Open Access
Add to BibSonomy
Abstract
Single-frequency fiber lasers at S-, C-, and L-bands play a crucial role in various applications such as optical network expansion, high-precision metrology, coherent lidar, and atomic physics. However, compared to the C-band, the S- and L-bands have wavelength deviations and suffer from excited-state absorption, which limits the output performance. To address this issue, a strategy called ion hybridization has been proposed to increase the differences in site locations of rare earth (RE) ions in the laser matrix, thereby achieving a broader gain bandwidth. This strategy has been applied to an Er3+/Yb3+ co-doped modified phosphate fiber (EYMPF), resulting in gain coefficients per unit length greater than 2 dB/cm at S-, C-, and L-bands. To demonstrate its capabilities, several centimeter-long EYMPFs have been used to generate single-frequency laser outputs at S-, C- and L-bands with kHz-linewidths, high signal-to-noise ratios (>70 dB), and low relative intensity noise (<–130 dB/Hz) in a compact short linear-cavity configuration.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Utilizing the dense wavelength division multiplexing (WDM) transmission systems, the capacity of optical transmission can be significantly increased by expanding the number of channels required by an optical communication system [1]. Currently, the combination of the long-wavelength band (L-band, 1565∼1625 nm) and the conventional band (C-band, 1530∼1565 nm) increases the available bandwidth by 60 nm and enables a total bandwidth of 95 nm [2,3]. However, this solution is only temporary. As a mid-term solution, it is important to expedite the exploitation of the short-wavelength band (S-band, 1460∼1530 nm) to adapt to the highly dynamic traffic patterns. Additionally, single-frequency fiber lasers (SFFL) with ultra-narrow linewidth have excellent temporal coherence, theoretical coherence length over millions of kilometers, low phase noise and a compact structure [4]. In addition to their application prospects in coherent communication systems, SFFLs are also attractive light sources for high-precision metrology, coherent lidar, gravitational wave detection, material micromachining, atomic clocks and quantum computing [5]. Compared with the C-band SFFL, S- and L-bands SFFLs have shorter and longer operation wavelengths, respectively, exhibiting wavelength-oriented characteristics such as higher resolution or lower scattering, which are beneficial for advanced applications in imaging, radar and sensing [6].
So far, there have been two main methods for building SFFLs: ring-cavity and linear-cavity configurations. The ring-cavity architecture, known for its wavelength tunability, is considerably employed in laser devices at extended wavebands [7,8]. However, additional architectures need to be adopted in ring-cavity configuration fiber lasers to guarantee single-longitudinal-mode (SLM) laser oscillation, such as passive multiple-ring cavity, integration of two cascaded fiber Fabry-Perot filters, and unpumped Er3+-doped fiber as a dynamic grating [9,10]. Despite the incorporation of these integration-unfriendly designs, ring-cavity fiber lasers still suffer from mode hopping problems. The short linear-cavity design is more conducive to achieving stable and high-quality laser output in a highly integrated device. However, laser outputs at S- and L-bands are challenging due to the deviation from the center wavelength and serious excited-state absorption, which means elevated gain coefficients within the S-band and L-band are crucial in this particular scenario. Only by utilizing such high-gain fiber, along with meticulously designed gratings and specific fiber lengths, can the output at extended wavebands outperform other longitude modes. Hence, developing a fiber that provides exceptional amplification at S-, C- and L- bands is of utmost importance.
The application of traditional Er3+-doped silica fiber (EDF) is restricted in short linear-cavity SFFLs due to the low rare-earth (RE) solubility (∼1019 ions/cm3) and gain coefficient (∼0.1 dB/cm at S- and L-bands). Previous studies on S- and L-bands SFFLs primarily employed a ring-cavity configuration with meters-long EDFs, leading to poor integration and a risk of mode hopping [11,12]. Er3+/Yb3+ co-doped phosphate fibers with a solubility of ∼1021 ions/cm3 and gain coefficient of ∼5 dB/cm at C-band have been a favorable gain medium for 1.5 µm SFFL [13]. Nevertheless, to achieve laser output at extended wavebands, it is necessary to broaden the gain bandwidth of phosphate fibers. Modifying the fiber composition and the local environment around RE ions is expected to be an effective method.
Herein, an ion-hybridized strategy to introduce variations in the RE sites has been employed to an Er3+/Yb3+ co-doped modified phosphate fiber (EYMPF) to broaden the bandwidth. The effectiveness of this strategy has been confirmed through theoretical molecular dynamics (MD) simulations and experimental characterizations using electron paramagnetic resonance (EPR). As a result, the fiber exhibits a broad effective linewidth of 82 nm and gain coefficients higher than 2 dB/cm at S-, C- and L-bands. This enables stable outputs with linewidths of less than 10 kHz, signal-to-noise ratios (SNR) greater than 70 dB, and relative intensity noise (RIN) less than –130 dB/Hz at the wavelengths of 1519 nm (S-band), 1550 nm (C-band), and 1571 nm (L-band) in EYMPFs shorter than 2 cm.
2. Experimental section
The EYMPF was fabricated using the rod-in-tube technique, with cladding and core glass compositions of 75(Al(PO3)3-KPO3-Ba(PO3)2)-25(BaF2-BaSO4) and 75(Al(PO3)3-KPO3-Ba(PO3)2)-21(BaF2-BaSO4)-2.5Yb2O3-1.5Er2O3 in molar percentage, respectively. Both the cladding and core glasses were prepared by utilizing high-purity reagents (>99.9%) through the conventional melt-quenching technique. The refractive indexes of the core and cladding glasses at 1.5 µm were measured to be 1.5389 and 1.5325, respectively, using a prism coupler (Metricon Model 2010). The self-designed EYMPF has a core diameter of 7.6 µm and a numerical aperture (NA) of 0.14 at 1.5 µm. The propagation loss measured by a cut-back method is ∼0.07 dB/cm at 1310 nm.
The absorption spectrum was recorded using a Lambda 900 UV/VIS/NIR double-beam spectrophotometer (Perkin-Elmer, USA) with an accuracy of 1 nm. The photoluminescence spectrum was measured using a Triax 320 fluorescence spectrometer (Jobin-Yvon, France) with an excitation light source of 980 nm laser diode (LD), and the fluorescence decay curve was recorded with a TDS3012c oscilloscope (Tektronix, USA). The gain characteristics of EYMPF were tested by utilizing a tunable laser operating at 1500∼1625 nm. The input signal light, with an attenuated power of 20 µW, was coupled into a 5-cm-long EYMPF along with the pumping laser fixed at 574 mW from one end. With the signal optical intensity being considerably lower than the saturation optical intensity, the tested gain coefficient approximates the net gain coefficient. The output power of the amplified signal light was measured at the other end of the gain fiber.
3. Results and discussion
3.1 Design: ion hybridization strategy
To broaden the luminescent bandwidth, an ion hybridization strategy has been proposed based on the cognition that changes in the distribution and local environment of RE ions can affect luminescent behaviors [6,14]. The presence of different cations and anions can adjust the polymerization degree of the long- and medium-range-order structure, as well as the polarization degree of the electron cloud around RE ions in the short-range order. To explain this mechanism, a theoretical analysis was conducted using MD simulation to characterize the local environment around the RE ions. In Fig. 1(a), the modeled structure of Er3+-doped Al(PO3)3-KPO3-Ba(PO3)2 glass with approximately 10000 atoms is depicted. The model was generated using the melt and quench process based on a randomly initialized configuration equilibrated at 8000 K. The partial charge pairwise potentials and short-range interactions in the form of Buckingham potentials were used in the simulation. The potential parameters, relaxation setup, and processing of structural descriptors followed the methods outlined in [15].
Fig. 1. (a) Screenshot of the simulated glass structure. (b) RDF of Er-P, Er-Al, Er-K and Er-Ba. Inset: CN analysis. (c) RDF and angle distribution function (ADF) of Er-O, Er-BO (Bridging Oxygen) and Er-NBO. (d-e) Redrawn version of 2D-HYSCORE and EDFS EPR results from [17]. (f) Schematic diagram of hybridized ligands.
Figure 1(b) displays the radial distribution function (RDF) and coordination number (CN) for Er-P, Er-Al, Er-K and Er-Ba. The first peak in the RDF corresponds to the mean bond lengths between Er-P, Er-Al, Er-K and Er-Ba, which are 3.62 Å, 3.42 Å, 3.78 Å and 3.75 Å, respectively. This indicates that these cations are all located in the sub-nearest layer to Er. These structural data are consistent with previous literature reports, further validating the accuracy of the simulated model [15,16]. Around Er, P is the most abundant with a CN of 5.82. The distribution of K and Ba around Er is secondary, with CNs of 0.72 and 1.06, respectively, while the distribution of Al is the least. Additionally, the distribution of near-neighbor oxygen ions around the RE ions is analyzed. As shown in Fig. 1(c), the bond length and CN of Er-O are 2.3 Å and 6, respectively. The bond angle distribution confirms that non-bridging oxygen (NBO) surrounds Er in an octahedral six-coordinated manner, with a predominant bond angle of approximately 90 degrees. The minor 160-degree peak in the O-Er-O connection may be attributed to five-coordinated Er.
In addition to the cations in the second shell around RE ions, the anions in the first shell around RE are also capable of hybridization. Previous studies have demonstrated that the introduction of fluorides and sulfates into phosphate can effectively modify the local environment of RE ions. Figure 1(d) and (e) show a redrawn version of the pulse sequence of the two-dimensional hyperfine sublevel correlation (2D-HYSCORE) and the echo-detected field sweep (EDFS) EPR from our previous work, illuminating the modification effect [17]. The split 19F signal observed on the non-diagonal line in Fig. 1(d) indicates strong coupling, suggesting the formation of a RE-F chemical bond. The shift and broadening of the peak in Fig. 1(e) indicate a difference in the occupation site of RE ions after sulfate addition, which implies a potential direct linkage between sulfate anions and RE ions.
Through theoretical and experimental analysis, a comprehensive picture of the complex ligands around RE ions in the modified fiber has been achieved, including network formers [PO4], intermediates [AlOn], modifiers [KOn], [BaOn], F and [SO4], as depicted in Fig. 1(f). The results confirm that the ligand hybridization strategy effectively facilitates site-to-site differences of RE ions, which is advantageous for achieving inhomogeneous broadening. Furthermore, different ligands exert varying electrostatic forces on the electron cloud surrounding RE ions, which can modify the radiation transition behavior. Specifically, the presence of phosphate in the vicinity reduces the likelihood of phase separation and concentration quenching compared to a rigid silica network [18]. Additionally, small-field-strength cations such as K and Ba decrease the local basicity of RE sites, increase the degree of overlap between the 4f and 5d orbitals, and enhance the 1.5 µm luminescence of Er3+ [19].
3.2 Property: comparison between EYMPF and other fibers
After the theoretical design and compositional optimization, the Er3+/Yb3+ co-doped modified phosphate laser glass and fiber were prepared. The absorption and emission spectra, as well as gain coefficients, are displayed in Fig. 2. The relevant properties and comparison with silica and phosphate fibers are summarized in Table 1. The absorption cross-section (σa) at 976 nm is 5.4 × 10−20 cm2, significantly higher than silica (0.2 × 10−20 cm2) and another phosphate glass (1.5 × 10−20 cm2) [20,21]. Meanwhile, the emission bandwidth (Δλeff) of EYMPF reaches 82 nm, which is 2 times that of silica and 1.5 times that of another phosphate glass [22,23]. The emission cross-sections (σe) are 2.8 × 10−21 cm2 at 1515 nm, 6.5 × 10−21 cm2 at 1550 nm, and 3.0 × 10−21 cm2 at 1603 nm, respectively. Notably, due to the wider emission bandwidth, the emission cross-sections at non-central wavelength are almost 3 times that of silica glass [24]. Fluorescence lifetime is crucial for population inversion and efficient output. In EYMPF, even with heavy doping (8 mol% RE3+), the lifetime can still reach 7.23 ms as given in Supplement 1, longer than silica glass with lower doping concentration.
Fig. 2. Spectral characteristics of EYMPF. (a) Absorption spectrum. (b) Emission spectrum. (c) Absorption and emission cross-sections. (d) Gain coefficients.
Table 1. Comparison between properties of silica, phosphate and EYMPF
The net gain coefficient of EYMPF at 1550 nm reaches 5.2 dB/cm, surpassing that of silica fiber and comparable to phosphate fiber. Additionally, at S- and L-bands, the EYMPF demonstrates high gain coefficients of 2.3 dB/cm at 1515 nm and 2.5 dB/cm at 1600 nm, respectively, which are 10 times greater than silica and 2 times that of phosphate fiber [25–27]. The gain coefficient exhibits a positive correlation with the emission cross-section and the population in the upper energy level. In the case of EYMPF, the elevated gain coefficient observed at S- and L-bands stems from its broad bandwidth, which results in a heightened emission cross-section, coupled with a relatively longer fluorescence lifetime ensured by the optimized melting and dehydration processes. Through adjusting the fabrication and purification processes, these properties could be improved in the future. Overall, the main strength of EYMPF lies in its remarkably broad bandwidth, which leads to high σe and gain coefficients at extended wavebands, rendering it a promising gain medium for S- and L-bands SFFLs.
3.3 Application: SFFLs at S-, C-, and L-bands
To evaluate the applicability of the EYMPF for S-, C-, and L-bands, compact distributed brag reflector (DBR) SFFLs are constructed. The experimental setup of the SFFLs is illustrated in Fig. 3. The inset graphs depict the as-drawn fiber, revealing a well-preserved core/cladding structure. The two end facets of cm-long EYMPF are butt-coupled to one spectrally narrow-band fiber Bragg grating (NB-FBG) and one spectrally wide-band fiber Bragg grating (WB-FBG). The FBGs are directly imprinted in standard Corning SMF-28e fibers. The center wavelengths of WB-FBG and NB-FBG are 1518.9 nm, 1550.2 nm, and 1570.9 nm for S-, C- and L-bands SFFLs, respectively. The WB-FBGs have a reflectivity of 99.9% at center wavelength and a 3-dB bandwidth of ∼0.4 nm, while the NB-FBGs have a reflectivity of ∼70% at center wavelength and a 3-dB bandwidth of 0.072∼0.096 nm. The cavity is pumped by a 976 nm laser diode (LD) through a 976/1550 nm wavelength-division multiplexer (WDM) fiber coupler with a wavelength tolerance of ±30 nm centered at 1550 nm. The output laser from the NB-FBG end travels through the WDM and isolator (ISO). All the components are connected by standard fibers. The optical spectra were measured using an optical spectrum analyzer (AQ6370D, Yokogawa, Japan) with a resolution of 0.02 nm. All experiments were carried out at room temperature and the key parameters are summarized in Table 2.
Fig. 3. Experimental setup of the compact DBR short-cavity SFFL. WB-FBG, wide band fiber Bragg grating; EYMPF, Er3+/Yb3+ co-doped modified phosphate glass fiber; NB-FBG, spectrally narrow-band fiber Bragg grating; WDM, wavelength-division multiplexer; ISO, isolator; LD, laser diode. Inset: picture and electron micrograph image of the EYMPF
Table 2. Key parameters of EYMPF-based SFFLs
The optical spectrum of the S-band SFFL is presented in Fig. 4(a). The laser has a central wavelength of 1519 nm and an SNR exceeding 70 dB. No signs of ASE or parasitic lasing appear in the spectrum. The longitudinal mode characteristics are measured using a scanning Fabry-Perot interferometer (FPI) with a free spectral range (FSR) of 1.5 GHz. The fine spectral structure of the laser signal reveals only two main peaks within one FSR, confirming the SLM operation. To measure the laser linewidth, a delayed self-heterodyne method was performed with a 10-km-long single-mode fiber delay under full power operation (∼450 mW). The result is presented in Fig. 4(b). The typical heterodyne signal is fitted to a Lorentzian profile to estimate the spectral linewidth, which is found to be 200 kHz at a 20-dB linewidth, indicating a 10-kHz linewidth. The relationship between the output power at 1519 nm and the pump power at 976 nm is displayed in Fig. S2 (a) (see Supplement 1). The laser exhibits a threshold of approximately 90 mW and a slope efficiency of 8.3%. To assess the RIN of the fiber laser, an electrical spectrum analyzer is employed, and the results are given in Fig. S2 (b). The RIN spectra are dominated by peaks at the relaxation oscillation frequency (ROF), which occurs at around 0.2 MHz, with an RIN level of approximately −85 dB/Hz. The RIN level then gradually decreases as the frequency increases. At frequencies above 10 MHz, the RIN of the SFFL is less than −135 dB/Hz.
Fig. 4. The laser spectrum and self-heterodyne spectrum of the (a-b) S-, (c-d) C- and (e-f) L-bands SFFLs. Inset graphs: longitudinal mode characteristics measured by a scanning Fabry-Pérot interferometer.
Similarly, SFFLs operating at C- and L-bands have been constructed. The C-band SFFL shows a central wavelength of 1550 nm, an SNR exceeding 75 dB, and a 3dB-linewidth of 8 kHz. According to Fig. S3, the slope efficiency is 9.6%, and the ROF is 1.2 MHz, resulting in an RIN of approximately −95 dB/Hz, which decreases toward high frequencies. For frequencies above 10 MHz, the RIN of the SFFL is approximately −135 dB/Hz. The laser parameters of the L-band SFFL are presented in Fig. 4 and Fig. S4, with a central wavelength of 1571 nm, an SNR higher than 75 dB, and a linewidth of 6 kHz. The slope efficiency and obtained RIN above 10 MHz are 9.2% and −130 dB/Hz, respectively. All lasers exhibit good power stability according to a 2-hour stability record with root mean square (RMS) noises less than 1.5% at S-, C- and L-bands, as illustrated in Fig. S5.
The above results confirm the potential application of EYMPF in the S-, C-, and L-bands SFFLs. It should be pointed out that SFFLs operating at extended wavebands usually require the employment of as long fiber as possible to ensure efficient output lasing, which contradicts the necessity of enlarging the mode spacing for SLM operation. The outstanding amplification capability of EYMPF at S-, C-, and L-bands now permits the use of only centimeter-long fibers to achieve single-frequency laser outputs at these bands. Furthermore, a comparison between the EYMPF-based DBR SFFL in our study and the ring-cavity and DFB (distributed feedback Bragg reflector) configuration lasers utilized in previous research is provided in Table 3. Despite the advantage in wavelength tunability, the compactness and stability have been sacrificed in ring-cavity configuration laser devices considering the tens-of-meters-long fiber used. In contrast, the EYMPF-based DBR SFFL displays superior performance in terms of linewidth, stability and integration. Compared to EDF-based DFB lasers, the EYMPF-based DBR SFFLs demonstrate superior amplification capabilities across multiple bands, along with a shorter working length and the absence of etching requirements. Furthermore, it should be noted that there is still potential for further enhancement of laser performance based on EYMPF. Lasing further down at S-band and further up in at L-band could be achieved with even narrower linewidths with a higher reflectivity NB-FBG.
Table 3. Comparison of current work with ring-cavity and DFB configuration SFFLs at S-, C- and L-bands
4. Conclusions
A modification mechanism that expands the gain bandwidth by hybridizing the coordination environment of rare earth ions has been proposed. As a demonstration, an Er3+/Yb3+ co-doped modified phosphate fiber with promising amplification ability at S-, C-, and L-bands has been designed and fabricated, which allows for the generation of ultra-narrow-linewidth single-frequency outputs at multi bands with centimeter-long fibers. The single-longitudinal-mode characteristics of these outputs include a linewidth of 10 kHz, SNR greater than 70 dB, and RIN lower than –125 dB/Hz at S-band, a linewidth of 8 kHz, SNR greater than 75 dB, and RIN lower than –130 dB/Hz at C-band, a linewidth of 6 kHz, SNR greater than 75 dB, and RIN lower than –120 dB/Hz at L-band. Besides, these outputs are devoid of issues like mode hopping and redundant structures commonly found in ring cavities. The ion-hybridized fiber offers a reliable solution for obtaining high-performance single-frequency output across extended wavebands, and this work opens the door for further exploration and utilization of hybridization strategy in other matrices or novel wavebands.
Funding
State Key Laboratory of Luminescent Materials and Devices (Skllmd-2023-01); Fundamental Research Funds for the Central Universities (2023ZYGXZR099); National Natural Science Foundation of China (52130201, 52172003, 62275082).
Acknowledgments
The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (52130201, 52172003, 62275082), Fundamental Research Funds for the Central Universities (2023ZYGXZR099) and the State Key Lab of Luminescent Materials and Devices, South China University of Technology (Skllmd-2023-01).
Disclosures
The authors declare no conflicts of interest.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Supplemental document
See Supplement 1 for supporting content.
References
1. J. Y. Liu, J. X. Zhang, J. Liu, et al., “1-Pbps orbital angular momentum fibre-optic transmission,” Light: Sci. Appl. 11(1), 202 (2022). [CrossRef]
2. B. J. Puttnam, R. S. Luís, G. Rademacher, et al., “S-, C-and L-band transmission over a 157 nm bandwidth using doped fiber and distributed Raman amplification,” Opt. Express 30(6), 10011–10018 (2022). [CrossRef]
3. R. Emmerich, M. Sena, R. Elschner, et al., “Enabling SCL-band systems with standard C-band modulator and coherent receiver using coherent system identification and nonlinear predistortion,” J. Lightwave Technol. 40(5), 1360–1368 (2022). [CrossRef]
4. S. J. Fu, W. Shi, Y. Feng, et al., “Review of recent progress on single-frequency fiber lasers,” J. Opt. Soc. Am. B 34(3), A49–A62 (2017). [CrossRef]
5. D. Grassani, E. Tagkoudi, H. Guo, et al., “Mid infrared gas spectroscopy using efficient fiber laser driven photonic chip-based supercontinuum,” Nat. Commun. 10(1), 1553 (2019). [CrossRef]
6. X. Y. Huang, H. H. Cheng, W. Luo, et al., “Er-activated hybridized glass fiber for laser and sensor in the extended wavebands,” Adv. Opt. Mater. 9(24), 2101394 (2021). [CrossRef]
7. X. Li, J. Mou, L. Xiong, et al., “Fractional-order double-ring erbium-doped fiber laser chaotic system and its application on image encryption,” Opt. Laser Technol. 140, 107074 (2021). [CrossRef]
8. F. Kalantarifard, P. Elahi, G. Makey, et al., “Intracavity optical trapping of microscopic particles in a ring-cavity fiber laser,” Nat. Commun. 10(1), 2683 (2019). [CrossRef]
9. R. Xiang, C. C. Wang, and L. Lu, “Laser Doppler velocimeter using the self-mixing effect of a fiber ring laser with ultra-narrow linewidth,” J. Opt. 48(3), 384–392 (2019). [CrossRef]
10. C.-H. Yeh, T.-T. Huang, H.-C. Chien, et al., “Tunable S-band erbium-doped triple-ring laser with single-longitudinal-mode operation,” Opt. Express 15(2), 382–386 (2007). [CrossRef]
11. Z. H. Jiang and Q. Y. Zhang, “The formation of glass: a quantitative perspective,” Sci. China Mater. 58(5), 378–425 (2015). [CrossRef]
12. C.-H. Yeh and C.-W. Chow, “Broadband wavelength-tunable single-longitudinal-mode erbium-doped fiber ring laser using saturable-absorber filter,” Laser Phys. Lett. 7(2), 158–163 (2010). [CrossRef]
13. Q. Y. Zhang, L. L. Hu, and Z. H. Jiang, “Thermodynamic study on elimination of platinum inclusions in phosphate laser glasses for inertial confinement fusion applications,” Chin. Sci. Bull. 44(7), 664–668 (1999). [CrossRef]
14. M. J. Weber, D. C. Ziegler, and C. A. Angell, “Tailoring stimulated emission cross sections of Nd3+ laser glass: Observation of large cross sections for BiCl3 glasses,” J. Appl. Phys. 53(6), 4344–4350 (1982). [CrossRef]
15. J. C. Du, L. Kokou, J. L. Rygel, et al., “Structure of cerium phosphate glasses: molecular dynamics simulation,” J. Am. Ceram. Soc. 94(8), 2393–2401 (2011). [CrossRef]
16. J. J. Zhao, X. X. Xu, P. C. Li, et al., “Structural origins of RF3/NaRF4 nanocrystal precipitation from phase-separated SiO2–Al2O3–RF3–NaF glasses: a molecular dynamics simulation study,” J. Phys. Chem. B 123(13), 3024–3032 (2019). [CrossRef]
17. Y. Ji, C. Y. Shao, D. D. Chen, et al., “Modification of structure and properties of fluoro-sulfo-sodium alumino-phosphate glass through additional fluorine and sulfur agents,” J. Non-Cryst. Solids 617, 122515 (2023). [CrossRef]
18. J. C. Du and A. N. Cormack, Atomistic Simulations of Glasses: Fundamentals and Applications (Wiley, 2022).
19. S. Tanabe, T. Hanada, T. Ohyagi, et al., “Correlation between Eu151 Mössbauer isomer shift and Judd-Ofelt Ω6 parameters of Nd3+ ions in phosphate and silicate laser glasses,” Phys. Rev. B 48(14), 10591–10594 (1993). [CrossRef]
20. C. S. Yang, X. C. Guan, W. Lin, et al., “Efficient 1.6 µm linearly-polarized single-frequency phosphate glass fiber laser,” Opt. Express 25(23), 29078–29085 (2017). [CrossRef]
21. A. Meissner, P. Kucireka, J. Lib, et al., “Simulations and experiments on resonantly-pumped single-frequency Erbium lasers at 1.6 µm,” Proc. SPIE 8599, 85990H (2013). [CrossRef]
22. Q. Wang, N. Dutta, and R. Ahrens, “Spectroscopic properties of Er doped silica glasses,” J. Appl. Phys. 95(8), 4025–4028 (2004). [CrossRef]
23. F. X. Wang, F. Song, S. X. An, et al., “Er3+/Yb3+-codoped phosphate glass for short-length high-gain fiber lasers and amplifiers,” Appl. Opt. 54(5), 1198–1205 (2015). [CrossRef]
24. M. A. Khamis and K. Ennser, “Theoretical model of a thulium-doped fiber amplifier pumped at 1570 nm and 793 nm in the presence of cross relaxation,” J. Lightwave Technol. 34(24), 5675–5681 (2016). [CrossRef]
25. S. W. Harun, K. Dimyati, K. K. Jayapalan, et al., “An overview on S-band erbium-doped fiber amplifiers,” Laser Phys. Lett. 4(1), 10–15 (2007). [CrossRef]
26. S. H. Xu, Z. M. Yang, Z. M. Feng, et al., “Gain and noise characteristics of single-mode Er3+/Yb3+ Co-doped phosphate glass fibers,” International Nanoelectronics Conference633–635 (IEEE, 2008).
27. Z. Zhai and J. K. Sahu, “Extending L-Band Gain to 1625 nm Using Er3+: Yb3+ Co-Doped Silica Fibre Pumped by 1480 nm Laser Diodes,” Opt. Lett. 46(23), 5834 (2021). [CrossRef]
28. X. L. Zou and T. Izumitani, “Spectroscopic properties and mechanisms of excited state absorption and energy transfer upconversion for Er3+-doped glasses,” J. Non-Cryst. Solids 162(1-2), 68–80 (1993). [CrossRef]
29. W. C. Wang, B. Zhou, S. H. Xu, et al., “Recent advances in soft optical glass fiber and fiber lasers,” Prog. Mater. Sci. 101, 90–171 (2019). [CrossRef]
30. H. Ahmad, M. Z. Zulkifli, A. A. Latif, et al., “Tunable single longitudinal mode S-band fiber laser using a 3 m length of erbium-doped fiber,” J. Mod. Optic. 59(3), 268–273 (2012). [CrossRef]
31. S. J. Feng, Q. H. Mao, Y. Y. Tian, et al., “Widely tunable single longitudinal mode fiber laser with cascaded fiber-ring secondary cavity,” IEEE Photonics Technol. Lett. 25(4), 323–326 (2013). [CrossRef]
32. R. A. Perez-Herrera, A. Ullan, D. Leandro, et al., “L-band multiwavelength single-longitudinal mode fiber laser for sensing applications,” J. Lightwave Technol. 30(8), 1173–1177 (2012). [CrossRef]
33. N. Photonics, “NKT Photonics Koheras BasiK, low noise, sub-kHz linewidth, single frequency fiber laser” (NKT Photonics, 2023), https://photonicshop.co.uk/products/nkt-photonics-koheras-basik-low-noise-sub-khz-linewidth-frequency-fiber-laser.
