Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

1.7-μm Tm-doped fiber laser intracavity-pumped by an erbium/ytterbium-codoped fiber laser

Open Access Open Access

Abstract

In this paper, we demonstrate an efficient 1.7-μm Tm-doped fiber laser whose cavity was embedded in a 1560 nm erbium/ytterbium-codoped fiber laser cavity, which enabled bidirectional pumping and made full use of the circulating pump in the parent laser cavity. A rate equation model was developed to optimize the fiber length and output coupling for a desired output power. In the experiment, a maximum output power at 1720 nm of 1.13 W was obtained under 10 W of 976 nm diode pump power, which correlated well with our modeling. The slope efficiency from the multimode 976 nm diode pump to 1720 nm output was 13.5%, while the slope efficiency in terms of launched 1560 nm pump power reached 62.5%. By using a short Tm-doped fiber to minimize signal reabsorption, a high signal-to-noise ratio over 65 dB was achieved. The prospect for further power scaling was also discussed based on our developed model.

© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement

1 Introduction

In recent years, laser sources operating in the wavelength region of 1.7 μm have received intense attention due to their extensive use in applications such as optical coherence tomography, acne vulgaris treatment and polymer processing [16]. It is also predicted that the efficiency of Dy-doped fiber lasers operating in the mid-infrared region can be improved significantly when 1.7 μm lasers are employed as the pump source [7,8]. Tm-doped fibers have a broad emission band extending from 1650 nm to 2100 nm, and are regarded as a promising route to achieving lasing in this wavelength region. 1.7 μm lasing in these fibers can be directly generated from the 3F43H6 transition in Tm. However, the combination of large reabsorption loss and gain competition makes it challenging to achieve efficient 1.7 μm laser emission compared with that at 1.8-2 μm. Strong quasi-three-level behavior at 1.65-1.8 μm requires high-brightness pumping to achieve a sufficient population inversion [9]. Furthermore, severe amplified spontaneous emission (ASE) and/or parasitic lasing at longer wavelengths have also decreased the signal-to-noise ratio (SNR) achievable from these sources.

At present, research on 1.7 μm Tm-doped fiber lasers predominantly focuses on improvement of output power and efficiency. These aspects of performance can be enhanced through the use of 1.55-1.60 μm Er-doped or Er/Yb-codoped fiber lasers as the pump source, as this benefits low quantum defect [914]. The fundamental-transverse-mode pump light in the fiber core (facilitated by fiber-to-fiber coupling) also enhances the pump absorption. Based on this pump scheme, the output power and efficiency of 1.7 μm Tm-doped fiber lasers have been improved dramatically. In 2015, researchers at the University of Southampton demonstrated the first ten-watt-level 1.7 μm Tm-doped fiber laser with output power above 12 W and a slope efficiency of 63% [9]. In 2019, by employing a self-built high-power 1580 nm pump system, the output power at 1726 nm was further improved to 47 W. The optical efficiency reached up to 80%, which was near the theoretical maximum Stokes efficiency (∼91.5%) [12]. It should be noted that both of these approaches utilized a homemade Tm-doped fiber. Other demonstrations of 1.7 μm fiber lasers in the literature are instead based on commercial Tm-doped fiber, and exhibit slope efficiencies typically no more than 50%. Recently, we demonstrated an efficient 1720nm Tm-doped fiber ring laser with optimized output coupling. A slope efficiency with respect to the launched 1570 nm pump power of 46.5% was achieved, this in spite of the large insertion loss of the circulator. This is the highest reported efficiency among 1.7-μm fiber ring lasers based on commercial Tm-doped fiber till now [14].

The multimode laser diode, which exhibits advantages in high pump power, compactness and low cost, is generally used as the pump source for high-power fiber lasers. However, in the 1.7 μm wavelength region, large reabsorption loss makes it extremely difficult to achieve a sufficient population inversion and net gain in Tm-doped fibers using conventional cladding-pumping scheme with 790 nm laser diode [12]. Here, we present an alternative approach using a dual-cavity structure which includes an external 1560 nm Er/Yb-codoped fiber cavity, cladding-pumped by a multimode 976 nm laser diode and an internal 1720nm Tm-doped fiber cavity embedded within the Er/Yb-codoped fiber cavity. The 1560 nm laser cavity serves as the in-band pump source for the internal 1.7 μm fiber laser. The high-power intracavity 1560 nm pump laser makes it feasible to use a short Tm-doped fiber which limits reabsorption loss and yet has sufficient pump absorption. Furthermore, a bidirectional pump scheme also facilitates efficient and low-threshold operation. A rate equation model was used to analyze the energy transfer process between the Er/Yb-codoped and Tm-doped fibers. The output characteristics of this 1.7 μm fiber laser were experimentally studied with different Tm-doped fiber lengths, and the results were in good agreement with calculations. By using a 5 m Er/Yb-codoped fiber and a 1.4 m Tm-doped fiber, a maximum 1720 nm output power of 1.13 W and a high SNR exceeding 65 dB were obtained under 10 W of 976 nm pump power, corresponding to a diode to 1.7 μm slope efficiency of 13.5%. The potential power scaling of the design was also theoretically investigated and the calculated results showed that the efficiency of the system can be increased to over 20% via minimization of the fusion splicing losses.

2 Experimental setup

The experimental setup of the 1.7 μm Tm-doped fiber laser is shown in Fig. 1. The external cavity is formed by two 1560 nm high reflectivity (HR, R>99.5%) fiber Bragg gratings (FBG1 and FBG4) with a 3 dB bandwidth of ∼0.5 nm. The pump light is provided by a multimode 976 nm laser diode (BWT, DS3-51412-0714) and is launched through a combiner, with a signal fiber, GDF 6/125 from Nufern Inc. A length of double-cladding Er/Yb-codoped fiber (Nufern, PM-EYDF-6/125-HE) with a cladding absorption coefficient at 976 nm of 3.5 dB/m is used to generate the 1560 nm laser field which is confined in the cavity by FBG1 and FBG4. Due to the low power extraction offered by FBG1 and FBG4, the intracavity field intensity at 1560 nm is very high. The residual 976 nm pump power is removed by a homemade cladding mode stripper (CMS) where the fiber cladding is recoated with a high refractive index polymer. The 1.7 μm laser cavity, comprises a 1720nm HR FBG (FBG2, R>99.5%, Δλ∼0.8 nm), an output FBG (FBG3, Δλ∼0.2 nm) and a length of Tm-doped fiber (Nufern, SM-TSF-9/125) which is all positioned after the Er/Yb-codoped fiber. The doping concentration of this Tm-doped fiber is around 0.3 wt.%, which limits reabsorption loss. Since the Tm-doped fiber is embedded in the linear cavity of the 1560 nm laser, it is bidirectionally pumped by the circulating 1560 nm laser. This scheme enhances the pump absorption so that the shorter Tm-doped fiber with lower reabsorption loss at 1.7 μm can be used. A 1560/1720nm wavelength division multiplexer (WDM) is spliced after FBG4 to filter the small amount of undesired 1560 nm laser which is present in the laser output. The end face of output fiber is cleaved at an angle of 8° to minimize residual optical feedback.

 figure: Fig. 1.

Fig. 1. Experimental setup of the 1720nm Tm-doped fiber laser. All the FBGs are written in standard SMF-28e fiber.

Download Full Size | PDF

3 Rate equation model

Figure 2 shows the simplified schematic of the 1720 nm Tm-doped fiber laser. Pumped at 976 nm (Pp1), the 1560 nm laser field (Ps1) was generated from the Er/Yb-codoped fiber and launched into the Tm-doped fiber to generate 1720 nm laser (Ps2) emission. The residual forward 1560 nm laser could be reflected and recycled in the Tm-doped fiber to increase the pumping efficiency.

 figure: Fig. 2.

Fig. 2. The theoretical schematic of the 1720nm Tm-doped fiber laser. PR FBG: partial-reflection fiber Bragg grating. The blue and orange arrows denote the propagation of 1560 nm and 1720nm laser fields. L1 and L2 are the lengths of the Er/Yb-codoped fiber and Tm-doped fibers, respectively.

Download Full Size | PDF

Owing to the higher doping concentration and larger absorption cross section (at 976 nm) of Yb ions, the Er/Yb-codoped fiber exhibits a much higher pump absorption compared with singly Er-doped fiber. In Er/Yb-codoped fibers, pump energy is mainly absorbed by Yb ions and transferred to the neighboring ground-state Er ions through cross relaxation. According to Refs [15,16], the steady-state pump and signal power evolutions are determined by:

$$\begin{aligned} \frac{{\textrm{d}P_{\textrm{p}1}^ \pm (z)}}{{\textrm{d}z}} ={\pm} P_{\textrm{p}1}^ \pm (z)&\{{{\varGamma _{\textrm{p}1}}[{{\sigma_{{{\textrm{Yb}}\_{\textrm e}}}}({\lambda_{\textrm{p}1}}){N_{{{\textrm{Yb}}\_2}}}(z) - {\sigma_{{{\textrm{Yb}}\_{\textrm a}}}}({\lambda_{\textrm{p}1}}){N_{{{\textrm{Yb}}\_1}}}(z)} } \\ &{ { - {\sigma_{{{\textrm{Er}}\_{\textrm a}}}}({\lambda_{\textrm{p}1}}){N_{{{\textrm{Er}}\_1}}}(z)} ]- {\alpha_{\textrm{p}1}}} \}\end{aligned}$$
$$\frac{{\textrm{d}P_{\textrm{s1}}^ \pm (z)}}{{\textrm{d}z}} ={\pm} P_{\textrm{s1}}^ \pm (z)\{{{\varGamma _{\textrm{s1}}}[{{\sigma_{{{\textrm{Er}}\_{\textrm e}}}}({\lambda_{\textrm{s1}}}){N_{{{\textrm {Er}}\_2}}}(z) - {\sigma_{{{\textrm{Er}}\_{\textrm a}}}}({\lambda_{\textrm{s1}}}){N_{{{\textrm{Er}}\_1}}}(z)} ]- {\alpha_{\textrm{s1}}}} $$
The in-band pump scheme in Tm-doped fiber enables a low quantum defect, which thus improves the conversion efficiency. When pumped at 1560 nm, the Tm-doped fiber laser can be simplified to a two-level system and the power distributions of the pump and signal can be determined by [17,18]:
$$\frac{{\textrm{d}P_{\textrm{p2}}^ \pm (z)}}{{\textrm{d}z}} ={\pm} P_{\textrm{p2}}^ \pm (z)\{{{\varGamma _{\textrm{p2}}}[{{\sigma_{{{\textrm{Tm}}\_{\textrm e}}}}({\lambda_{\textrm{p2}}}){N_{{{\textrm {Tm}}\_2}}}(z) - {\sigma_{{{\textrm {Tm}}\_{\textrm a}}}}({\lambda_{\textrm{p2}}}){N_{{{\textrm {Tm}}\_1}}}(z)} ]- {\alpha_{\textrm{p2}}}} \}$$
$$\frac{{\textrm{d}P_{\textrm{s2}}^ \pm (z)}}{{\textrm{d}z}} ={\pm} P_{\textrm{s2}}^ \pm (z)\{{{\varGamma _{\textrm{s2}}}[{{\sigma_{{{\textrm{Tm}}\_{\textrm e}}}}({\lambda_{\textrm{s2}}}){N_{{{\textrm {Tm}}\_2}}}(z) - {\sigma_{{{\textrm{Tm}}\_{\textrm a}}}}({\lambda_{\textrm{s2}}}){N_{{{\textrm{Tm}}\_1}}}(z)} ]- {\alpha_{\textrm{s2}}}} \}$$
In Eqs. (1)–(4), Pp and Ps represent the pump and signal power, respectively; Γp and Γs are the overlapping factors for the pump light and the signal light, respectively; σYb_e, σEr_e and σTm_e are the emission cross sections of Yb, Er and Tm ions, while σYb_a, σEr_a and σTm_a are the corresponding absorption cross sections; NYb_1 and NYb_2 are the population densities of 2F7/2 and 2F5/2 levels in Yb, respectively; NEr_1 and NEr_2 are the population densities of 4I15/2 and 4I13/2 levels in Er, respectively; NTm_1 and NTm_2 are the population densities of 3H6 and 3F4 levels in Tm, respectively; αp and αs are the intrinsic losses at the pump wavelength and signal wavelength for the Er/Yb-codoped fiber and Tm-doped fiber, respectively.

Based on the laser setup, the boundary conditions of the 976 nm pump, 1560 nm laser and 1720nm laser are given as follows:

$$\begin{aligned}P_{\textrm{p1}}^ + ({0}) &= P, P_{\textrm{p1}}^ - ({L_1}) = 0 \\ P_{\textrm{s1}}^ + (0) &= {R_1}({1 - {\delta_1}} )({1 - {\delta_2}} )P_{\textrm{s1}}^ - (0 )\\ P_{\textrm{p2}}^ + ({L_1}) &= (1 - {\delta _2})P_{\textrm{s1}}^ + ({L_1}) \\ P_{\textrm{s1}}^ - ({L_1}) &= (1 - {\delta _1})P_{\textrm{p2}}^ - ({L_1})\\ P_{{\rm s2}}^- (L_1 + L_2) &= R_3P_{{\rm s2}}^ + (L_1 + L_2) \\ P_{\textrm{s2}}^ - ({L_1} + {L_2}) &= {R_3}P_{\textrm{s2}}^ + ({L_1} + {L_2})\\ P_{\textrm{p2}}^ - ({L_1} + {L_2}) &= {R_4}P_{\textrm{p2}}^\textrm{ + }({L_1} + {L_2})\end{aligned}$$
Where P denotes the 976 nm pump power launched into the Er/Yb-codoped fiber. L1 and L2 are the lengths of the Er/Yb-codoped fiber and Tm-doped fiber. Ri (i=1,2,3,4) are the reflectivities of four FBGs. δj (j=1,2) represents the splicing loss between the Er/Yb-codoped fiber and SMF-28e fiber. The splicing loss was experimentally measured using a homemade 1570 nm fiber laser. Considering that the Er/Yb-codoped fiber exhibits strong absorption at 1570 nm, we used its passive matched fiber (GDF-6/125) instead in the measurement. The GDF-6/125 fiber with a CMS was spliced to SMF-28e fiber to measure the loss δ1. Then, another piece of SMF-28e fiber was spliced to the GDF-6/125 fiber to measure the loss δ2. Based on this scheme, the transmission losses from SMF-28e fiber to GDF-6/125 fiber and from GDF-6/125 fiber to SMF-28e fiber were measured as 0.5 dB and 1 dB, thus the values δ1 and δ2 were 0.1 and 0.2, respectively. Such large splicing losses can be attributed to the mismatch of the mode field diameter (MFD) and numerical aperture (NA). The MFDs (at 1560 nm) of SMF-28e and Er/Yb-codoped fiber are 11.1 μm and 8.7 μm [19]; while the NAs of these two fibers are 0.14 and 0.18, respectively. For efficient 1.7 μm laser output, the Er/Yb-codoped fiber length, Tm-doped fiber length and the reflectivity of output FBG are calculated. The simulations are carried out by the finite difference method and the specific parameters (and their values) used in the calculation are summarized in Table 1.

Tables Icon

Table 1. Parameters used in the calculation [15,2023]

The high loss of the splice between the Er/Yb-codoped fiber and SMF-28e fiber meant that it was very susceptible to damage by the circulating laser field. The maximum launched 976 nm pump power in the experiment was therefore limited to 10 W to protect the laser system, and simulations were carried out at this pump power. The optimal Er/Yb-codoped fiber length was calculated to be 5 m; this ensured sufficient pump absorption as well as low signal reabsorption loss. In this case, the influence of Tm-doped fiber length and reflectivity of FBG3 on 1.7 μm output power were calculated. The results of this modeling are shown in Fig. 3(a). A maximum output power of 1.26 W can be achieved when using a 1.6 m long Tm-doped fiber and an output FBG with 45% reflectivity. It can be seen that there is a large region within which the output power is above 1.2 W, indicating that the output power of current laser is somewhat flexible with regard to the choice of Tm-doped fiber length and output coupling. This is because the pump absorption in the Tm-doped fiber effectively acts as negative feedback to the 1560 nm parent laser. The optimal Tm-doped fiber length is therefore in the range of 1.4 and 2 m, with corresponding reflectivity of the output FBG in the range 30%-50%. Fig. 3(a) also indicates that a laser with a lower output coupling allows the use of a shorter Tm-doped fiber to produce similar output power levels. Fig. 3(b) plots the 1.7 μm output power as a function of the length of Tm-doped fiber with the reflectivity of FBG3 fixed at R=45%. When the Tm-doped fiber length is less than 0.3 m, the laser cannot oscillate due to the limited gain. Thereafter the 1.7 μm laser power grows with the increasing Tm-doped fiber length, till the reabsorption loss causes a decrease in the output power when the fiber length exceeds 1.6 m.

 figure: Fig. 3.

Fig. 3. (a) Theoretical output powers as functions of Tm-doped fiber length and the reflectivity of FBG3; (b) theoretical output powers versus Tm-doped fiber length with a R=45% FBG3. The numbers besides the contour curves in (a) are the output powers.

Download Full Size | PDF

4 Experimental results and discussion

Experiments were performed with the configuration shown in Fig. 1. Based on the simulation results above, an FBG with a reflectivity of 45% was chosen as the output coupler of the 1.7 μm laser cavity. The output powers as functions of the launched 976 nm diode pump power with different Tm-doped fiber lengths are shown in Fig. 4(a). A linear increase of output power with the launched pump power was observed for all the cases. At 10 W of 976 nm pump power, a maximum output power of 1.13 W was obtained using a 1.4 m long Tm-doped fiber, while the optical efficiency and slope efficiency were 11.3% and 13.5%, respectively. As expected from our modeling, the output power dropped slightly with longer Tm-doped fibers owing to larger cavity loss induced by signal reabsorption. The measured and calculated slope efficiency and threshold pump power (in terms of 976 nm diode power) of the 1.7 μm laser with different Tm-doped fiber lengths are presented in Fig. 4(b). With the Tm-doped fiber length increasing from 1 to 2 m, the measured laser threshold increased from 1.8 W to 3 W and the measured slope efficiency increased from 11% to 15.3%, which were both in good agreement with calculations.

 figure: Fig. 4.

Fig. 4. (a) Power transfer of the 1720nm fiber laser with different Tm-doped fiber lengths; (b) Measured and calculated results of laser threshold and slope efficiency versus Tm-doped fiber length.

Download Full Size | PDF

To investigate the power transfer characteristic from 1560 nm in the parent cavity to the 1720nm output, a 99:1 (at 1560 nm) 2×2 coupler was inserted between the CMS and FBG2 to monitor the forward and backward 1560 nm powers. As shown in the Fig. 5 (a), the maximum 1560 nm power launched into the 1.7 μm cavity was inferred to be 2.1 W, while the corresponding backward “residual” 1560 nm pump power was around 0.2 W, revealing a total pump absorption (at 1560 nm) over 90%. Fig. 5(b) shows the 1720nm output power versus the launched and absorbed 1560 nm laser power. A relatively low threshold (versus the launched 1560 nm power) of around 0.3 W was achieved. Additionally, the slope efficiency was also improved significantly. The slope efficiency with respect to launched and absorbed 1560 nm power were 62.5% and 68%, respectively. To the best of our knowledge, our setup exhibits the highest efficiency (from 1.5 μm to 1.7 μm) among reported 1.7 μm fiber lasers based on commercial Tm-doped fibers. It should be noted that the slope efficiency (in terms of launched 1560 nm pump power) was lower than the Stokes efficiency limit (∼90%). The difference was attributed to the launched 1560 nm pump not being completely absorbed by the Tm-doped fiber. Furthermore, the intrinsic reabsorption loss was limited, but not eliminated in spite of a short Tm-doped fiber.

 figure: Fig. 5.

Fig. 5. (a) Forward and backward 1560 nm laser power (inferred by monitored powers of a 2×2 coupler) versus the launched 976 nm pump power; (b) 1720nm output power versus launched and absorbed 1560 nm power.

Download Full Size | PDF

An optical spectrum analyzer (OSA, YOKOGAWA, AQ6375) with a resolution of 0.05 nm was used to record the laser spectrum. Fig. 6 shows the laser spectrum at an output power of 1.1 W. The laser wavelength is 1720.04 nm. Due to the use of FBGs and a short Tm-doped fiber with low doping concentration, the ASE at 1.8-2 μm was suppressed effectively. The results show a high SNR of greater than 65 dB, which exceeds the SNR of previously reported watt-level 1.7 μm Tm-doped fiber lasers (which were usually no more than 55 dB). Due to the use of the 1560 nm HR FBG and WDM, there was very little 1560 nm pump left in the output. The inset in Fig. 6 shows the zoomed spectrum with linear scale demonstrating a 3 dB linewidth of around 75 pm (7.6 GHz at 1720nm). The output power fluctuation recorded by the power meter within 30 minutes was less than 1% (RMS) at an output power of 1.1 W.

 figure: Fig. 6.

Fig. 6. Measured laser spectrum at ∼1.1 W output power. Inset: zoomed laser spectrum with a linear scale.

Download Full Size | PDF

Note that the cavity loss in the current laser configuration mainly comes from the splicing losses between Er/Yb-codoped fiber or GDF 6/125 fiber and standard SMF-28e fiber. If the splicing losses could be reduced, better laser performance is expected. To investigate the extent to which the performance could be improved, we modeled the 1.7 μm output power as a function of the Tm-doped fiber length and reflectivity of the output FBG, with the assumption of zero splicing/coupling losses within the laser; the results are plotted in Fig. 7(a). Compared with the current experiment, the optimal Tm-doped fiber length is further shortened to around 1 m and the maximum output power is increased to 2.01 W with a 50% output coupling. The theoretical 1720 nm output power as a function of 976 nm pump power with 1 m Tm-doped fiber and 50% output coupling is shown in Fig. 7(b). A higher slope efficiency of 22.5% and a lower threshold of 1 W could be achieved. The inset in Fig. 7(b) shows the power transfer from the 1560 nm field in the parent cavity to the 1720 nm output. In this design, the maximum launched and absorbed 1560 nm powers are increased to 3.8 W and 2.8 W respectively, compared with those of 2.1 W and 1.9 W in the current experiment. The slope efficiency with respect to the launched and absorbed 1560 nm powers are 58% and 79%, respectively. So, by improving the matching between fibers and reducing the coupling losses within the system, substantial improvements to the output power, laser efficiency and laser threshold can be achieved.

 figure: Fig. 7.

Fig. 7. (a) Theoretical 1720nm output powers as a function of Tm-doped fiber length and the reflectivity of output FBG without splicing loss; (b) theoretical power transfer from launched 976 nm pump to 1720nm output power. Inset: Theoretical power transfer from 1560 nm power to 1720nm output power. The numbers beside the contour curves in (a) are the output powers.

Download Full Size | PDF

The risk of splice joint damage limited the pump power to 10 W in the current experiment. Using our numerical model, we determine that 15 W output power at 1720nm can be obtained under 100 W launched diode pump, assuming the splicing losses can be eliminated. Considering the power handling of other devices including the FBG, combiner and WDM, it is possible for the intracavity-pumped Tm-doped laser to be further power scaled to tens of watts, with efficiency comparable with that of the record 1.7 μm power demonstrated with homemade Tm-doped fiber [9,12]. However, it should be emphasized that the main advantage of our setup is that it allows good conversion efficiency at low-to-moderate power levels since the circulating pump power is efficiently utilized. In other words, only several watts of primary diode pump are needed to produce efficient watt-level 1.7 μm output, which is impossible with traditional schemes. This is also the advantage compared to the cascaded Raman random fiber lasers in 1.7 μm region which generally have extremely high threshold (tens of watts) and broad laser linewidth (several nanometers) [24,25]. Moreover, our intracavity scheme also exhibits a high degree of flexibility in that it is quite tolerant of a range of fiber lengths and FBG reflectivity.

5 Conclusion

In conclusion, a 1.7 μm Tm-doped fiber laser, the cavity of which was embedded in an Er/Yb-codoped fiber laser, was demonstrated in this paper. Two HR FBGs were employed to generate a high-power intracavity 1560nm field which was then used to efficiently pump a short length of Tm-doped fiber. The dependences of laser power on Tm-doped fiber lengths and output couplings were theoretically and experimentally investigated in detail. The monolithic device, pumped directly by the 976nm diode yielded 1.13 W output at 1720nm under a multimode pump power of 10 W. Since the short Tm-doped fiber minimized the reabsorption loss and long-wavelength ASE, an excellent SNR over 65dB was achieved. Further optimization of the laser design, notably a reduction in splicing loss offers the prospect of greater power scaling and optical efficiency greater than 20%. This work highlights a new method of developing compact and efficient fiber laser sources with high SNR in the 1.7 μm wavelength region.

Funding

National Natural Science Foundation of China (62075159, 61975146); Major Scientific and Technological Innovation Projects of Key R&D Plans in Shandong Province (2019JZZY020206).

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. I. Mingareev, F. Weirauch, A. Olowinsky, L. Shah, P. Kadwani, and M. Richardson, “Welding of polymers using a 2 μm thulium fiber laser,” Opt. Laser Technol. 44(7), 2095–2099 (2012). [CrossRef]  

2. V. V. Alexander, K. Ke, Z. Xu, M. N. Islam, M. J. Freeman, B. Pitt, M. J. Welsh, and J. S. Orringer, “Photothermolysis of sebaceous glands in human skin ex vivo with a 1,708 nm Raman fiber laser and contact cooling,” Laser Surg. Med. 43(6), 470–480 (2011). [CrossRef]  

3. M. Wu, K. Jansen, A. F. W. van der Steen, and G. van Soest, “Specific imaging of atherosclerotic plaque lipids with two-wavelength intravascular photoacoustics, Biomed,” Opt. Express 6(9), 3276–3286 (2015). [CrossRef]  

4. U. Sharma, E. W. Chang, and S. H. Yun, “Long-wavelength optical coherence tomography at 1.7 μm for enhanced imaging depth,” Opt. Express 16(24), 19712–19723 (2008). [CrossRef]  

5. C. Anselmo, J. Y. Welschinger, J. P. Cariou, A. Miffre, and P. Rairoux, “Gas concentration measurement by optical similitude absorption spectroscopy: methodology and experimental demonstration,” Opt. Express 24(12), 12588–12599 (2016). [CrossRef]  

6. S. P. Chong, C. W. Merkle, D. F. Cooke, T. Zhang, H. Radhakrishnan, L. Krubitzer, and V. J. Srinivasan, “Noninvasive, in vivo imaging of subcortical mouse brain regions with 1.7 μm optical coherence tomography,” Opt. Lett. 40(21), 4911–4914 (2015). [CrossRef]  

7. X. Xiao, Y. Xu, H. Guo, P. Wang, X. Cui, M. Lu, Y. Wang, and B. Peng, “Theoretical Modeling of 4.3 μm Mid-Infrared Lasing in Dy3+-Doped Chalcogenide Fiber Lasers,” IEEE Photonics J. 10(2), 1–11 (2018). [CrossRef]  

8. R. S. Quimby, L. B. Shaw, J. S. Sanghera, and I. D. Aggarwal, “Modeling of cascade lasing in Dy: chalcogenide glass fiber laser with efficient output at 4.5 μm,” IEEE Photonics Technol. Lett. 20(2), 123–125 (2008). [CrossRef]  

9. J. M. O. Daniel, N. Simakov, M. Tokurakawa, M. Ibsen, and W. A. Clarkson, “Ultra-short wavelength operation of a thulium fibre laser in the 1660-1750nm wavelength band,” Opt. Express 23(14), 18269–18276 (2015). [CrossRef]  

10. J. Zhang, Q. Sheng, S. Sun, C. Shi, S. Fu, W. Shi, and J. Yao, “1.7-μm thulium fiber laser with all-fiber ring cavity,” Opt. Commun. 457, 124627 (2020). [CrossRef]  

11. X. Xiao, H. Guo, M. Lu, Z. Yan, H. Wang, Y. Wang, Y. Xu, C. Gao, X. Cui, Q. Guo, and B. Peng, “Watts-level super-compact narrow-linewidth Tm-doped silica all-fiber laser near 1707nm with fiber Bragg gratings,” Laser Phys. 26(11), 115103 (2016). [CrossRef]  

12. M. D. Burns, P. C. Shardlow, P. Barua, T. L. Jefferson-Brain, J. K. Sahu, and W. Andrew Clarkson, “47 W continuous-wave 1726nm thulium fiber laser core-pumped by an erbium fiber laser,” Opt. Lett. 44(21), 5230–5233 (2019). [CrossRef]  

13. X. Xiao, H. Guo, Z. Yan, H. Wang, Y. Xu, M. Lu, Y. Wang, and B. Peng, “3 W narrow-linewidth ultra-short wavelength operation near 1707nm in thulium-doped silica fiber laser with bidirectional pumping,” Appl. Phys. B 123(4), 135 (2017). [CrossRef]  

14. L. Zhang, J. Zhang, Q. Sheng, S. Sun, C. Shi, S. Fu, X. Bai, Q. Fang, W. Shi, and J. Yao, “Efficient multi-watt 1720nm ring-cavity Tm-doped fiber laser,” Opt. Express 28(25), 37910–37918 (2020). [CrossRef]  

15. Q. Han, J. Ning, and Z. Sheng, “Numerical Investigation of the ASE and Power Scaling of Cladding-Pumped Er–Yb Codoped Fiber Amplifiers,” IEEE J. Quantum Electron. 46(11), 1535–1541 (2010). [CrossRef]  

16. W. Yu, P. Yan, Q. Xiao, T. Qi, D. Li, and M. Gong, “Power scalability of a continuous-wave high-power Er-Yb co-doped fiber amplifier pumped by Yb-doped fiber lasers,” Appl. Opt. 60(7), 2046–2055 (2021). [CrossRef]  

17. P. Peterka, B. Faure, W. Blanc, M. Karásek, and B. Dussardier, “Theoretical modelling of S-band thulium-doped silica fibre amplifiers,” Opt. Quantum Electron. 36(1-3), 201–212 (2004). [CrossRef]  

18. S. D. Jackson and T. A. King, “Theoretical Modeling of Tm-Doped Silica Fiber Lasers,” J. Lightwave Technol. 17(5), 948–956 (1999). [CrossRef]  

19. D. Marcuse, “Loss analysis of single-mode fiber splices,” Bell Syst. Tech. J. 56(5), 703–718 (1977). [CrossRef]  

20. M. A. Khamis and K. Ennser, “Model for a thulium-doped silica fiber amplifier pumped at 1558 nm and 793 nm,” Int. J. Eng. Adv. Technol. 5(4), 76–80 (2016).

21. S. D. Jackson, “The spectroscopic and energy transfer characteristics of the rare earth ions used for silicate glass fibre lasers operating in the shortwave infrared,” Laser Photonics Rev. 3(5), 466–482 (2009). [CrossRef]  

22. G. Canat, J.-C. Mollier, J.-P. Bouzinac, G. M. Williams, B. Cole, L. Goldberg, Y. Jaouën, and G. Kulcsar, “Dynamics of high-power erbium-ytterbium fiber amplifiers,” J. Opt. Soc. Am. B 22(11), 2308–2318 (2005). [CrossRef]  

23. E. Yahel and A. Hardy, “Efficiency optimization of high-power, Er3+-Yb3+-codoped fiber amplifiers for wavelength-division-multiplexing applications,” J. Opt. Soc. Am. B 20(6), 1189–1197 (2003). [CrossRef]  

24. J. Dong, L. Zhang, H. Jiang, X. Yang, W. Pan, S. Cui, X. Gu, and Y. Feng, “High order cascaded Raman random fiber laser with high spectral purity,” Opt. Express 26(5), 5275–5280 (2018). [CrossRef]  

25. Y. Zhang, J. Song, J. Ye, J. Xu, T. Yao, and P. Zhou, “Tunable random Raman fiber laser at 1.7 µm region with high spectral purity,” Opt. Express 27(20), 28800–28807 (2019). [CrossRef]  

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.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (7)

Fig. 1.
Fig. 1. Experimental setup of the 1720nm Tm-doped fiber laser. All the FBGs are written in standard SMF-28e fiber.
Fig. 2.
Fig. 2. The theoretical schematic of the 1720nm Tm-doped fiber laser. PR FBG: partial-reflection fiber Bragg grating. The blue and orange arrows denote the propagation of 1560 nm and 1720nm laser fields. L1 and L2 are the lengths of the Er/Yb-codoped fiber and Tm-doped fibers, respectively.
Fig. 3.
Fig. 3. (a) Theoretical output powers as functions of Tm-doped fiber length and the reflectivity of FBG3; (b) theoretical output powers versus Tm-doped fiber length with a R=45% FBG3. The numbers besides the contour curves in (a) are the output powers.
Fig. 4.
Fig. 4. (a) Power transfer of the 1720nm fiber laser with different Tm-doped fiber lengths; (b) Measured and calculated results of laser threshold and slope efficiency versus Tm-doped fiber length.
Fig. 5.
Fig. 5. (a) Forward and backward 1560 nm laser power (inferred by monitored powers of a 2×2 coupler) versus the launched 976 nm pump power; (b) 1720nm output power versus launched and absorbed 1560 nm power.
Fig. 6.
Fig. 6. Measured laser spectrum at ∼1.1 W output power. Inset: zoomed laser spectrum with a linear scale.
Fig. 7.
Fig. 7. (a) Theoretical 1720nm output powers as a function of Tm-doped fiber length and the reflectivity of output FBG without splicing loss; (b) theoretical power transfer from launched 976 nm pump to 1720nm output power. Inset: Theoretical power transfer from 1560 nm power to 1720nm output power. The numbers beside the contour curves in (a) are the output powers.

Tables (1)

Tables Icon

Table 1. Parameters used in the calculation [15,2023]

Equations (5)

Equations on this page are rendered with MathJax. Learn more.

d P p 1 ± ( z ) d z = ± P p 1 ± ( z ) { Γ p 1 [ σ Yb _ e ( λ p 1 ) N Yb _ 2 ( z ) σ Yb _ a ( λ p 1 ) N Yb _ 1 ( z ) σ Er _ a ( λ p 1 ) N Er _ 1 ( z ) ] α p 1 }
d P s1 ± ( z ) d z = ± P s1 ± ( z ) { Γ s1 [ σ Er _ e ( λ s1 ) N Er _ 2 ( z ) σ Er _ a ( λ s1 ) N Er _ 1 ( z ) ] α s1
d P p2 ± ( z ) d z = ± P p2 ± ( z ) { Γ p2 [ σ Tm _ e ( λ p2 ) N Tm _ 2 ( z ) σ Tm _ a ( λ p2 ) N Tm _ 1 ( z ) ] α p2 }
d P s2 ± ( z ) d z = ± P s2 ± ( z ) { Γ s2 [ σ Tm _ e ( λ s2 ) N Tm _ 2 ( z ) σ Tm _ a ( λ s2 ) N Tm _ 1 ( z ) ] α s2 }
P p1 + ( 0 ) = P , P p1 ( L 1 ) = 0 P s1 + ( 0 ) = R 1 ( 1 δ 1 ) ( 1 δ 2 ) P s1 ( 0 ) P p2 + ( L 1 ) = ( 1 δ 2 ) P s1 + ( L 1 ) P s1 ( L 1 ) = ( 1 δ 1 ) P p2 ( L 1 ) P s 2 ( L 1 + L 2 ) = R 3 P s 2 + ( L 1 + L 2 ) P s2 ( L 1 + L 2 ) = R 3 P s2 + ( L 1 + L 2 ) P p2 ( L 1 + L 2 ) = R 4 P p2  +  ( L 1 + L 2 )
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.