The output spectra of port I and port II at a pump power of 49 W in a linear scale, insert: the corresponding spectra in a logarithmic scale for comparison. (b) The filtered spectra of port I and port II at a pump power of 49 W in a logarithmic scale.
Therefore, our all-fiber multi-cavity passively Q-switched laser based on SA fiber successfully generates stable dual-channel nanosecond pulses with an average power of 35.4 W, a pulse width of about 45 ns, and an optical conversion efficiency of about 72%. For a single channel, the corresponding maximum pulse energy is greater than 180 µJ, and the highest peak power is greater than 4 kW. Such a dual-channel high-power nanosecond fiber seed provides an opportunity to simplify the traditional multi-stage amplifiers and realize nanosecond pulses with high power and large energy via single-stage amplifier.
3. Single-stage amplifier of the multi-cavity nanosecond seed
To evaluate the single-stage amplifying capability of the multi-cavity passively Q-switched nanosecond seed, we constructed a power amplifier as shown in Fig. 5. The all-fiber power amplifier consists of a dual-channel nanosecond seed and two amplifying sections (amplifier I and amplifier II). The structure of the seed laser is consistent with Fig. 1, while amplifier I (Amp I) and amplifier II (Amp II) are identical. The amplifying sections are composed of a 2.5-meter-long YDF with a core/cladding diameter of 100/400 µm, and a cladding absorption coefficient of 6 dB/m at 915 nm. Six pieces of 976 nm LDs with a power of 180 W each are employed to pump the active fiber of the amplifier through a (6 + 1) × 1 signal/pump combiner. It should be noted that the fiber-pigtailed ISO with bandpass filter has a core/cladding diameter of 20/125 µm, while the core/cladding diameter of the input and output signal fiber for the (6 + 1) × 1 combiner is 30/250 µm and 50/400 µm, respectively, enabling a mode field transition between the multi-cavity seed and the amplifier. And cladding pump strippers (CPS) are spliced with the active fiber to eliminate the light in the cladding. Besides, the fiber pigtailed end caps are applied at the end of the system to avoid unexpected end reflection and fiber facet damage. Notably, Amp I and Amp II are also pumped synchronously. To achieve reliable power scaling, all optical components except the end caps are placed on water-cooled heat sinks for thermal management.
After filtering the multi-cavity nanosecond seed at Port I and Port II by an ISO with bandpass filter, the average power of the spectrally pure nanosecond pulses is 15.6 W for Port I and 16.5 W for Port II. Then the pulses with a repetition frequency of 95.6 kHz are injected into Amp I and Amp II, respectively, for power scaling. Figure 6 shows the dependence of output power and optical conversion efficiency on pump power. It is discernible that the system’ s output power exhibits a linear growth trajectory with the growing pump power, culminating at a maximum value of 1405 W when subjected to a pump power of 1945 W. Notably, the Amp I and Amp II manifest closely aligned average powers, with a value of 703 W for Amp I and 702 W for Amp II at the maximum output power. And the optical conversion efficiency of the system stays almost stable after saturation, which is about 72% at the maximum output power. Moreover, the evolution of the pulse width of Amp I and Amp II with the pump power are identical, indicating their robust synchronization, as evident in Fig. 7(a). The pulse widths are widened from 45 ns of the multi-cavity seed to 62 ns at a pump power of 1945 W. The pulse widening can be attributed to pulse distortion induced by gain saturation effect [26,27]. In fiber amplifiers, the leading edge of the pulse consumes a greater proportion of the inverted population compared to the trailing edge, resulting in a higher gain in the leading edge and consequential pulse distortion and pulse width alteration. The inset in Fig. 7(a) specifically employs Amp II as an illustrative case to visually convey the process of pulse distortion as output power grows. It becomes clear that the pulse with higher output power exhibits augmented intensity in the leading edge compared to the one with lower average power, resulting in a widening of pulse width. Moreover, the pulse profiles of Amp I and Amp II at a pump power of 1945 W are the same, showing a pulse width of about 62 ns (see Fig. 7(b)). At this point, the oscilloscope trace on a time scale of 90 µs substantiates the temporal stability of the pulse train, which operates at a repetition frequency of 95.6 kHz. As a result, both Amp I and Amp II have a pulse energy of about 7.3 mJ and a peak power of 118 kW at the maximum output power.
Fig. 7. (a) The pulse width of Amp I and Amp II versus pump power, insert: the pulse profiles of Amp II at different output powers. (b) The pulse profiles of Amp I and Amp II at a pump power of 1945 W, insert: the oscilloscope trace of the stable pulse train.
The spectra of Amp I and Amp II at different output powers are illustrated in Fig. 8(a) and Fig. 8(b), respectively. With the growing output power, the spectra of the two amplifiers are broadened by nonlinear effects such as self-phase modulation. At the maximum output power, the spectral range of the two amplifiers spans from 1015 nm to 1120 nm with a 3 dB linewidth of 1.8 nm. An amplified spontaneous emission (ASE) envelope appears within the 1030 nm band of the spectra. And the difference between laser peak and ASE peak is 50 dB for Amp I, 49 dB for Amp II. In addition, no SRS peak in 1120 nm wavelength band is seen at a signal-to-noise ratio of 60 dB. This shows that the all-fiber nanosecond laser with high power and large energy achieved by single-stage amplifier has excellent performance. Although the spectra suggest potential for further expansion in the output power of the single-stage amplifier, the feasibility of such power scaling is constrained by the available pump power.
Additionally, Fig. 9(a) and Fig. 9(b) depict the beam quality of Amp I and Amp II at the maximum output power, respectively. The beam quality of Amp I and Amp II is nearly the same, with measured M2 values of approximately 11.47, 11.58, respectively. This poor beam quality can be primarily ascribed to the use of extra-large mode-area gain fiber in the amplifier, which inherently supports a multitude of modes. The pursuit of improved beam quality prompts consideration of gain fibers with reduced core diameters, exemplified by the 50/400 µm YDF. However, the smaller core diameter can introduce heightened susceptibility to SRS, and consequently impose limitations on the scalability of output power. Furthermore, due to the mode mixing effect in the extra-large mode-area gain fiber, it is arduous to investigate the nonlinear effects related to beam quality, such as transverse mode instability. Considering the tremendous development of tapered fibers in recent years, the employing of tapered fibers in combination with our high-power multi-cavity nanosecond seed in the future may enable the realization of all-fiber single-stage amplifiers with high average power and large pulse energy while maintaining high beam quality.
4. Conclusion
In this work, we successfully demonstrated a novel passively Q-switched multi-cavity all-fiber laser by employing a piece of Yb-doped fiber as SA. Stable dual-channel nanosecond pulses running at 1064 nm can be obtained with a total average power of greater than 35 W, a minimum pulse width of 45 ns, an optical conversion efficiency of 72%. For a single channel, the corresponding maximum pulse energy exceeds 180 µJ and the highest peak power is greater than 4 kW. Moreover, by power scaling the dual-channel high-power seed laser, an all-fiberized single-stage amplifier with a total average power of 1405 W is realized, which delivers a single-port average power of greater than 700 W and a pulse energy of about 7.3 mJ. To our knowledge, this is the highest average power of an all-fiber nanosecond amplifier achieved by single-stage amplifier. The results show that our multi-cavity passively Q-switched fiber seed laser can simplify the conventional multi-stage amplifiers, supplying a novel scheme for solving the nonlinear effect accumulation in the multi-stage amplifiers. Furthermore, our single-stage amplifier with high power and large energy affords a promising source for industrial processing such as laser ablation and surface treatments, and this synchronous dual-channel laser output is also of great benefit to enhance processing efficiency.
Funding
CAS Project for Young Scientists in Basic Research (YSBR-065); National Natural Science Foundation of China (62175230, 62225507, U2033211); Scientific Instrument Developing Project of the Chinese Academy of Sciences (YJKYYQ20200001); National Key Research and Development Program of China (2022YFB3607800).
Acknowledgment
The authors of this paper would like to thank Professor Jing-yuan Zhang for his precious time and valuable suggestions in the preparation and revision of the manuscript.
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available but may be obtained from the authors upon reasonable request.
Reference
1. V. Veiko, Y. Karlagina, M. Moskvin, et al., “Metal surface coloration by oxide periodic structures formed with nanosecond laser pulses,” Opt. Lasers Eng. 96, 63–67 (2017). [CrossRef]
2. K. Tsubakimoto, H. Yoshida, and N. Miyanaga, “600 W green and 300 W UV light generated from an eight-beam, sub-nanosecond fiber laser system,” Opt. Lett. 42(17), 3255–3258 (2017). [CrossRef]
3. T. Heiderscheit, N. G. Shen, Q. H. Wang, et al., “Keyhole cutting of carbon fiber reinforced polymer using a long-duration nanosecond pulse laser,” Opt. Lasers Eng. 120, 101–109 (2019). [CrossRef]
4. L. D. Zhang, J. Y. Zhang, X. Wang, et al., “Design of coherent wideband radiation process in a Nd3+-doped high entropy glass system,” Light: Sci. Appl. 11(1), 12 (2022). [CrossRef]
5. F. D. Zhang, H. Liu, C. Suebka, et al., “Corrosion behaviour of laser-cleaned AA7024 aluminium alloy,” Appl. Surf. Sci. 435, 452–461 (2018). [CrossRef]
6. R. T. Su, P. Zhou, X. L. Wang, et al., “Kilowatt high average power narrow-linewidth nanosecond all-fiber laser,” High Power Laser Sci. Eng. 2, e3 (2014). [CrossRef]
7. Y. M. Cai, Q. X. Zhu, J. H. Shi, et al., “High-Power Yb-Doped and All Fiber-Based Nanosecond MOPA Laser,” IEEE Photonics J. 14(5), 1–5 (2022). [CrossRef]
8. C. Y. Ning, S. Z. Zou, H. J. Yu, et al., “Monolithic all-fiberized nanosecond laser with kilowatt average power and megawatt peak power,” Opt. Laser Technol. 157, 108701 (2023). [CrossRef]
9. W. Liu, S. Ren, P. F. Ma, et al., “Impact of amplified spontaneous emission noise on the SRS threshold of high-power fiber amplifiers,” Chin. Phys. B 32(3), 034202 (2023). [CrossRef]
10. H. Y. Ying, J. Q. Cao, Y. Yu, et al., “Raman-noise enhanced stimulated Raman scattering in high-power continuous-wave fiber amplifier,” Optik 144, 163–171 (2017). [CrossRef]
11. Y. Lu and X. J. Gu, “All-fiber passively Q-switched fiber laser with a Sm-doped fiber saturable absorber,” Opt. Express 21(2), 1997–2002 (2013). [CrossRef]
12. S. W. Moore, D. B. S. Soh, S. E. Bisson, et al., “400 µJ 79 ns amplified pulses from a Q-switched fiber laser using an Yb3+-doped fiber saturable absorber,” Opt. Express 20(21), 23778–23789 (2012). [CrossRef]
13. D. C. Jin, R. Y. Sun, S. Y. Wei, et al., “Nanosecond Yb-doped monolithic dual-cavity laser oscillator with large core fiber,” IEEE Photonics Technol. Lett. 27(14), 1477–1480 (2015). [CrossRef]
14. V. V. Dvoyrin, V. M. Mashinsky, and E. M. Dianov, “Yb-Bi pulsed fiber lasers,” Opt. Lett. 32(5), 451–453 (2007). [CrossRef]
15. B. Dussardier, J. Maria, and P. Peterka, “Passively Q-switched ytterbium- and chromium-doped all-fiber laser,” Appl. Opt. 50(25), E20–E23 (2011). [CrossRef]
16. A. S. Kurkov, E. M. Sholokhov, and O. I. Medvedkov, “All fiber Yb-Ho pulsed laser,” Laser Phys. Lett. 6(2), 135–138 (2009). [CrossRef]
17. S. Yang, Y. Y. Yang, L. Zhang, et al., “Alternated Q-switched and gain-switched dual-pulse Yb fiber laser with Tm-Ho co-doped fiber as saturable absorber,” Opt. Laser Technol. 113, 159–163 (2019). [CrossRef]
18. T. Y. Tsai and Z. C. Lee, “1030-nm passively Q-switched ytterbium-doped fiber laser using excited-state absorption of thulium fiber,” Opt. Lett. 45, 17–20 (2020). [CrossRef]
19. A. A. Fotiadi, A. S. Kurkov, and I. Razdobreev, “All-fiber passively Q-switched Ytterbium laser,” in Conference on Lasers and Electro-Optics (CLEO), (IEEE, 2005). [CrossRef]
20. T. Y. Tsai, Y. C. Fang, H. M. Huang, et al., “Saturable absorber Q- and gain-switched all-Yb3 + all-fiber laser at 976 and 1064 nm,” Opt. Express 18(23), 23523–23528 (2010). [CrossRef]
21. Y. Lu and X. J. Gu, “Kilowatt peak power pulses from a passively Q-switched Yb-doped fiber laser with a smaller-core Yb-doped fiber as a saturable absorber,” IEEE Photonics J. 6(3), 1–7 (2014). [CrossRef]
22. V. V. Dvoyrin, “Pulsed fiber laser with cross-modulation of laser cavities,” in Conference on Lasers and Electro-Optics (CLEO), (IEEE, 2012). [CrossRef]
23. D. C. Jin, R. Y. Sun, H. X. Shi, et al., “Stable passively Q-switched and gain-switched Yb-doped all-fiber laser based on a dual-cavity with fiber Bragg gratings,” Opt. Express 21(22), 26027–26033 (2013). [CrossRef]
24. J. Q. Zhou, Y. Lu, B. He, et al., “Q-switched laser in an SMS cavity for inhibiting nonlinear effects,” Appl. Opt. 54(19), 6080–6084 (2015). [CrossRef]
25. R. Herda, S. Kivisto, and O. G. Okhotnikov, “Dynamic gain induced pulse shortening in Q-switched lasers,” Opt. Lett. 33(9), 1011–1013 (2008). [CrossRef]
26. L. M. Frantz and J. S. Nodvik, “Theory of pulse propagation in a laser amplifier,” J. Appl. Phys. 34(8), 2346–2349 (1963). [CrossRef]
27. D. N. Schimpf, C. Ruchert, D. Nodop, et al., “Compensation of pulse-distortion in saturated laser amplifiers,” Opt. Express 16(22), 17637–17646 (2008). [CrossRef]
