We report on a newly designed and fabricated ytterbium-doped large mode area fiber with an extremely low NA (~0.04) and related systematic investigations on fiber parameters that crucially influence the mode instability threshold. The fiber is used to demonstrate a narrow linewidth, continuous wave, single mode fiber laser amplifier emitting a maximum output power of 3 kW at a wavelength of 1070 nm without reaching the mode-instability threshold. A high slope efficiency of 90 %, excellent beam quality, high temporal stability, and an ASE suppression of 70 dB could be reached with a signal linewidth of only 170 pm.
© 2016 Optical Society of America
Average power scaling of narrow linewidth solid state lasers with nearly diffraction limited beam quality is a challenging task. Power scaling is typically limited by nonlinear optical effects (NLE) and/or degradation of beam quality due to thermo-optical effects. However, some applications such as beam combining make high demands on beam quality and linewidth of the laser sources and request even higher power levels [1,2].
As one of the most promising solid state laser concepts, fiber lasers are known for their ability to emit laser light with high average power and excellent beam quality combined with easy thermal management, high reliability, compact size, and high efficiency [3,4]. During the last decades average power scaling of ytterbium doped fiber lasers with nearly diffraction limited beam quality increased exponentially . In 2009 further power scaling stopped rapidly due to the appearance of a new limitation called mode instabilities (MI) .
Today, power scaling of single mode fiber lasers is still not straightforward and is mainly limited due to NLE especially for fibers with relatively small mode field diameters and MI especially for fibers with relatively large mode field diameters operating on the edge of true single-mode guidance. NLE such as stimulated Raman scattering (SRS) and stimulated Brillouin scattering (SBS)  lead to energy transfer to (usually) unwanted wavelengths and thus limit narrow linewidth power scaling. MI is a threshold-like effect, which leads to significant, temporally dynamic power transfer from the fundamental mode to higher order modes as soon as the system characteristic threshold power has been reached. This power transfer originates from thermally induced long-period index gratings . As a result of these limitations, today’s average output power of directly diode pumped laser systems with high electrical to optical efficiency of typically more than 30 % is limited to 1.0-2.5 kW output power [8–10]. Few systems with higher average power and good beam quality have been reported: In 2009 an output power of 10 kW from an industrial single mode fiber laser based on tandem pumping technology was claimed but unfortunately no beam quality measurements have been published [11,12]. Recently, a 3 kW laser system based on laser architecture with free-space components  as well as a 3.15 kW monolithic MOPA (master oscillator power amplifier) system  were demonstrated. The spectral linewidth of these three exemplary systems was dependent on the output power and showed a broadband spectral linewidth in the range of some nanometers at highest output power level due to NLE (self-phase modulation, SBS, and SRS) causing spectral broadening. Thus they are unsuitable for applications requiring narrow linewidth spectra as mentioned above. Furthermore, beam quality of the systems in  and  was characterized via M2-measurement. However, the results in  showed that for a clear proof of single-mode emission and the proof of MI faster measurement techniques are required, which temporally resolve the dynamic of MI in the kHz range. This is important, especially if the fiber laser system emits “near” diffraction limited output, which is predestinated to show MI. Unfortunately, such systematic experimental investigations towards the MI threshold are missing for the systems presented in [13,14] but they are important to judge the system performance, compare fiber laser system as well as understand the physical origin of limitations in order to develop mitigation strategies. Furthermore, a narrow linewidth laser system with an output power of 2.9 kW and a spectral linewidth of 0.31 nm was reported on . The beam quality of the system was degraded at an output power higher than 2.5 kW by MI and thus is not suitable for single mode requiring applications at maximum output power.
In this paper we systematically investigate the impact of two important parameters, namely the average heat load and the guided mode content controlled by the fiber design, its fabrication and fiber handling. These two parameters have previously been identified to crucially impact the MI threshold [17,18]. From these investigations a fiber design is deduced and fabricated that simultaneously prevent from both NLE and MI. The fiber design is based on a step index fiber with large mode area and low numerical aperture (NA) fabricated by modified chemical vapor deposition (MCVD)-technology getting along without complex nano- or micro-structuring techniques. As a result of optimizing the fiber core diameter, and the ytterbium concentration, an output power of 3 kW with excellent slope efficiency of 90 %, single-mode beam quality (), and a spectral linewidth of 170 pm could be reached. No evidence for NLE or MI was observed and further power scaling was pump power limited. To the best of our knowledge, our results present a narrow linewidth (170 pm), single mode fiber amplifier with highest output power ever reported on.
2. Preform fabrication and experimental setup
The aim was to design and fabricate fibers simultaneously preventing from both, NLE and MI. Up to date, the highest average output power of a single mode, narrow linewidth fiber amplifier reported on is 2.3 kW . The results were achieved with a fiber amplifier built from a commercially available fiber (Nufern YDF) with a core diameter of 20 µm and a NA of 0.06. Further power scaling was limited due to NLE. A necessary approach for further power scaling thus is to increase the core diameter to prevent from NLE. In order to still guarantee single mode guiding and to simultaneously prevent from MI, for fibers with increased core diameter the NA has to be reduced. Fibers with reduced NA were realized by reducing the ytterbium concentration, which also has some additional important advantages: firstly, the pump light absorption length is increased and secondly photodarkening losses are reduced. Both result in a reduced average heat load, which is advantageous to prevent from MI .
As dopants for the fiber core ytterbium, aluminum and phosphorus were chosen. Ytterbium acts as the active ion, which of course is essential for the laser process. Aluminum serves as solutizer for the ytterbium ions. Phosphorus has several features: firstly, it also serves as solutizer for the ytterbium ions [19,20]. Secondly, it is known to reduce photodarkening  and thus is advantageous for raising the MI threshold. Thirdly, it can help to reduce the refractive index in combination with aluminum. Separate doping with aluminum, phosphorus as well as ytterbium increases the refractive index. However, there is an anomalous behavior when aluminum and phosphorus dopants are mixed: For equal aluminum and phosphorus concentrations the refractive index nearly stays unaffected compared to undoped SiO2 . This behavior was used to fabricate extremely low NA fibers, where the NA mainly depends on the ytterbium concentration.
For fiber preform fabrication we have chosen the well-proven MCVD technology in combination with a solution doping step. One porous, phosphorous doped SiO2 layer was deposited on the inside of a fused silica tube (F300) and presintered to tailor its porosity. Afterwards, aluminum and ytterbium ions were doped into this layer by using the solution doping method . Typical doping concentrations of around 1 mol% phosphorus, 1 mol% aluminum, and 0.1 mol% ytterbium were chosen. Hence, our fabricated fibers preforms contain only half of the ytterbium concentration of the commercially available fiber used in . For the high power amplifier tests double clad fibers are required, guiding the signal in the core and the pump light in the inner cladding. Thus the preforms were jacketed to adjust the required core to cladding ratio of around 1:20. Afterwards they were grinded into an octagonal shape to increase the pump light absorption and finally drawn to a fiber. The fiber absorption length was determined to be 30 m concerning a cladding absorption of more than 95 % of pump light at 976 nm.
The experimental high power amplifier setup for testing the Yb-doped fibers is shown in Fig. 1. The pump light at a wavelength of 976 nm (Laserline GmbH, wavelength stabilized by water cooling) was coupled into the pump core (from the right in Fig. 1) while the narrow linewidth (170 pm) seed signal of around 10 W at a central wavelength of 1070 nm was coupled into the Yb-doped signal core in counter-propagational direction (from the left in Fig. 1). To enhance the SBS threshold, the spectral linewidth of 170 pm was reached by phase modulation of a commercially available external cavity diode laser operating single-frequency. Dichroic mirrors separated the amplified signal from the pump light as well as the seed from the transmitted pump light. The amplified signal was analyzed in terms of spectral and temporal behavior, beam quality and output power. In order to detect the MI threshold, a small fraction (< 0.01%) of the high-power near field signal was detected with an InGaAs photo diode (rise time 10 ns) connected to an oscilloscope. Since the beam was larger than the detector, only a spatial fraction was detected resulting in a constant signal for stable mode operation and a strongly modulated signal for unstable mode operation due to the spatially fluctuating power distribution of the unstable mode mix as discussed in .
All fibers under test were inserted into the amplifier setup in Fig. 1 and characterized one after the other. The amplifier systems reproducibility was carefully checked in numerous repetitions of similar experiments.
3. Experimental results
In a first step a first preform with an ytterbium concentration of around 0.09 mol%, a phosphorus concentration of around 1 mol% and an aluminum concentration of around 0.8 mol%, as can be extracted from the electron probe micro analysis (EPMA)  results shown in Fig. 2(a), was fabricated. The doping concentration resulted in a very small core NA of 0.042. The NA was calculated from the mean refractive index within the core, defined as central region, where the entire index profile is above the level of the pure silica cladding level [Fig. 4(a)]. All mode calculations take into account the complete index profile of the fiber, thus this definition is only used for adjusting the core to cladding diameter ratio during fabrication of the fiber. The preform was jacketed and drawn to a fiber with a core diameter of 24.5 µm and an inner cladding diameter of 500 µm (Fiber I).
In a first amplifier test a fiber length comparable to the fiber length used in  was chosen to stay comparable. Thus, Fiber I with a length of 18 m was cooperated into the setup shown in Fig. 1 with a bending diameter of around 1 m. Due to the reduced ytterbium concentration of Fiber I the amplifier was driven ineffectively leading to a large amount of pump light that was not absorbed in the amplifier fiber of 18 m lengths. To prevent the seed source from the residual pump light a high power mode stripper was added at the fiber side, where the seed was coupled in. A maximum signal output power of 1 kW could be achieved with Fiber I. Further power scaling was limited by MI as is shown in Fig. 2(b) (green circles). Above the MI threshold the photodiode signal was strongly modulated, which is reflected in an increase of standard deviation of the time traces , shown in Fig. 2(b).
In order to increase the MI threshold a second experiment with an increased fiber length of 30 m was performed to reduce the average heat load. Setting up an efficiently running amplifier with the increased fiber length of 30 m could only be realized due to the reduced Yb concentration, which results in a longer pump power absorption length. An output power of 1.6 kW could be demonstrated without the occurrence of MI, as is shown in Fig. 2(b) (green squares). Further power scaling was limited by fiber destruction, probably caused by coating defects. Nevertheless, the experiment proves that the average heat load is a key aspect for MI: the lower the average heat load, the higher the MI threshold.
Theoretically the MI threshold is predicted for an average heat load of 34 W/m . We calculated an average heat load of only 10 W/m even for the experiment with the short fiber of 18 m. Thus, the MI threshold should theoretically not be reached in this experiment. Consequently, a second parameter seems to additionally influence the MI threshold crucially. Previous investigations indicated that also the mode content influences the MI threshold, because higher order modes can seed MI . To clarify the dependence of the mode content on the MI threshold, we repeated the experiment with the 18 m long fiber. This time the bending diameter was reduced to around 0.5 m to discriminate higher order modes more effectively. The mode instability threshold was increased up to 1.3 kW as is shown in Fig. 2(b). This experiment confirms that discrimination of higher order modes increases the MI threshold.
Additionally, we analyzed the higher order mode content via S2 measurement with a commercial device (FMA-100 Fiber Mode Analyzer, Interfiber Analysis) . The results are shown in Fig. 4(b): for a bending diameter of around 0.5 m (comparable to the amplifier setup configuration) a significant amount of higher order mode content is guided in the fiber, which can be concluded by the peak at a delay of around 2 ps clearly representing the LP 11 mode. This mode could be suppressed by tighter bending to diameters smaller than 0.3 m (also shown in Fig. 3(b). However, for these small bending diameters the attenuation for the fundamental mode, as well, is significant, which would reduce the amplifier performance.
Taking into account the results with Fiber I one promising approaches to further increase the MI threshold arises: The fibers need to be single mode. This can be arranged by reducing the V-parameter by either reducing the core diameter or the core NA. Both approaches were realized by fabricating two new fibers, namely Fiber II and Fiber III as schematically shown in Fig. 3.
The refractive index profiles of all three fibers are shown in Fig. 4(a). Fiber II was fabricated from the same preform as Fiber I, but it was drawn to a slightly smaller diameter of 22.0 µm. Fiber III was drawn from a newly designed preform with a slightly reduced NA of around 0.040 (NA = 0.042 for Fiber I and Fiber II). The concentration of the ytterbium ions was comparable and the preform was drawn to a fiber with a core diameter of 24.5 µm like Fiber I. The mode content of both fibers was again analyzed via S2-measurement [Fig. 4(b)]. For both fibers no higher order mode content (>50 dB suppression) could be observed even at large bending diameters of around 1 m indicating that Fiber II as well as Fiber III are single mode.
Fiber II and Fiber III were inserted into the same high power amplifier setup that has already been used to characterize Fiber I (Fig. 1). For both fibers a length of 30 m was chosen for optimized pump light absorption. Both fibers were bended to a diameter of around 1 m to avoid significant attenuation of the fundamental mode.
Fiber II showed a very good amplifier performance with a very high slope efficiency of 90 % with respect to the absorbed pump power. A maximum output power of 2.7 kW with a linewidth of 170 pm and without any indications of MI was reached. The standard derivation of the time trace of the photodiode signal is plotted in Fig. 6 showing no significant increase. This experiment in comparison with the results performed with Fiber I impressively demonstrates the crucial dependence of the mode content on the MI threshold. Fiber II has exactly the same fiber parameters as Fiber I since they were drawn from the same preform except of the core diameter influencing the guided mode content. Just by reducing the core diameter from 24.5 µm to 22.0 µm the MI threshold could be increased from 1 kW to more than 2.7 kW for comparable bending diameters. Up to the maximum output power we have not seen any indication for NLE especially SBS, which is expected to be the dominant effect for such narrow linewidth signals, yet. We have measured the SBS threshold for a signal with a reduced linewidth of 100 pm to be around 1.7 kW in a preliminary experiment using the same setup. SBS was detected as powerful spikes in the back scattered signal. Taking the linewidth dependence of SBS  into account, the SBS threshold for a signal linewidth of 170 pm would be expected to be at 2.8 kW. Thus, the maximum output power of 2.7 kW seems to be very close to the SBS threshold.
To avoid the destruction of the setup by high power Brillouin Peaks travelling in counter-propagational direction to the signal, we stopped further power scaling with Fiber II.
Due to the larger core diameter and the reduced NA, which both contribute to the effective mode area from 290 µm (Fiber II) up to 360 µm2 (Fiber III), an increased SBS threshold of around 3.35 kW is calculated for Fiber III concerning a signal linewidth of 170 pm. The results of the amplifier tests with Fiber III are shown in Figs. 5(a)–5(b) and Fig. 6. Fiber III again showed a very high slope efficiency of 90 % [Fig. 5(a)] and a maximum output of 3 kW was reached. No indications for MI [Fig. 6] or NLE [Fig. 5(b)] were observed. Further power scaling was pump power limited. The signal was single mode (as already proven by the S2 measurement in unpumped configuration) and a beam quality of M2 = 1.3 was measured (using the knife edge method , which is less sensitive towards residual signal light guided in the cladding). The amount of amplified spontaneous emission (ASE) was suppressed by more than 70 dB below the peak power at the signal wavelength as can be seen in Fig. 5(b). A total amount of 99.5 % of the signal power was contained within the spectral linewidth of 170 pm at a central wavelength of 1067 nm. At the maximum output power of 3.0 kW, a pump power of 4.54 kW was launched into the fiber, resulting in an optical efficiency of 66 %. A residual pump power of 330 W was measured at a launched pump power of 4.5k W. The pump power coupling efficiency was measured to be 80 %. Thus, the pump absorption of the fiber was determined to be 10.4 dB.
The only experiment, where the MI threshold could directly be measured was the experiment with Fiber I with a length of 18 m [Fig. 2(b)]. The experiments with longer fiber lengths (30 m) were limited by other effects for all three fibers under test. From the results presented up to know we can clearly identify the average heat load as an important parameter influencing the MI threshold. The guided mode content is supposed to also have a crucial impact. However, it cannot clearly be clarified from the presented results, if the increased output powers for Fiber II and Fiber III dominantly depend on this parameter. In order to clarify this point an additional experiment was performed with Fiber III with a fiber length of 18 m and a bending diameter of around 1 m. Thus, all experimental conditions in comparison to the first experiment with Fiber I were kept constant allowing for a direct comparison. The results for Fiber III regarding the temporal behavior of the photodiode signal are shown in Fig. 6. Up to an output power of 2.2 kW no significant increase of the standard deviation was observed, indicating that the system was driven below the MI threshold. Further power scaling was limited by fiber destruction. Comparing the results of Fiber I and Fiber III with comparable experimental conditions, we can conclude that the MI threshold was significantly increased from 1 kW (Fiber I) to more than 2.2 kW (Fiber III) due to the reduction of the NA, which improved the single mode guiding properties. This confirms that single-mode guiding properties have a great impact on the MI threshold.
An overview of the essential fiber parameters, the maximum output power, and the limitations for the fibers under test is given in Table 1. For the calculation of modal parameters we solved the scalar Helmholtz wave-equation using a finite difference approximation of the Laplacian, taking into account the refractive index profile of the fiber.
A single mode, narrow linewidth, directly diode pumped fiber amplifier with a maximum output power of 3 kW was demonstrated. No identifications for NLE or MI were observed. The results could be achieved by several iterative optimization steps of the fiber design based on systematic fiber characterization. The fiber design is based on a step index profile realized by MCVD-technology and getting along without complex nano- or micro-structuring techniques.
Table 2 gives an overview of the actions that were taken to realize a fiber design free from NLE and MI up to an output power of 3 kW. In comparison to the commercial fiber used in , where a narrow linewidth, single mode output power of 2.3 kW was demonstrated, two actions were taken: the core diameter was increased and the ytterbium concentration was reduced. As listed in Table 2, an increased core diameter results in increased effective mode areas, which reduce the intensity within the fiber core and thus increases the nonlinear threshold. The MI threshold could be increased by reducing the ytterbium concentration within the fiber core (second action). On the one hand the reduction of the ytterbium concentration was used to reduce the core NA, to realize single mode guiding. The influence of the mode content on the MI threshold was investigated in different experiments. Firstly, it was demonstrated that the MI threshold can be increased for improving the single-mode guiding properties. Fiber I was found to be a few-mode fiber. By reducing the core NA Fiber III was realized, which was proven to be a single mode fiber with significantly high MI threshold for comparable experimental conditions. Secondly, it was demonstrated, that the bending diameter impacts the MI threshold. Higher order modes can successfully be suppressed by reducing the fiber bending diameter leading to an increased MI threshold. Further consequences of reduced ytterbium concentrations are the increase of the pump power absorption length and the reduction of photodarkening losses, which both reduce the average heat load. The MI threshold could significantly be increased by increasing the fiber length of Fiber I from 18 m (inefficient pump power absorption) to 30 m (efficient pump power absorption) under comparable experimental conditions.
The S2-measurement method as a fiber characterization tool at low power operation was shown to be suitable to predict the trend of the MI threshold for specific fibers, for the first time.
To the best of our knowledge, our results present a narrow linewidth (170 pm), single mode directly diode pumped fiber amplifier with highest output power ever reported on.
We thank our colleagues Dr. Jens Kobelke and Dr. Jörg Bierlich from the Leibniz IPHT Jena for drawing our preforms to fibers.
References and links
1. C. Wirth, O. Schmidt, I. Tsybin, T. Schreiber, R. Eberhardt, J. Limpert, A. Tünnermann, K. Ludewigt, M. Gowin, E. ten Have, and M. Jung, “High average power spectral beam combining of four fiber amplifiers to 8.2 kW,” Opt. Lett. 36(16), 3118–3120 (2011). [CrossRef] [PubMed]
2. S. McNaught, C. Asman, H. Injeyan, A. Jankevics, A. Johnson, G. Jones, H. Komine, J. Machan, J. Marmo, M. McClellan, R. Simpson, J. Sollee, M. Valley, M. Weber, and S. Weiss, “100-kW Coherently Combined Nd:YAG MOPA Laser Array,” OSA Technical Digest Series (CD) (2009), paper FThD2. [CrossRef]
3. Y. Xiao, F. Brunet, M. Kanskar, M. Faucher, A. Wetter, and N. Holehouse, “1-kilowatt CW all-fiber laser oscillator pumped with wavelength-beam-combined diode stacks,” Opt. Express 20(3), 3296–3301 (2012). [CrossRef] [PubMed]
4. C. Jauregui, J. Limpert, and A. Tünnermann, “High-power fibre lasers,” Nat. Photonics 7(11), 861–867 (2013). [CrossRef]
5. M. N. Zervas and C. A. Codemard, “High Power Fiber Lasers: A Review,” IEEE J. Sel. Top. Quantum Electron . 20(5), 219–241 (2014).
6. T. Eidam, C. Wirth, C. Jauregui, F. Stutzki, F. Jansen, H.-J. Otto, O. Schmidt, T. Schreiber, J. Limpert, and A. Tünnermann, “Experimental observations of the threshold-like onset of mode instabilities in high power fiber amplifiers,” Opt. Express 19(14), 13218–13224 (2011). [CrossRef] [PubMed]
7. C. Jauregui, H.-J. Otto, F. Stutzki, J. Limpert, and A. Tünnermann, “Simplified modelling the mode instability threshold of high power fiber amplifiers in the presence of photodarkening,” Opt. Express 23(16), 20203–20218 (2015). [CrossRef] [PubMed]
8. L. Huang, W. Wang, J. Leng, S. Guo, X. Xu, and X. A. Cheng, “Experimental investigation on evolution of the beam quality in a 2 kW high power fiber amplifier,” IEEE Photonics Technol. Lett. 26(1), 33–36 (2014). [CrossRef]
9. J. Nold, M. Strecker, A. Liem, R. Eberhardt, T. Schreiber, and A. Tünnermann, “Narrow Linewidth Single Mode Fiber Amplifier With 2.3 kW Average Power,” in European Conference on Lasers and Electro-Optics - European Quantum Electronics Conference (2015), paper CJ_11_4.
10. Q. Fang, W. Shi, Y. Qin, X. Meng, and Q. Zhang, “2.5 kW monolithic continuous wave (CW) near diffraction-limited fiber laser at 1080 nm,” Laser Phys. Lett. 11(10), 105102 (2014). [CrossRef]
11. G. Overton, “IPG Photonics offers world’s first 10 kW single-mode production laser,” http://www.laserfocusworld.com/articles/2009/06/ipg-photonics-offers-worlds-first-10-kw-single-mode-production-laser.html, Laser Focus World (Published 06/17/2009), 12/09/2015.
12. M. O’Connor, V. Gapontsev, V. Fomin, M. Abramov, and A. Ferin, “Power Scaling of SM Fiber Lasers toward 10kW,” in Conference on Lasers and Electro-Optics/International Quantum Electronics Conference, OSA Technical Digest (CD) (2009), paper CThA3. [CrossRef]
13. V. Khitrov, J. D. Minelly, R. Tumminelli, V. Petit, E. S. Pooler, “3kW single-mode direct diode-pumped fiber laser,” Proc. SPIE 8961, 89610 (2014).
14. H. Yu, H. Zhang, H. Lv, X. Wang, J. Leng, H. Xiao, S. Guo, P. Zhou, X. Xu, and J. Chen, “3.15 kW direct diode-pumped near diffraction-limited all-fiber-integrated fiber laser,” Appl. Opt. 54(14), 4556–4560 (2015). [CrossRef] [PubMed]
15. H. J. Otto, F. Stutzki, F. Jansen, T. Eidam, C. Jauregui, J. Limpert, and A. Tünnermann, “Temporal dynamics of mode instabilities in high-power fiber lasers and amplifiers,” Opt. Express 20(14), 15710–15722 (2012). [CrossRef] [PubMed]
16. Z. Huang, X. Liang, C. Li, H. Lin, Q. Li, J. Wang, and F. Jing, “Spectral broadening in high-power Yb-doped fiber lasers employing narrow-linewidth multilongitudinal-mode oscillators,” Appl. Opt. 55(2), 297–302 (2016). [CrossRef] [PubMed]
17. N. Haarlammert, B. Sattler, A. Liem, M. Strecker, J. Nold, T. Schreiber, R. Eberhardt, A. Tünnermann, K. Ludewigt, and M. Jung, “Optimizing mode instability in low-NA fibers by passive strategies,” Opt. Lett. 40(10), 2317–2320 (2015). [CrossRef] [PubMed]
19. H. Zimer, M. Kozak, A. Liem, F. Flohrer, F. Doerfel, P. Riedel, S. Linke, R. Horley, F. Ghiringhelli, S. Desmoulins, M. Zervas, J. Kirchhof, S. Unger, S. Jetschke, T. Peschel, T. Schreiber, “Fibers and fiber-optic components for high-power fiber lasers,” Proc. SPIE 7914, 791414 (2011).
21. D. J. DiGiovanni, J. B. MacChesney, and T. Y. Kometani, “Structure and properties of silica containing aluminum and phosphorus near the AlPO4 join,” J. Non-Cryst. Solids 113(1), 58–64 (1989). [CrossRef]
22. O. Arnould and F. Hild, “EPMA Measurements of Diffusion Proles at the Submicrometre Scale,” Mikrochim. Acta 139(1-4), 3–10 (2002). [CrossRef]
23. J. W. Nicholson, A. D. Yablon, S. Ramachandran, and S. Ghalmi, “Spatially and spectrally resolved imaging of modal content in large-mode-area fibers,” Opt. Express 16(10), 7233–7243 (2008). [CrossRef] [PubMed]
24. A. Kobyakov, M. Sauer, and D. Chowdhury, “Stimulated Brillouin scattering in optical fibers,” Adv. Opt. Photonics 2(1), 1 (2010). [CrossRef]
25. ISO 11146 “Lasers and laser-related equipment– Test methods for laser beam parameters– Beam width, divergence, angle and beam propagation factor” (2005).