Various rare earth doped single crystal YAG and sesquioxide fibers have been drawn using the laser heated pedestal growth (LHPG) method. Crystalline core/clad fibers have been successfully fabricated using a two-step hydrothermal growth method applied to core fibers. The development of a two-step process is essential to the growth of reasonable cladding layers. Various shapes of the cladding layer were observed, i.e., octagon, hexagon, square and intermediate shapes. EDX study shows a stoichiometry of YAG composition of the clad and there is no significant diffusion of the rare earth ion across the core/clad interface. No birefringence was observed under cross-polarized light indicating that a negligible stress between core/clad crystals was formed during hydrothermal process. Acid etching is an effective method to obtain faceted YAG fibers with a reduced diameter in a controlled manner. An etch rate of ∼0.15 µm/min was measured using 50:50 mixtures of phosphoric acid and sulfuric acid at elevated temperature. Scattering loss of 0.05 dB/cm and a net peak gain of 19 dB was measured from 10% Yb:YAG core/YAG clad fiber. The hydrothermal technique is a versatile epitaxial method to growth undoped YAG cladding of various thicknesses onto doped YAG core fibers and has a great potential to pave the way forward for improving laser performance.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
Since the demonstration of high-power fiber lasers using double clad structures in 1988 , fiber lasers, especially those based on silica glass, have been developed to provide several kW’s of output power and this class of fiber lasers is widely used in commercial applications such as cutting, welding and medical applications [2–4]. Much effort has been made since then to devolep higher power and higher efficiency fiber lasers based on silica glass [5–7]. Unfortunately, the silica fibers have several intrinsic issues that ultimately limit power scaling. These relate to their low thermal conductivity (∼1W/m/K) coupled with low usage temperature, low dopant concentrations (<1%), and thermal effects such as thermal mode instability (TMI) and thermal lensing. In addition, for single frequency operation, the high stimulated Brillouin (SB) cross-section of glass leads to SB scattering which limits the achievable power [8–10].
Recently, there have been research efforts to replace silica glass with more robust and efficient materials such as ceramics and single crystals to overcome the intrinsic issues of silica fibers and to enable higher power fiber lasers. In principle, an all-crystalline fiber is advantageous in many respects compared to glass-based fibers. For example, yttria-alumina garnet (YAG) has ∼100 times lower stimulated Brillouin cross-section, more than 10 times greater thermal conductivity, much higher melting temperatures (>2000 °C), excellent environmental stability, and the propensity for high rare earth doping levels (10%) which can result in shorter fiber lengths, lower nonlinear effects, and the potential for high peak and average laser power output. We have reported a successful fabrication of polycrystalline YAG cladding onto various rare-earth doped (RE:YAG) crystal fibers by the magnetron sputtering method . Kim et al, reported the fabrication of a polycrystalline 1.5% Ho:YAG core fiber cladded with SF57 Schott glass . More recently, a group reported Yb:YAG-core/undoped-YAG-clad crystal fibers grown by the hybrid liquid phase epitaxial (LPE) crystal growth method .
Dawson, et al. modeled single frequency fiber lasers and reported that Yb:YAG fiber lasers can scale to power levels on the order of 16.9 kW, nearly 9 times that of Yb:Silica fiber before nonlinear and thermal issues have a detrimental impact . Based upon an analysis of thermal and nonlinear effects, Ho-doped silica fibers would be limited to ∼3 kW single frequency power for spectral or coherent beam combination at 2.1 µm. By comparison, an all-crystalline Ho:YAG fiber could generate 35 kW of single frequency eye-safer laser power . However, realization of high quality all-crystalline fibers is very challenging due to the high melting temperatures of the crystals and lack of processing protocol to fabricate the required core/clad waveguide structures while maintaining proper refractive indices and diameters of the core and cladding. The ideal scenario would be the continuous drawing of a crystal fiber with the in-situ fabrication of the core/clad structure as we often see from co-drawn glass fibers . However, it is almost impossible to apply the same process designed for glasses to crystalline materials because it requires an apparatus that withstands much higher temperatures. Crystalline materials, unlike glasses, have a very low viscosity upon melting which ensures the core and cladding crystals become immediately mixed together upon melting preventing the formation of sharp core/clad fiber structures. The only option we have left is to apply cladding onto a core fiber as a post-process at lower temperature such as the LPE method utilized to apply crystal cladding onto a crystal core fiber . The feasibility of crystal core/clad fibers has been demonstrated using the LPE process, but reproducibility and fabrication of long length fiber is still unproven with that method.
Here we report a successful growth of a single crystal cladding around a single crystal core using the hydrothermal growth process, a solution growth method that enables deposition of a high quality crystal cladding at temperatures much lower than the melt temperature of the fiber core and lower than that used in the LPE process. Hydrothermal growth of bulk YAG crystals was pioneered in 1969 by Puttbach et al where they reported an optimum condition to control the grow rate of the crystals  and further investigation on the phase equilibrium on hydrothermal growth of YAG was followed by Kolb and Laudise in 1975 . Since then, very little research has been done on the hydrothermal growth of YAG until in 2012 when a group from Clemson University reported the successful growth of thick layers of YAG doped with various transition metals and rare earths at a much faster growth rate of 0.5 mm/side/week on a flat <100 > YAG substrate . They also reported a dramatic reduction in interfacial scattering between layers by applying a very narrow thermal gradient at the early stages of growth. However, their study was limited to flat substrates with a predetermined crystal orientation and no study was made on the hydrothermal growth on round surfaces like fibers where multiple crystallographic orientations are present.
The hydrothermal process of YAG is typically carried out between 300°C to 800°C and at 10–30 kpsi pressure, which is significantly lower than the melting temperature of 1950°C for YAG. The crystal growth is typically carried out in a two-zone high-pressure vessel as shown in Fig. 1. The lower zone, containing a salt solution of the desired crystal constituent elements, is heated to temperatures and pressures above the supercritical point of the solvent to make a fully saturated solution. The temperature in the upper zone of the chamber is held at a slightly lower temperature. The temperature gradient results in super-saturation in the cooler region forcing the material to come out of solution and facilitating crystal growth. The nutrients in the bottom zone provide continuous feed of material. In this work, the rare earth doped single crystal fiber grown from our LHPG process is used as a seed crystal in the crystal growth zone of the hydrothermal vessel. The hydrothermal process has been used to grow various single crystals, including doped and undoped YAG on bulk material but not on fiber until now. Here, we report a successful fabrication of YAG cladding as thick as 110 µm on RE-doped YAG fibers grown by the laser heated pedestal growth (LHPG) technique.
2.1 Growth of crystal core fibers
The LHPG process was used to grow rare earth-(RE) doped single crystal fiber cores from a larger diameter (typically between 100 µm and 1 mm) RE-doped feed crystals. Figures 2(a)-(b) show the state-of-the-art LHPG facility at NRL and a schematic diagram of the system, respectively. In the LHPG apparatus, light from a CO2 laser is first expanded to about 25 mm diameter beam by means of a Galilean beam expander. The beam then enters a reflexicon where it is converted into a 100 mm diameter hollow ring-shaped or donut beam by a precisely-aligned right angle conical mirror and a 45° toroidal mirror. This donut beam is directed by a turning mirror with a hole in it, to an annular parabolic focusing mirror. The 1 mm diameter feed crystal is fed through the turning mirror from below such that the laser is focused directly onto its end forming a bead of melted material in the molten zone. A seed crystal is fed downward from above through the center of the focusing mirror until it impacts the bead. The seed crystal is then withdrawn upwards from the molten zone forming a single crystal fiber while the feed crystal is simultaneously advanced upward into and replenishing the molten zone. The melt and the single crystal fiber are kept in place solely by surface tension, hence, this fiber growth method does not require crucibles unlike the conventional crystal growth techniques such as Czochralski  and Bridgman–Stockbarger methods. [21–22] This type of containerless growth also permits the synthesis of materials with extremely high melting points. The position and diameter of both the feed crystal and the single crystal fiber are continuously monitored as are the shape of the molten zone and any power fluctuations in the CO2 laser. An automated feedback algorithm controls the draw speed and adjusts the laser power as needed to maintain the desired fiber diameter and keep diameter fluctuations to about 1% or better. Figures 2(c)-(d) show photograph images of an actual fiber draw where a fiber diameter is reduced from 100 µm to 17 µm using the state of the art LHPG machine. Various RE doped single crystal YAG fibers have been drawn using the LHPG process in diameters ranging from 17-500 µm in lengths of over 2 m with fiber diameter variations of < 1%. The fiber draw rates are about 1∼2 mm/min. These fibers show excellent mechanical strength and flexibility.
2.2 Fabrication of crystalline cladding by hydrothermal growth
For the hydrothermal growth of an undoped YAG cladding layer onto a RE-doped YAG core fiber, we have used Y2O3 and Al2O3 as feedstock, and K2CO3 as the mineralizer. We find that mineralizer concentrations between 0.25 - 0.50 M K2CO3 are a satisfactory range for fibers with initial diameters of 100 µm. The temperature ranges in the hot (bottom) zone of the autoclave and in the growth zone where the seed fiber crystal is suspended are varied between 550∼600 °C. A suitable baffle is inserted above the feedstock and the entire container welded shut and placed in a high-pressure autoclave. The autoclave is pressurized to an appropriate pressure (ca. 150 MPa) and heated as described above using ceramic resistance band heaters.
2.3 Acid etching of Yb: YAG core fiber
For an acid etching experiment, we have used a method similar to the one reported by Feldman, where a YAG fiber is immersed in an acid bath containing 50:50 mixture of H2SO4 and H3PO4 at 225°C.  The fiber was recovered, rinsed with H2O followed by acetone. The diameter of the fiber was measured after a predetermined time of etching and was used for calculating the fiber etch rate. A typical acid etching was carried out in a beaker containing 50:50 ratio of phosphoric and sulfuric acid. First, fiber is gently immersed in an acid bath and heated up to 225°C, and we measured the change is diameter and shape of the fiber every hour. Note that the fiber is placed in the beaker diagonally, such that only the ends of the fiber touch the beaker.
3. Results and discussion
3.1 LHPG fiber growth
We have used the LHPG technique at NRL to grow various RE-doped YAG single crystal fibers with diameters ranging from 500µm to 17µm. Typically, a commercial single crystal rod of 1 mm x 1 mm x 20 mm was used as a feed rod when available. In cases where a single crystal source was not available, ceramic feed rods were made in-house by hot-pressing [24–26] and cut and used as feed rods for growing fibers by the LHPG technique. Figure 3(a) shows some examples of feed rods used in this study. Various RE-doped single crystal YAG fibers as well as sesquioxide fibers have been drawn in diameters ranging from 17-500 µm in lengths of over 2 m with fiber diameter variations of <1%. Figure 3(b) shows a photograph image of a 250µm diameter fiber along the fiber length where text printed on the paper under the fiber is clearly seen through the fiber indicating a decent quality of the fiber. Figure 3(c) shows a looped fiber with a bend diameter of ∼1 cm indicating good mechanical stability and robustness of the fiber. In Fig. 3(d)-(e), endface images of single crystal fibers of various diameters are shown. It is noted by the red arrows, that the fiber cores are not perfectly round shape and rather curved multigonal shape with dull edges regardless of the diameter of the fibers. It is noted that we used the length of the diagonal direction as a diameter of the fiber unless specified. (See Fig. 3(d)-(e))
3.2 Acid etching of crystal fibers
For a single mode operation in fiber lasers, a thin core diameter is required. In an effort to obtain smaller diameters, we have carried out an etching experiment with a 50:50 mixtures of phosphoric acid and sulfuric acid at 225 °C using 110 µm diameter fiber. Figure 4(a) shows a plot of diameter of the fiber vs. time of exposure to mixed acids. We monitored the change in the dimension of the fiber in vertical, horizontal, and diagonal directions before and after etching in hot acid for a pre-determined exposure time. The crystal fibers seem to be generally etched at a constant rate, although the rate may vary with temperature. A rough calculation shows an etch rate ∼0.15 µm/min which is slower than the rate of ∼0.6 µm/min as reported in reference 23 and we believe the lower etch rated is mainly due to the difference in the actual temperature of the acid bath. Figure 4(b) shows microscope images of endfaces of fibers etched for different times in the acid mixture. A facet formation (red arrows) is clearly observed after 2 hours etching and the diameter of the fiber was reduced. The fiber diameter was further reduced after 4 hours acid etching at 225°C. After prolonged etching, the endface of the fiber became more hexagonal in shape. A sharp edge formation was also observed from the surface along the fiber. Figure 4(c) shows a photographic image of a surface along the fiber length that underwent 4 hour etching where a sharp edge is shown. The acid etching is a very useful method to obtain faceted YAG fibers with a reduced diameter in a controlled manner and also it might be beneficial for realizing high power fiber lasers especially in a double clad structure where the pump light is injected in the inner clad and the purpose of the outer clad is to guide the pump light. In such cases, it is desirable to have an inner cladding breaking the circular symmetry, for example by being off-set or non-round, to overcome whispering gallery modes and increase the absorption of pump light into the fiber core. Since the fiber core and inner cladding are crystalline materials, the thermal gradients inside the core are significantly reduced, thereby enabling higher output powers. Another benefit of acid etching of YAG is an increase in thermal loading capability attributed by the increase in tensile strength of the acid etched fiber . This is beneficial in high-power fiber lasers where high-pump powers lead to high thermally induced stresses that may exceed the tensile strength of RE:YAG fibers. Providing extra strength to such rods is essential for their employment in high-power lasers. A detailed crystallography study on the effect of acid etching on YAG fiber will be published elsewhere.
3.3 Hydrothermal growth of YAG clad
Hydrothermal growth is a very promising technique for applying undoped YAG cladding at a lower temperature. Growth of a crystal on a substrate with well-defined crystal orientation is straightforward since the crystal will grow epitaxially following the crystal orientation of the substrate. However, it is very challenging to grow a single crystal cladding on a round surface, or an approximately round surface such as a crystal fiber. There are three main challenges. First, because the surface of the seed fiber is approximately round, a true single crystal layer cannot be grown since there are essentially an infinite number of nucleation sites on the surface of the fiber. This can lead to an extremely large number of crystals with slightly different orientations growing simultaneously and eventually impinging on each other. This can lead to some spectacular failures of epitaxial growth if not treated properly (Fig. 5(a)-(b)). Secondly, it is often observed that single crystal is growing in a non-symmetrical manner (Fig. 5c) and special care is required to center the core fiber. The last and the biggest challenge is that controlling the growth rate, through proper nutrient concentration, processing temperature and pressure is extremely difficult and sometimes the crystal core fiber is completely dissolved away during the hydrothermal growth process unless the processing conditions are well satisfied. To resolve these critical issues, we have designed a unique two step growth process where the majority of Miller indices as well as the diameter of the round fiber are first reduced in the etch-back step and that is followed by the actual crystal growth on the etched facets. Figure 6(a) shows an example of the fiber underwent the two-step growth process. The diameter of the original fiber was reduced from 0.1 mm to 0.04 mm after the etch-back process. We were able to grow YAG crystal cladding sufficient to yield a 0.45 mm thick diameter fiber upon multiple successive crystal growth process. Figure 6(b) shows YAG cladded fibers of different thickness obtained by this two-step process. Typically, different conditions are employed for the etch-back step and the growth step. The desired temperature difference between the dissolution zone and the deposition zone, the etch-back time, and the nutrient concentration must be identified iteratively to have a well-defined faceted core fiber. The development of a two step-growth process is absolutely essential to the growth of reasonable cladding layers on the fibers.
The rate of etch-back has not been quantitatively identified but is believed to be controllable by controlling the temperature of the crystal growth chamber. In the hydrothermal process all the nutrients are in a saturated state in the bottom dissolving zone as shown in Fig. 1 while the fiber is suspended in the upper growth region. Ideally, the temperature in the growth region could be maintained higher than normal (with a smaller ΔT between the growth and dissolution region). At this stage etch-back of the core fiber will be enhanced without any cladding growth. The time at this condition can be controlled until the fiber surface is etched to reveal a well-defined crystal facet. The temperature in the growth region can then be dropped to promote epitaxial cladding growth. We focused a considerable amount of effort on the two separate steps, namely the etch-back and the subsequent cladding growth step. For the etch-back step, a narrow temperature range was found to be optimal, namely in the vicinity of 575-580˚C. A critical parameter of this step is the thermal differential, which matters at least as much as the actual temperatures. The convection in the tube initiates some dissolve-back that is sufficient to dissolve enough of the high Miller surfaces back to either 111, 110 or 100 depending on the ramp up times. Once the growth temperature equilibrates in the autoclave, transport growth initiates and good quality epitaxial growth can be achieved. Thus, we found that the differential for the etch-back step should range between 2-5˚ in the regime between 575 and 580˚C. The etch-back time can vary depending on the final desired thickness of the core. The thermal profiles are subtly different for the growth step after the dissolve back process. For the growth step, we found that the temperature stays approximately the same but the gradient is increased to 10˚ to 570-580˚C.
We observed various shapes and number of facets from hydrothermally grown YAG cladding. Figure 7(a)-(c) shows some examples of the fibers with different shapes, i.e., octagon, hexagon, and square. Some intermediate shapes were also observed (Figures 7(d)-(f)). They generally show excellent interfacial properties where there is no noticeable stress or birefringence that is often seen from dissimilar materials systems such as crystal core and glass cladding. It is advantageous to use YAG based materials for both core and cladding since all the optical, physical and chemical properties are basically identical. It is noted that the number of facets decreased as the thickness of cladding is increased and we believe it is closely related to the growth rate of different crystal growth directions as suggested by Hartman that the growth form of a single crystal substrate will eventually switch from the fastest growing faces to the slowest and reach an equilibrium state . Initially the crystal grows in directions with the fastest growth rate but eventually would be bound by the lowest energy plane that has the slowest growth rate. A detailed crystallography study on the growth mechanism and facet formation will be discussed in a subsequent communication. Figures 8(a)-(b) show photograph images of the endface of 208 µm thick (vertical and horizontal) Yb:YAG core/YAG clad fiber under cross-polarized light. There was no noticeable birefringence observed which is a good indication that there is minimal stress along the core and clad. It is noted that the shape of clad is close to perfect square. Initial core fiber thickness was ∼100 µm and the diameter was reduced to ∼58µm during the etch-back process. A very thick cladding of ∼118 µm was obtained on a diagonally around a core fiber. The shape of core turned into a rounded octagon with apparent corners after etch-back and the sides of the octagonal core and square cladding are facing almost perpendicular to each other. Energy-dispersive X-ray spectroscopy (EDX) was used to characterize the interfacial properties and composition of the core and clad. Figure 8(c) shows an EDX scan across core/clad indicating no noticeable diffusion within the resolution of the measurement (∼1 µm). Analysis of the cladding reveals a composition close to the desired Y3Al5O12 stoichiometry of undoped YAG phase (Table 1). In addition, there is a sharp gradient in the detection of the concentration of Yb at the core-clad interface. This suggests minimal diffusion of rare-earth ions across the interface occurred during the hydrothermal growth of cladding. Diffusion of the dopant would cause scattering losses and less confinement of light in the core.
3.4 Scattering and gain measurement
The scattering loss was calculated by looking at the attenuation of the brightness of the HeNe light along the fiber. HeNe light was coupled into the cladded YAG fiber (∼210/240 µm of core/clad ratio, 0.076 NA-calculated) through a commercial multimode silica fiber (20/250 µm of core/clad ratio, 0.08 NA). A visible camera equipped with a microscope objective lens was used to zoom into the 1 mm section of the YAG fiber to capture the scattered HeNe light. Figure 9 shows images of light scattered out of the side of the fiber near the fiber ends. Gain, focus and distance of the camera from the outer surface of the YAG fiber was kept constant. The camera was held by a Piezo based computer-controlled translation stage to scan the incremental length of the fiber surface along the entire length of the 3.2 cm long fiber. Total power scattering out from the surface of the segment of the fiber facing the camera was calculated by integrating brightness of the individual pixels in the image. By comparing the net amount of scattered light at the input and output end of the fiber, the scattering loss was calculated to be 0.05 dB/cm.
A 100 µm 10% Yb:YAG core, undoped YAG fiber structure, where cladding was fabricated by hydrothermal crystal growth method, was used for the measurements. (33 mm long and 230µm thick) The index of undoped YAG is lower than that of doped YAG. Based on the index of 10% Yb:YAG and undoped YAG, the numerical aperture (NA) of the fiber core was ∼0.1. Gain was measured using a simple setup comprises of laser diodes, an isolator to prevent feedback and combiner to couple both pump and signal light into the YAG fiber (Fig. 10). DFB laser at 1030 nm was used as a seed and a 970 nm diode was used as a pump source operating at 10% duty cycle to minimize the average power, thereby reducing the heating of the sample. Both pump and seed were launched into the core of the 10% Yb:YAG fiber. The launched seed-signal temporally overlapping the pump was 0.268 mW. A net peak gain of 19 dB was measured from 10% Yb:YAG core/YAG clad fiber. For single crystal fiber structures to be a viable technology, mating the fiber to silica fiber technology is critical. We have demonstrated splicing of unclad 100 µm single crystal YAG fiber to 65 µm core/125 µm clad silica fiber. A multimode splice loss of 0.33 dB was measured with a splice tensile strength of ∼50 kpsi .
Various rare earth doped single crystal YAG and sesquioxide fibers have been drawn using LHPG method. All crystalline core/clad fibers where thermal and optical properties are superior over glass based fibers have been successfully fabricated using a two-step hydrothermal growth method. The development of a two-step process, where majority of miller indices as well as the diameter of the round fiber are first reduced in the etch-back step and that is followed by the actual crystal growth on the etched facets, is essential to the growth of reasonable cladding layers on the fibers. Various shapes of the cladding layer were observed, i.e., octagon, hexagon, cubic and intermediate shapes. EDX study shows a stoichiometry of YAG composition of the clad and there is no noticeable diffusion along the core/clad interface. The microscopy image under cross-polarized light shows no birefringence indicating there is only negligible stress during hydrothermal process. The acid etching is an alternative method to obtain faceted YAG fibers with a reduced diameter in a controlled manner. An etch rate of ∼0.15µm/min was measured using 50:50 mixtures of phosphoric acid and sulfuric acid at 225 °C. Scattering loss of 0.05 dB/cm and a net peak gain of 19 dB was measured from 10% Yb:YAG core/YAG clad fiber. The hydrothermal technique is a versatile epitaxial method to growth undoped YAG cladding of various thicknesses onto doped YAG core fibers. Our present work shows the first demonstration of hydrothermal method to grow single crystal cladding onto a single crystal core fiber. Once fully developed, an all crystalline fiber laser capable of producing high output power from a single fiber will be realized.
1. E. Snitzer, H. Po, F. Hakimi, R. Tumminelli, and B. C. McCollum, OSA Technical Digest Series (Optical Society of America, 1988), paper PD5 https://doi.org/10.1364/OFS.1988.PD5
2. V. Fomin, M. Abramov, and A. Ferin, “10 kW single mode fiber laser,” in 5th International Symposium on High-Power Fiber Lasers and Their Applications, St. Petersburg Russia (2010).
3. L. Quintino, A. Costa, R. Miranda, and D. Yapp, “Welding with high power fiber lasers - A preliminary study,” Mater. Des. 28(4), 1231–1237 (2007). [CrossRef]
4. M. Zervas and C. Codemard, “High Power Fiber Lasers: A Review,” IEEE J. Sel. Top. Quantum Electron. 20(5), 219–241 (2014). [CrossRef]
5. C. C. Baker, E. J. Friebele, A. A. Burdett, D. L. Rhonehouse, J. Fontana, W. Kim, S. R. Bowman, L. B. Shaw, J. Sanghera, J. Zhang, R. Pattnaik, M. Dubinskii, J. Ballato, C. Kucera, A. Vargas, A. Hemming, N. Simakov, and J. Haub, “Nanoparticle doping for high power fiber lasers at eye-safer wavelengths,” Opt. Express 25(12), 13903–13915 (2017). [CrossRef]
6. C. C. Baker, E. J. Friebele, A. A. Burdett, C. G. Askins, M. P. Hunt, B. A. Marcheschi, J. Fontana, J. R. Peele, W. Kim, J. Sanghera, J. Zhang, R. K. Pattnaik, L. D. Merkle, M. Dubinskii, Y. M. Chen, I. A. Dajani, and C. Mart, “Nanoparticle Doping for Improved Er-doped Fiber Lasers,” Fiber Lasers XIV: Technology and Systems, Proceedings of SPIE Volume: 10083 UNSP 100831P (2017).
7. C. C. Baker, E. J. Friebele, A. A. Burdett, D. Rhonehouse, W. Kim, and J. Sanghera, “Rare earth co-doping for increased the efficiency of resonantly pumped Er-fiber lasers,” Opt. Mater. Express 9(3), 1041–1048 (2019). [CrossRef]
8. J. W. Dawson, M. J. Messerly, R. J. Beach, M. Y. Shverdin, E. A. Stappaerts, A. K. Sridharan, P. H. Pax, J. E. Heebner, C. W. Siders, and C. P. J. Barty, “Analysis of the scalability of diffraction-limited fiber lasers and amplifiers to high average power,” Opt. Express 16(17), 13240 (2008). [CrossRef]
9. D. C. Brown and H. J. Hoffman, “Thermal , Stress , and Thermo-Optic Effects in High Average Power Double-Clad Silica Fiber Lasers,” IEEE J. Quantum Electron. 37(2), 207–217 (2001). [CrossRef]
10. C. Jauregui, T. Eidam, H. Otto, F. Stutzki, F. Jansen, J. Limpert, and A. Tünnermann, “Physical origin of mode instabilities in high-power fiber laser systems,” Opt. Express 20(12), 12912 (2012). [CrossRef]
11. J. Myers, W. Kim, B. Shaw, S. Bayya, S. Qadri, D. Rhonehouse, C. Askins, J. Peele, R. Thapa, R. Bekele, C. McClain, and J. Sanghera, “Development of Thin Film Claddings for Single Crystal Optical Fiber,” OSA Advanced Photonics Congress, Novel Optical Materials And Applications, NoTu4D.4, Zurich, Switzerland (July 2-5, 2018) (2018).
12. H. Kim, R. Hay, S. McDaniel, G. Cook, N. Usechak, A. Urbas, K. Shugart, H. Lee, A. Kadhim, D. Brown, B. Griffin, G. Fair, R. Corns, S. Potticary, G. Hopkins, K. Averett, D. Zelmon, T. Parthasarathy, and K. Keller, “Lasing of surface-polished polycrystalline Ho:YAG (yttrium aluminum garnet) fiber,” Opt. Express 25(6), 6725 (2017). [CrossRef]
13. M. Dubinskii, J. Zhang, V. Fromzel, Y. Chen, S. Yin, and C. Luo, “Low-loss ‘crystalline-core/crystalline-clad’ (C4) fibers for highly power scalable high efficiency fiber lasers,” Opt. Express 26(4), 5092–5095 (2018). [CrossRef]
14. J. W. Dawson, M. J. Messerly, J. E. Heebner, P. H. Pax, A. K. Sridharan, A. L. Bullington, R. J. Beach, C. W. Siders, C. P. J. Barty, and M. Dubinskii, “Power scaling analysis of fiber lasers and amplifiers based on non-silica materials,” SPIE768611, 768611 (2010).
15. B. Shaw, S. Bayya, W. Kim, J. Myers, C. Askins, J. Peele, D. Rhonehouse, R. Thapa, S. Bowman, R. Gattass, D. Gibson, and J. Sanghera, “Progress in Development of Clad Single Crystal Fibers,” 18th Annual Directed Energy Symposium, March 7-11, 2016, Albuquerque, NM.
16. W. Kim, V. Nguyen, B. Shaw, L. Busse, C. Florea, D. Gibson, R. Gattass, S. Bayya, F. Kung, G. Chin, and F. Miklos, I. Aggarwal and J. Sanghera, “Recent progress in chalcogenide fiber technology at NRL,” J. Non-Cryst. Solids 431, 8–15 (2016). [CrossRef]
17. R. C. Puttbach, R. R. Monchamp, and J. W. Nielsen, in H. S. Peiser ed., “Hydrothermal Growth of Y3Al5O12 in Crystal Growth,” Pergamon, Oxford, 1967, pp. 569–571
18. E. Kolb and R. Laudise, “Phase equilibria of Y3Al5O12, hydrothermal growth of Gd3Ga5O12 and hydrothermal epitaxy of magnetic garnets,” J. Cryst. Growth 29(1), 29–39 (1975). [CrossRef]
19. C. D. McMillen, M. Manna, J. Fan, L. Zhu, and J. Kolis, “Revisiting the Hydrothermal growthof YAG,” J. Cryst. Growth 356, 58–64 (2012). [CrossRef]
20. J. Czochralski, “A new method for the measurement of the crystallization rate of metals,” Zeitschrift für Physikalische Chemie 92, 219 (1918).
21. P. W. Bridgman, “Certain Physical Properties of Single Crystals of Tungsten, Antimony, Bismuth, Tellurium, Cadmium, Zinc, and Tin,” Proc. Am. Acad. Arts Sci. 60(6), 305–383 (1925). [CrossRef]
22. D. C. Stockbarger, “The Production of Large Single Crystals of Lithium Fluoride,” Rev. Sci. Instrum. 7(3), 133–136 (1936). [CrossRef]
23. R. Feldmana, Y. Shimonya, E. Lebiusha, and Y. Golan, “Effect of hot acid etching on the mechanical strengthof ground YAG laser elements,” J. Phys. Chem. Solids 69(4), 839–846 (2008). [CrossRef]
24. W. Kim, G. Villalobos, C. Baker, J. Frantz, B. Shaw, S. Bayya, B. Sadowski, M. Hunt, B. Rock, I. Aggarwal, and J. Sanghera, “Overview of transparent optical ceramics for high-energy lasers at NRL,” Appl. Opt. 54(31), F210–F221 (2015). [CrossRef]
25. W. Kim, G. Villalobos, C. Baker, J. Frantz, B. Shaw, S. Bayya, B. Sadowski, M. Hunt, I. Aggarwal, and J. Sanghera, “Ceramic windows and gain media for high-energy lasers,” Opt. Eng. 52(2), 021003 (2012). [CrossRef]
26. W. Kim, G. Villalobos, C. Baker, J. Frantz, B. Shaw, S. Bayya, B. Sadowski, M. Hunt, I. Aggarwal, and J. Sanghera, “Laser Based on Highly Doped Lu2O3 Ceramics,” Journal of Directed Energy 5(1), 93–104 (2013).
27. E. Hartmann and E. Beregi, “Face statistics on the dissolution forms of garnet crystals,” J. Cryst. Growth 128(1-4), 74–81 (1993). [CrossRef]
28. R. Thapa, D. Gibson, R. R. Gattass, C. Askins, W. Kim, S. Bayya, L. B. Shaw, and J. S. Sanghera, “Fusion splicing of highly dissimilar YAG crystal fiber and silica fiber with reaction bonding,” Opt. Mater. Express 6(8), 2560–2566 (2016). [CrossRef]