We present a MHz level high repetition rate gain-switched Ho-doped fiber laser at 2.103 μm using a h-shaped mode-locked Tm-doped fiber laser as the pump, for the first time. A NPR mode-locked Tm-doped ring cavity was designed as the seed, which could generate h-shaped pulses at 1.985 μm with a fundamental repetition rate of 1.435 MHz. It had a pump-dependent ns scale pulse duration with an almost unchanged peak amplitude. Then its output power and pulse energy were significantly scaled to 3.92W and 2.71 μJ, respectively based on a one-stage Tm-doped fiber based amplifier. After that, the amplified pulses were fed to a linear cavity Ho-doped fiber laser formed by a HR FBG and a perpendicularly cleaved fiber end, resulting in stable gain-switched pulses at 2.103 μm with a repetition rate of 1.435 MHz, which marks the record of gain-switched fiber lasers in this spectral region. At this repetition rate, the maximum output power of 1.13 W and pulse energy of 0.79 μJ were achieved at a high slope efficiency of 75.4% with respect to its absorbed pump power, giving the shortest pulse duration of 10.8 ns. The results indicated that the h-shaped pulses were an alternative choice for high repetition rate gain-switching in an all-fiber configuration.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Pulsed fiber lasers in the 2 μm spectral region have attracted increasing attention in the recent decade, owing to their potential and widespread applications in remote sensing (e.g., coherent Doppler lidar wind detection [1,2]), laser surgery (e.g., ablation , lithotripsy ), polymer material processing (e.g., plastic ), pumping for longer wavelength mid-infrared (Mid-IR) generation via nonlinear frequency conversion (e.g., OPO ) and Mid-IR supercontinuum generation , etc. For ns~μs scale short pulse generation, Q-switching and gain-switching were usually adopted. Compared to fast developed Q-switching technique strongly pushed by the rise of nonlinear materials e.g., graphene , topological insulator , black phosphorus , transition metal dichalcogenides , etc., gain-switching implemented by pulse pumping always remained its distinct advantages. First, it is more compact and simple in structure, since no additional intra-cavity devices are required except laser gain and feedbacks. Second, it is easier for output power/energy scaling which is usually limited by the damage threshold of the saturable absorber for passive Q-switching. Third, its output characteristics can be controlled by tuning the pump source thus is more flexible. Fourth, gain-switching is a more common approach owing to its wavelength independence. So far, many demonstrations about 2 μm gain-switched fiber lasers have been reported [12–19], including some proposed new ways e.g., fast switching , hybrid pumping  for better performance. Most efforts were deployed based on the platform of Tm-doped fiber (TDF) laser since its rich and mature pump sources available, e.g., ~0.79 μm , ~1.06 μm [15,16], ~1.55 μm [12,17], and ~1.9 [18,19], which can provide efficient emission at the typical range of 1.9~2.05 μm. Recently, we extended the longest laser wavelength of TDF laser based on 3F4→4H6 transition to 2.198 μm by using a pair of FBGs to select the cavity gain, however, the small emission cross section at this long wavelength led to more populations participated in the parasitic processes and consequently a low slope efficiency of 17.82% . In some cases, efficient laser at longer wavelengths than 2.05 μm is eagerly demanded. For example, the useful atmospheric transmission window 2.05~2.3 μm can serve the laser lidar better as a result of lower Rayleigh scattering losses than at <2.05 μm. Moreover, its pulsed version is also much more preferred by nonlinear frequency conversion using Zinc Germanium Phosphide (ZGP), which has a strong defect related absorption at wavelengths <2.1 μm .
Compared to TDF, Ho-doped fiber (HDF) has an extended efficient emission wavelength region, with a typical spectral range of 2.05~2.17 μm . Besides, its strong ground state absorption around 1.9 μm, well overlapped with the emission band of TDF laser, enables the fast gain-switching by efficient in-band pumping. In the past decade, many research works on gain-switched HDF laser were carried out [23–25]. In 2009, K. S. Wu et al. reported the first gain-switched HDF laser at 2.106 μm pumped by a gain-switched TDF laser at 1.909 μm . The maximum repetition rate and shortest pulse duration were 80 kHz and 70 ns, respectively (not simultaneously). After that, S. Hollitt et al. presented a linearly polarized gain-switched HDF laser at 2.104 μm using a gain-switched TDF laser pump source at 1.95 μm, yielding a maximum repetition rate of 600 kHz and the shortest pulse duration of 85 ns . Also they obtained a smaller pulse duration of 38 ns at the repetition rate of 300 kHz under unpolarized condition . Then, J. H. Geng et al. exploited an actively Q-switched TDF laser at 1.95 μm to pump a short cavity HDF laser and achieved single-frequency gain-switched pulses at 2.05 μm with a repetition rate of 20 kHz and the shortest pulse duration of 7~8 ns . Recently, we also demonstrated gain-switching of fluoride-glass based HDF laser at ~2.07 μm by using ~3 μm Q-switched [25,26] or gain-switched  pulses from the cascaded transition as the pump. Tens of kHz repetition rate and μs scale pulse duration were obtained. In these reports, however, the repetition rates were all clamped at hundreds of kHz. Even in the whole 2 μm spectral region, there were few reports on MHz or higher level gain-switching , limited by the pump source. But high repetition rate short pulses have important applications in such as free space optical communication , ranging , laser radar , countermeasure , etc. Although actively modulated continuous wave (CW) laser  or actively Q-switched laser  and electronically modulated laser diode (LD)  have showed their potential in pumping for high repetition rate gain-switching, all required additional controllers thus complicating the scheme. As a contrast, mode-locking can generate a MHz or even GHz level high repetition rate easily just relying on a compact all-fiber configuration. Over the past several decades, many types of mode-locked pulses have been discovered e.g., conventional soliton , stretched pulse , self-similar pulse , dissipative soliton (DS) , and noise-like (NL) pulse , but all suffered from multipulse operation caused by wave-breaking in nonlinear fibers with the increased pump power, inhibiting pulse energy scaling. Therefore, MOPA was usually used to scale the output with a complex multi-stage structure. However, high peak power also easily led to spectrum broadening under nonlinear effects hence influencing pulse temporal behavior during amplifying. And these features excluded them from pumping for gain-switching.
Recently, a new soliton formation named dissipative soliton resonance (DSR) was theoretically predicted using complex cubic-quantic Ginzburg-Landau equation by W. Chang et al. in 2008  and first experimentally observed in an Er-doped fiber laser at 1.5 μm by X. Wu et al. in 2009 . Compared to the mode-locked pulses mentioned above, this type of pulse has a rectangular temporal profile free from wave-breaking and besides owns some unique properties such as large pulse duration, low pulse peak power and high pulse energy. With the increased pump power, its pulse duration increases monotonously with an infinitely scaled pulse energy in principle. These features make this type of pulse a promising candidate as the pump for gain-switching at a high repetition rate. Recently, S. D. Chowdhury et al. adopted one-stage amplified DSR pulses generated from a mode-locked Er/Yb codoped fiber laser at 1.56 μm as the pump to demonstrate a gain-switched TDF fiber laser at 1.94 μm. The shortest pulse duration and maximum repetition rate were 256 ns and 750 kHz, respectively (not simultaneously) . Based on this scheme, they additionally introduced a 793 nm CW LD to build up a hybrid-pumped cavity to further scale the output power and pulse energy of the gain-switched pulses, giving a maximum repetition rate of 1.3 MHz . Very recently, another new pulse with a unique h-shaped temporal profile was discovered by J. Q. Zhao et al. reported in a 2 μm mode-locked Tm/Ho codoped fiber laser based on NOLM . Despite its h-shaped temporal profile with a higher peak amplitude at the pulse leading edge, it inherited some advantages of DSR pulse favorable by gain-switching e.g., large pump-dependent pulse duration, freedom from wave-breaking, high pulse energy, etc. This motivated us to explore its potential in high repetition rate gain-switching at >2.1 μm which is useful but unserved.
In this paper, we experimentally present a high repetition rate gain-switched HDF laser at 2.103 μm in an all-fiber configuration, exploiting a one-stage amplified h-shaped NPR mode-locked TDF laser at 1.985 μm as the pump, for the first time. The output characteristics of the h-shaped pulses from the seed and amplifier were investigated. Finally, stable gain-switched pulse with a repetition rate of 1.435 MHz and the shortest pulse duration of 10.8 ns were obtained, giving a large output power of 1.13 W and pulse energy of 0.79 μJ at a high slope efficiency of 75.4% with respect to its absorbed pump power. To the best of our knowledge, this is the first MHz level gain-switched fiber laser at >2 μm.
2. Experimental setup
The constructed gain-switched system as shown in Fig. 1 can be divided into three parts. The first part is the seed which consists of a typical NPR mode-locked ring cavity which is more compact than the NOLM or NALM scheme. The pump from a commercial 793 nm LD (BWT) was launched into the gain fiber through a (2 + 1) × 1 combiner (IFT). The gain fiber was a 2.0 m double-cladding TDF (Coractive, DCF-TM-10/128) with an octagonal shaped inner cladding with a diameter of 128 μm and a numerical aperture (NA) of 0.45, and a circular core with a diameter of 10 µm diameter and a NA of 0.22. The absorption coefficient of ~4 dB/m measured at 793 nm  gave a nominally 94% total absorption at this length. A piece of SMF-28e fiber was spliced to the TDF to provide enough nonlinear shift difference for stable mode-locking. Note that the used SMF-28e fiber length was a key factor for achieving h-shaped pulses, which were resulted from the incompletely clamped peak powers . To obtain strong reverse saturable absorption, large nonlinearity provided by SMF-28e fiber was required. Thus a piece of long SMF-28e fiber, selected as 135.0 m, was utilized. After that, a 50:50 coupler at ~2 μm was used to output half of the intra-cavity laser. A polarization independent optical isolator (Advanced Photonics) was used to maintain the unidirectional propagation. The polarizer in this scheme was an in-house 45° titled fiber grating (TFG) same as that in our previous report , on both sides of which two polarizer controllers (PCs) were placed to adjust intra-cavity polarization. The total cavity length was about 137.2 m including 2.0 m TDF, 135.0 m SMF28e fiber additionally added, 2 m SMF28e fiber (from the pigtails of coupler, isolator, 45° TFG, and signal input port of the combiner), and 0.2 m double-cladding silica fiber (from the pigtail of signal output port of the combiner). The anomalous dispersion values of the TDF and the SMF-28e fiber at ~1.9 μm were about −84 ps2/km and −80 ps2/km, respectively . Therefore, the cavity net dispersion was roughly calculated to be −11.128 ps2, indicating the laser was operated in a large anomalous dispersion region. Then the output from the seed was split by a 90:10 coupler, in which 10% portion is for seed monitoring through another 80:20 coupler, 90% portion was fed to the following amplifier. The amplifier was a typical one-stage MOPA based on TDF. The isolator used here, same as the one in the ring cavity, aimed at preventing the amplified laser from going back to the seed. Its pump-to-gain structure was almost similar to the seed with a longer TDF (4.9 m) used in order to provide larger gain. Finally, the amplified laser was fed into a linear cavity gain-switched HDF laser formed by an in-house high-reflectance (HR) FBG centered at 2103.0 nm  and the perpendicularly cleaved fiber end. The HDF was made in house using MCVD technology with a solution doping method, having a core diameter of 10 μm and core-cladding refractive index difference of 0.006. The Ho ions dopant concentration was 4.5 × 1017 cm−3. Therefore, the used ~7 m fiber provided an about 75% absorption measured at 1.985 μm. A bandpass filter (Throlabs, FB2250-500) which had a nominally transmittance of ~59% at 2.1 μm was placed after the HDF end to remove the residual ~1.985 μm laser. Note that the output powers of the HDF laser referred in this paper were all corrected according to the transmittance of the filter. Temporal and spectral profiles of the output pulses were monitored by a InGaAs photodetector (EOT ET-5000F, USA) followed by an 8 GHz digital oscilloscope and an optical spectrum analyzer (Yokogawa AQ6375) with a resolution of 0.05 nm, respectively. The radio frequency (RF) spectrum was measured using a RF spectrum analyzer (YIAI, China, AV4033A, 30Hz-18GHz).
3.1 H-shaped mode-locked TDF laser seed
In the seed, CW laser was first generated when the launched pump power was increased to 0.65 W. The high laser threshold was primarily resulted from the high proportion laser output and long SMF-28e fiber induced high cavity loss. Further increasing the launched pump power to 1.19 W, typical Q-switched mode-locking (QSML) regime was observed. Once the launched pump power was enhanced to 2.20 W, CW mode-locking (CWML) was obtained by carefully adjusting the PCs, which could keep stable until the available maximum launched pump power of 5.64 W. Figure 2(a) displays the evolution of pulse envelope at the launched pump power range of 2.20 W~5.64 W. It is seen that the pulse exhibits an obvious h-shaped envelope especially under high power pumping, similar to the recent report . With the increased pump power while keeping the PCs positions fixed, the pulse became broader and broader gradually and the peak amplitude was almost unchanged, where wave-breaking was not observed. Considering the unique temporal profile, 3 dB width (or FWHM) was not suited to characterizing its pulse duration like the typical Gauss-like pulse any more. Thus, its pulse duration was not accurately quantized in this paper. Figure 2(b) shows the corresponding optical spectrum evolution at the same pump range, which looks broad and smooth. With the increased pump power, only spectrum intensity became stronger with the almost unchanged 3 dB bandwidth and center wavelength of ~10 nm and ~1985 nm, respectively. These features reminded us of the rectangular pulses under DSR condition despite the temporal profile shape [41, 46–48], indicating some relationship between them. Figure 2(c) shows the h-shaped pulse trains captured at the launched pump power of 5.64 W at the scanning range of 1.4 s (top) and 1.4 μs (bottom). High temporal stability could be concluded from the low amplitude fluctuation. Besides, the time interval of adjacent pulses of ~696 ns matched well with the cavity round trip time, suggested the laser was operated at the fundamentally mode-locked regime. At the same pump power, the RF spectra were also measured at a broad and narrow scanning ranges of 500 MHz and 0.9 MHz as shown in Figs. 2(d) and 2(e), respectively. A periodical damped frequency modulation but much weaker than that of DSR pulses was observed owing to the existence of narrow leading edges. Although the h-shaped pulse was also resulted from a power clamping effect, different from DSR pulse which was completely clamped, it contained an unclamped portion, therefore contributing to a higher peak amplitude leading edge . In Fig. 2(e), a fundamental repetition rate of 1.435 MHz with a signal-to-noise ratio (SNR) of ~50 dB was obtained. Figure 2(f) shows the output power and pulse energy of the seed as a function of the launched pump power. It is observed that both increase linearly with the launched pump power. At the maximum launched pump power of 5.64 W, the maximum output power of 235.0 mW and pulse energy of 163.7 nJ were gained at a slope efficiency of 4.7% with respect to the launched pump power.
3.2 TDF based one-stage amplifier
When the h-shaped pulses were generated from the mode-locked TDF ring cavity, about 90% portion was fed to the amplifier to scale. Figure 3(a) and its inset show the optical spectra and temporal single pulse waveforms of the h-shaped pulses through the amplifier without and with amplifying, all captured at the output end of the amplifier. It is observed that the optical spectrum is almost unchanged after amplifying except the intensity promotion. One point worth noting is that the whole spectrum bandwidth without amplifying seems narrower than that measured directly from the seed. This was primarily resulted from the absorption of the TDF in the amplifier when its pump was switched off. Besides, the specifically mismatched butt-coupled alignment for optical spectrum measurement here was also a contributable factor in order to avoid too high amplified power induced OSA damage. In the temporal domain, the pulse leading edge was stronger after amplifying owing to experiencing higher gain in the amplifier compared to the pulse trailing edge. These results suggested that amplification didn’t distort the h-shaped pulses. Figure 3(b) shows the amplified output power and output power as a function of the launched pump power of the amplifier when the seed power was fixed at the maximum value. Similarly, both output power and pulse energy were amplified linearly. At the maximum launched pump power of 8.48 W, the maximum output power of 3.92 W and pulse energy of 2.71 μJ were obtained at a slope efficiency of 45.1% with respect to the launched pump power.
3.3 Gain-switched HDF laser
The amplified h-shaped pulses were fed to a linear cavity HDF laser to induce gain-switching. When the absorbed pump power was increased to 0.55 W, stable gain-switched pulses were obtained with a repetition rate of 1.435 MHz same as its pump, which was the current highest repetition rate from gain-switched fiber lasers at >2 μm. This regime could maintain until the available maximum absorbed pump power of 1.99 W as displayed in Fig. 4(a). It is observed that the gain-switched pulses with a typical Gauss-like profile shape are as stable as its pump and a tens of ns time delay exist between them within a period. Here the absorbed instead of launched pump power was adopted to characterize the pump amount mainly considering large amount of residual pump. Different from the previous reports on DSR pulses pumped high repetition rate gain-switched TDF laser, no gain-switched pulses with the 1/n (n≥2) pump repetition rate were observed in the initial pump region beyond the laser threshold in this case. This mainly attributed to the shorter pulse duration induced lower duty cycle, which led to the populations on the laser upper level provided by the former pump pulse were almost released before the arrival of the next pump pulse . Figure 4(b) plots the output power, pulse energy, and pulse duration of gain-switching as a function of the absorbed pump power.
It is seen that the pulse duration decreases fast first and then tends to saturate with the further increased absorbed pump power. This phenomenon was common and mainly resulted from increased populations accumulated on the laser upper level as the increased pump power. In fact, the time delay decreased as well in this process although not given in this figure. At the maximum absorbed pump power of 1.99 W, the shortest gain-switched pulse duration of 10.8 ns was obtained, comparable to the current record (i.e.,7~8 ns ) from gain-switched fiber lasers in this wavelength region. While the output power and pulse energy increased linearly with the absorbed pump power, giving a maximum output power of 1.13 W and pulse energy of 0.79 μJ. The high slope efficiency of 75.4% with respect to the absorbed pump power was primarily attributed to the efficient in-band pump scheme. Although it was still lower than the theoretical Stoke-limited efficiency (94.4%), further efficiency improvement was possible by optimizing the output cavity feedback while reducing the silica fiber background loss at this wavelength. Moreover, the effect of pump pulse duration on gain-switched pulse duration was also studied at a fixed low absorbed pump power of 1.06 W corresponding to 2.06 W output power of the amplifier. Here, the pump of the amplifier was adjusted to keep its output power at 2.06 W when varying the seed pulse duration by adjusting the pump of the seed. We found the gain-switched pulse duration was almost changed at the range of 13~14 ns when varying the launched pump power of the seed from 3.7 W from 5.64 W, indicating the pump pulse duration had an ignored effect on the gain-switched pulse duration. If further decreasing the launched pump power of the seed, it was difficult for the amplifier to reach the gain-switched threshold. As a contrast, the cavity or active fiber length has an obvious influence on the gain-switched pulse duration which had been experimentally investigated in Tm- and Yb-doped gain-switched fiber lasers [50,51], although not checked in our case. Figure 4(c) shows the measured output spectrum at the absorbed pump power of 1.909 W using a butt-coupled alignment after removing the bandpass filter. It is seen that there are a large number of optical spectrum components peaked at ~1985 nm resulted from the residual pump. In contrast to that measured after the amplifier, the optical spectrum here became asymmetric due to experiencing stronger absorption in HDF for the short wavelength components . On its right side, there were stronger spectrum components centered at 2103.05 nm corresponding to the gain-switched signal with a 3 dB bandwidth of 0.1 nm as shown in Fig. 4(d), which matched well with the used FBG.
In our seed, the NL pulses, usually appeared in high-power pumped mode-locked fiber lasers at large anomalous dispersion regimes [45,52], were also achieved by adjusting the PCs. In contrast to that shown in Fig. 3(a), however, the optical spectrum of the NL pulses was obviously broadened towards long wavelength direction after amplifying, attributable to the quite high peak power. In this case, only unstable and irregular gain-switched components were obtained due to the degraded pump pulses. Besides, we could observe the continuous temporal conversion process from NL to h-shaped pulse regime or the inverse process by slightly adjusting one PC under some specific conditions. Specifically, from NL to h-shaped pulse regime, the pulse peak was reduced while the pulse trailing edge was broadened with a relatively flat top. It implied the h-shaped pulse (incompletely clamped) regime maybe an intermediate state between NL pulse (unclamped) and DSR pulse (completely clamped) regimes.
In this work, our target was to explore the potential of h-shaped mode-locked pulses in high repetition rate gain-switching. Consequently, gain-switching with a maximum repetition rate of 1.435 MHz was achieved depended by the seed. Although higher repetition rate of the seed could be gotten if shortening the cavity by further reducing the additionally introduced SMF-28e fiber length to <135 m, it became difficult to generate h-shaped pulses except NL pulses. If further increasing the SMF-28e fiber length, the h-shaped pulses were always available until it was increased to 240 m, with which rectangular DSR pulses were achieved instead due to the stronger intra-cavity reverse saturable absorption with a longer SMF-28e fiber. Although this type of pulse also owned potential in gain-switching HDF laser considering its behavior in gain-switched TDF lasers [42,43], these efforts were beyond the study scope of this paper. In order to further increase the seed repetition rate, shortening the cavity by replacing SMF-28e fiber by high nonlinear fiber was a feasible way. Besides, harmonic operation like that under DSR condition [53,54] was an alternative method as well. Compared to rectangular DSR pulse, h-shape pulse has a critical possible merit in principle favorable by high repetition rate gain-switching, i.e., higher relaxation oscillation frequency as a result of higher pulse leading edge which was the upper limit of gain-switched repetition rate.
In summary, we have demonstrated a MHz level high repetition rate gain-switched HDF laser at 2.103 μm using h-shaped pulses at 1.985 μm as the pump, for the first time. The h-shaped pulses were generated from a NPR mode-locked TDF ring cavity at 1.985 μm which included an in-house 45° TFG polarizer and a piece of long SMF-28e fiber. It has a pump-dependent ns scale pulse duration at a fixed peak amplitude, giving a fundamental repetition rate of 1.435 MHz. Then the h-shaped pulses were scaled using a one-stage TDF based amplifier without serious distortions. The maximum output power of 3.92 W and pulse energy of 2.71 μJ were obtained at a slope efficiency of 45.1% with respect to its launched pump power. Finally, the amplified h-shaped pulses were fed to a linear cavity HDF laser formed by a HR FBG and the perpendicularly cleaved fiber end. The stable gain-switched pulses center at 2.103 μm were generated with a same repetition rate of 1.435 MHz as its pump and the shortest pulse duration of 10.8 ns, which was the current highest repetition rate from gain-switched fiber lasers in the 2 μm spectral region. The maximum output power of 1.13 W and pulse energy of 0.79 μJ were obtained at a large slope efficiency of 75.4% with respect to its absorbed pump power. The results indicated that h-shape pulse was an alternative promising candidate for high repetition rate gain-switching. Further repetition rate scaling of the seed could resort to a shorter cavity or harmonic operation which was also our following work.
National Nature Science Foundation of China (Grant No. 61722503, 61435003 and 61421002), Open fund of Science and Technology on Solid-State Laser Laboratory, Fundamental Research Funds for the Central Universities (Grant No. ZYGX2016J068), International Scientific Cooperation Project of Sichuan Province (Grant No. 2017HH0046).
We would like to thank Kundong Mo to discuss with us about the seed design.
1. S. W. Henderson, P. J. M. Suni, C. P. Hale, S. M. Hannon, J. R. Magee, D. L. Bruns, and E. H. Yuen, “Coherent laser radar at 2μm using solid-state lasers,” IEEE Trans. Geosci. Remote Sens. 31(1), 4–15 (1993). [CrossRef]
2. G. J. Koch, J. Y. Beyon, B. W. Barnes, M. Petro, J. Yu, F. Amzajerdian, M. J. Kavaya, and U. N. Singh, “High energy 2 μm Doppler lidar for wind measurements,” Opt. Eng. 46(11), 116201 (2007). [CrossRef]
4. N. M. Fried, “Thulium fiber laser lithotripsy: an in vitro analysis of stone fragmentation using a modulated 110-watt Thulium fiber laser at 1.94 μm,” Lasers Surg. Med. 37(1), 53–58 (2005). [CrossRef] [PubMed]
5. J. H. Geng and S. B. Jiang, “Fiber lasers: The 2 μm market heats up,” Opt. Photonics News 25(7), 34–41 (2014). [CrossRef]
6. D. Creeden, P. A. Ketteridge, P. A. Budni, S. D. Setzler, Y. E. Young, J. C. McCarthy, K. Zawilski, P. G. Schunemann, T. M. Pollak, E. P. Chicklis, and M. Jiang, “Mid-infrared ZnGeP2 parametric oscillator directly pumped by a pulsed 2 microm Tm-doped fiber laser,” Opt. Lett. 33(4), 315–317 (2008). [CrossRef] [PubMed]
7. K. Yin, B. Zhang, L. Y. Yang, and J. Hou, “30 W monolithic 2-3 μm supercontinuum laser,” Photon. Res. 6(1), 123–126 (2018). [CrossRef]
8. M. Jiang, H. F. Ma, Z. Y. Ren, X. M. Chen, J. Y. Long, M. Qi, D. Y. Shen, Y. S. Wang, and J. T. Bai, “A graphene Q-switched nanosecond Tm-doped fiber laser at 2 μm,” Laser Phys. Lett. 10(5), 055103 (2013). [CrossRef]
9. Z. Q. Luo, C. Liu, Y. Z. Huang, D. D. Wu, J. Y. Wu, H. Y. Xu, Z. P. Cai, Z. Q. Lin, L. P. Sun, and J. Weng, “Topological-insulator passively Q-switched double-clad fiber laser at 2 μm wavelength,” IEEE J. Sel. Top. Quantum Electron. 20(5), 1–8 (2014). [CrossRef]
10. Y. Z. Wang, J. F. Li, L. Han, R. G. Lu, Y. X. Hu, Z. Li, and Y. Liu, “Q-switched Tm3+-doped fiber laser with a micro-fiber based black phosphorus saturable absorber,” Laser Phys. 26(6), 065104 (2016). [CrossRef]
11. Z. Q. Luo, Y. Z. Huang, M. Zhong, Y. Y. Li, J. Y. Wu, B. Xu, H. Y. Xu, Z. P. Cai, J. Peng, and J. Weng, “1-, 1.5-, and 2-μm fiber lasers Q-switched by a broadband few-layer MoS2 saturable absorber,” J. Lightwave Technol. 32(24), 4077–4084 (2014). [CrossRef]
13. Y. L. Tang and J. Q. Xu, “Hybrid-pumped gain-switched narrow-band thulium fiber laser,” Appl. Phys. Express 5(7), 072702 (2012). [CrossRef]
14. B. C. Dickinson, S. D. Jackson, and T. A. King, “10 mJ total output from a gain-switched Tm-doped fibre laser,” Opt. Commun. 182(1–3), 199–203 (2000). [CrossRef]
15. S. D. Jackson and T. A. King, “Efficient gain-switched operation of a Tm-doped silica fiber laser,” IEEE J. Quantum Electron. 34(5), 779–789 (1998). [CrossRef]
19. Y. L. Tang, F. Li, and J. Q. Xu, “High peak-power gain-switched Tm-doped fiber laser,” IEEE Photonics Technol. Lett. 33(13), 893–895 (2011). [CrossRef]
20. J. Li, Z. Sun, H. Luo, Z. Yan, K. Zhou, Y. Liu, and L. Zhang, “Wide wavelength selectable all-fiber thulium doped fiber laser between 1925 nm and 2200 nm,” Opt. Express 22(5), 5387–5399 (2014). [CrossRef] [PubMed]
21. E. Optics, “Infrared non-linear crystals,” www.eksmaoptics.com/repository/catalogue/pdfai/NLOC/nonlinear%20crystals/IR.pdf.
22. A. Hemming, N. Simakov, J. Haub, and A. Carter, “A review of recent progress in holmium-doped silica fibre sources,” Opt. Fiber Technol. 20(6), 621–630 (2014). [CrossRef]
24. S. Hollitt, N. Simakov, A. Hemming, J. Haub, and A. Carter, “A linearly polarised, pulsed Ho-doped fiber laser,” Opt. Express 20(15), 16285–16290 (2012). [CrossRef]
27. J. Li, H. Luo, L. Wang, Y. Liu, Z. Yan, K. Zhou, L. Zhang, and S. K. Turistsyn, “Mid-infrared passively switched pulsed dual wavelength Ho3+-doped fluoride fiber laser at 3 μm and 2 μm,” Sci. Rep. 5(1), 10770 (2015). [CrossRef] [PubMed]
28. C. Larat, M. Schwarz, J. Pocholle, G. Feugnet, and M. R. Papuchon, “High-repetition rate, short-pulse, diode-pumped solid state laser for space communications,” Proc. SPIE 2210, 565–571 (1994). [CrossRef]
29. P. Rieger and A. Ullrich, “Resolving range ambiguities in high-repetition rate airborne lidar applications,” Proc. SPIE 8186, 81860A (2011).
30. P. Rieger and A. Ullrich, “Resolving range ambiguities in high-repetition rate airborne light detection and ranging applications,” J. Appl. Remote Sens. 6(1), 063552 (2012). [CrossRef]
31. F. Huang, Y. Wang, J. Wang, and Y. Niu, “Study on application of high-repetition-rate solid state lasers in photoelectric countermeasure,” Infrared Laser Eng. 32(5), 465–467 (2003).
33. J. N. Zhang, Y. Wang, and D. Y. Shen, “High repetition rate gain-switched thulium fiber laser with an acousto-optic modulator,” IEEE Photonics Technol. Lett. 25(19), 1943–1946 (2013). [CrossRef]
34. H. M. Zhao, Q. H. Lou, J. Zhou, F. P. Zhang, J. X. Dong, Y. R. Wei, G. H. Wu, Z. J. Yuan, Z. J. Fang, and Z. J. Wang, “High-repetition-rate MHz acousto-optic Q-switched fiber laser,” IEEE Photonics Technol. Lett. 20(12), 1009–1011 (2008). [CrossRef]
35. L. E. Nelson, D. J. Jones, K. Tamura, H. A. Haus, and E. P. Ippen, “Ultrashort-pulse fiber ring lasers,” Appl. Phys. B 65(2), 277–294 (1997). [CrossRef]
36. K. Tamura, E. P. Ippen, and H. A. Haus, “Pulse dynamics in stretched-pulse fiber lasers,” Appl. Phys. Lett. 67(2), 158–160 (1995). [CrossRef]
38. X. Liu, “Numerical and experimental investigation of dissipative solitons in passively mode-locked fiber lasers with large net-normal-dispersion and high nonlinearity,” Opt. Express 17(25), 22401–22416 (2009). [CrossRef] [PubMed]
39. Y. Q. Huang, Y. L. Qi, Z. C. Luo, A. P. Luo, and W. C. Xu, “Versatile patterns of multiple rectangular noise-like pulses in a fiber laser,” Opt. Express 24(7), 7356–7363 (2016). [CrossRef] [PubMed]
40. W. Chang, A. Ankiewicz, J. M. Soto-Crespo, and N. Akhmediev, “Dissipative soliton resonances,” Phys. Rev. A 78(2), 023830 (2008). [CrossRef]
42. S. D. Chowdhury, A. Pal, D. Pal, S. Chatterjee, M. C. Paul, R. Sen, and M. Pal, “High repetition rate gain-switched 1.94 μm fiber laser pumped by 1.56 μm dissipative soliton resonance fiber laser,” Opt. Lett. 42(13), 2471–2474 (2017). [CrossRef] [PubMed]
44. J. Zhao, L. Li, L. Zhao, D. Tang, and D. Shen, “Cavity-birefringence-dependent h-shaped pulse generation in a thulium-holmium-doped fiber laser,” Opt. Lett. 43(2), 247–250 (2018). [CrossRef] [PubMed]
45. J. Li, Z. Yan, Z. Sun, H. Luo, Y. He, Z. Li, Y. Liu, and L. Zhang, “Thulium-doped all-fiber mode-locked laser based on NPR and 45°-tilted fiber grating,” Opt. Express 22(25), 31020–31028 (2014). [CrossRef] [PubMed]
46. W. Chang, J. M. Soto-Crespo, A. Ankiewicz, and N. Akhmediev, “Dissipative soliton resonances in the anomalous dispersion regime,” Phys. Rev. A 79(3), 033840 (2009). [CrossRef]
47. X. Liu, “Pulse evolution without wave breaking in a strongly dissipative-dispersive laser system,” Phys. Rev. A 81(5), 053819 (2010). [CrossRef]
49. H. Luo, J. Li, Y. Hai, X. Lai, and Y. Liu, “State-switchable and wavelength-tunable gain-switched mid-infrared fiber laser in the wavelength region around 2.94 μm,” Opt. Express 26(1), 63–79 (2018). [CrossRef] [PubMed]
52. J. Li, Z. Zhang, Z. Sun, H. Luo, Y. Liu, Z. Yan, C. Mou, L. Zhang, and S. K. Turitsyn, “All-fiber passively mode-locked Tm-doped NOLM-based oscillator operating at 2-μm in both soliton and noisy-pulse regimes,” Opt. Express 22(7), 7875–7882 (2014). [CrossRef] [PubMed]
53. Y. Lyu, X. Zou, H. Shi, C. Liu, C. Wei, J. Li, H. Li, and Y. Liu, “Multipulse dynamics under dissipative soliton resonance conditions,” Opt. Express 25(12), 13286–13295 (2017). [CrossRef] [PubMed]
54. G. Semaan1, A. Niang, M. Salhi, and F. Sanchez, “Harmonic dissipative soliton resonance passively mode- locked fiber laser,” in CLEO: QELS_Fundamental Science (OSA) (2018), paper FTh1M. 5.