Open Access
Add to BibSonomy
Abstract
Transparent ceramic Er:YAG laser rods were fabricated via the direct ink write (DIW) method with engineered doping profiles featuring an Er-doped core with endcaps and core-clad structures. Laser rods up to 11 cm in length were produced which required development of a scalable process. To achieve this, multiple improvements were implemented, including printing the rods horizontally on a substrate, rather than vertically, eliminating the need for an external support structure and using a sacrificial drying layer to mitigate warping and defects. Highly transparent rods were achieved with optical scatter levels as low as 0.5%/cm (at 543 nm). A small refractive index difference of 5.7 ppm was measured at the interface between the Er-doped core and the Lu-doped endcaps and cladding. These results demonstrate DIW as a straightforward method for making good optical quality laser rods with engineered doping profiles to improve laser performance.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Single crystals and glasses have been the primary solid-state laser hosts since the demonstration of the first ruby laser in 1960 [1,2]. However, transparent ceramics can offer similar performance to that of single crystals, as shown by Ikesue et al. in the 1990’s [3]. Polycrystalline ceramics offer advantages over single crystals for laser applications such as higher dopant concentrations, controlled doping profiles, lower processing temperatures, shorter fabrication time, improved fracture toughness, and near-net-shape fabrication [2,4]. In particular, the ability to control the doping profile within a laser gain medium opens the design space to many possibilities for improving laser efficiency, beam quality, amplified spontaneous emission suppression, and thermal management [2,4]. One particular material of interest is Er:YAG, a laser material typically operated at 1.6 or 2.94 µm (considered eye-safe) for infrared lasers with applications in dentistry, medicine, range finding, communication, and laser radar [5–9]. Improvements in performance of Er:YAG gain elements via ceramics fabrication and dopant control may therefore benefit a wide range of applications.
Conventional rod-shaped gain elements and a simplified zig-zag slab geometry (without the face tilts) are pictured in Fig. 1(a) [10]. In Fig. 1(b), we illustrate how thermally induced tension in the entrance/exit faces leads to bowing and fracture at high-power, which can be averted by the introduction of undoped endcaps onto the structure to convert the strain into stress (i.e., compression) [11]. Endcaps, which have previously been incorporated into structures by way of diffusion bonding, modify the nature of heat removal, reduce thermal lensing, and improve output power in YAG lasers [11–15]. The introduction of “side-cladding” can also offer a unique advantage, wherein the gain region is sandwiched between an index-matched, optically clear region parallel to the optical axis of the gain medium, illustrated in Fig. 1(c). With this extra cladding, the extraction can be rendered more efficient by way of higher mode-fill in an amplifier rod or slab since the beam is maintained at a small distance from the side of the gain medium, allowing it to tail-off via “gain guiding” [16]. Related core-clad YAG structures have also shown improved lasing characteristics [17].
Fig. 1. (a) Illustration of gain medium geometries of interest to the current work; (b) (Top) Standard rod experiencing bowing at the faces due to thermal gradient. (Bottom) Rod with endcaps to provide compressive force; (c) Proposed utility of the “side-clad” on the gain element in terms of mode-fill and beam propagation.
The purpose of this article is to illustrate the capabilities of additive manufacturing (AM) for ceramic laser gain media in which the structure is tailored to provide enhanced thermal and optical control, thereby improving the mechanical robustness and providing intrinsic mode control. AM methods can be deployed to “write” several green structures at a time, which are subsequently processed into fully consolidated transparent ceramics over several days and then ground and polished to the desired dimensions, providing a less complicated means of producing cladded structures, as well as structures not possible by means of diffusion bonding, such as doping gradients. The demonstrated precision achieved with spatially tailored transparent ceramics herein may inspire their incorporation into lasers in the future.
Direct ink write (DIW) is an AM method that has been demonstrated as an effective means to achieve radial doping profiles, such as “top-hat” or gradient, in cylindrical Nd:YAG laser rods [18,19]. This technique has been deployed to fabricate other YAG transparent ceramics with unique structures and slurry compositions (ceramic tubes by UV assisted ink and aqueous high solids loading inks) [20,21], as well as other types of ceramics for a wide range of applications [22]. In this paper, DIW was employed to fabricate rectangular prismatic Er:YAG laser slabs with endcaps and core-clad structures.
The introduction of Er into the YAG crystal structure increases the refractive index by approximately 2.10 × 10−4/at.% [23]; therefore Lu was added to the non-gain regions to match the refractive index to avoid reflections and beam distortion at the interfaces. Scaleup of the transparent ceramic DIW process enabled the fabrication of longer laser rods, from 6 cm previously to 11 cm in this work. The increase in length was achieved by printing the rods laying down rather than standing up, as was done in our previous work [18,19], eliminating the need for a support structure to keep the rods from falling over during the print process. With this new methodology, the primary limitation for the length is the size of the equipment used for post-processing (burn out, sintering, hot isostatic pressing) the ceramics.
2. Experimental procedure
2.1 Sample preparation
Nanoparticles of 0.5% Er:YAG ((Y0.995Er0.005)3Al5O12), undoped YAG, and undoped LuAG were purchased from Nanocerox, Inc. Two ink compositions were made, one loaded with 0.5% Er:YAG nanoparticles and the other with YAG and LuAG to form 1%Lu:YAG ((Y0.99Lu0.01)3Al5O12). These particles were dispersed into a shear-thinning viscous slurry containing a solvent, dispersant, and binder to create the inks, allowing them to be extruded through a 600 µm nozzle and retain their shape after deposition. The ink composition consisted of 40–45 vol% YAG nanoparticles, 40–45 vol% propylene carbonate for the solvent, 3–4 vol% 2-[2-(2-methoxyethoxy)ethoxy]acetic acid (MEEAA) for the dispersant, and approximately 12 vol% polyethylene glycol (PEG) for the binder. Tetraethyl orthosilicate (TEOS) was also added as a sintering aid to each ink at a concentration of 0.4 wt% (relative to the mass of YAG only). A drop of red food coloring was added to the Er:YAG ink to easily differentiate the inks during printing. To prepare the inks, the constituents were mixed in a Thinky centrifugal mixer at 2000rpm.
The inks were loaded into 10 or 30-mL syringe barrels and mixed in a Thinky centrifugal mixer at 2000rpm for 1 minute to remove air bubbles. The syringe plunger was then inserted, and the syringes were mixed in the Thinky again for 1 minute. Next, the syringes were installed on a DIW setup (Aerotech) as described in our previous work [18]. The laser gain media were printed either on a flat piece of porous PTFE (polytetrafluoroethylene) or glass, each sprayed with a release layer of PTFE to prevent the green body from sticking to the substrate.
The as-printed structures were dried in a drying oven at 45°C in air for 3 days. The resulting green bodies were burned out at 1000°C for 4 hours in air to remove the polymer binder and other residual organic molecules. Next, they were vacuum sintered in a tungsten mesh furnace at 1640°C for 8 hours to reach closed porosity. Finally, the samples were hot isostatically pressed (HIP) between 1750 and 1850°C for 4 hours in 200 MPa of Ar to collapse the remaining porosity and achieve transparency.
2.2 Characterization
Optical scatter was measured through the length of the laser rods using a 1 mW HeNe laser operating at 543 nm and a Labsphere 8” integrating sphere with integral photodiode. An electron microprobe was used to quantify and map the elemental composition across the interface from Er:YAG to Lu:YAG. The calculated change in refractive index using these compositional profiles was averaged across five data points for each point on the line in Fig. 7. A 532 nm interferometer was used to measure the transmitted optical path difference (OPD) across this interface. The OPD was then divided by the sample thickness and the surface tilt contribution subtracted to obtain the change in refractive index from the Lu-doped clad to the Er-doped core. The absorption spectrum was measured in the 0.5% Er:YAG region of a sample from 190 to 1700nm. Thermo Evolution and Shimadzu UV-3600i Plus UV-Vis-NIR spectrophotometers were employed to collect this full range, and the results combined into a single spectrum. Slit sizes were adjusted to correspond with the absorption linewidths, as follows: 190-990 nm, slits = 1 nm; 990-1450 nm, slits = 2 nm; 1450-1550 nm, slits = 0.2 nm and 1550-1700nm, slits = 2 nm. Samples were polished and thermally etched at 1450°C for 30 minutes and viewed with an optical microscope to determine the grain size. A Phenom ProX desktop scanning electron microscope was used to look at the fracture surfaces.
3. Results and discussion
An Er:YAG ink was used for the gain regions of the laser rods while a Lu:YAG ink was used for the non-gain regions. A concentration of 1.0 atomic % Lu was chosen to match the refractive index of the 0.5% Er-doped region, based on data of the relative refractive index increase for each respective ion [23].
Four different doping profiles were printed, as illustrated in Fig. 2(a). The simplest doping profile has an Er-doped center with Lu-doped endcaps. The next two structures are nominally the same, with an Er-doped core, endcaps, and side cladding on two of the four sides as might be employed for a zig-zag slab configuration; however, the core was printed either vertically (perpendicular to the substrate) or horizontally (parallel to the substrate). This influences the final structure and will be discussed later. These structures were printed with a rectangular cross-section of approximately 1 × 1 cm and a length of 14 cm, shrinking during processing by approximately 21% each, resulting in an approximately 11 cm long rod with an 8.8 cm Er-doped core. The final structure contains a roughly cylindrical Er-doped core surrounded by cladding and endcaps, with an overall length of 8 cm as-printed, shrinking to about 6 cm overall with a 5 cm long core after processing. Figure 2(b) shows an example of a short laser rod with a horizontally printed core during printing (colored red with food color, as described in the Experimental section).
Fig. 2. (a) Illustration of doping profiles printed using DIW (the red colored areas represent the Er-doped ink while the white areas represent Lu-doped ink). (b) Printing a rod with a horizontally printed Er-doped core. (c) Two printing patterns, A and B, were utilized. (d) (Top) A single sacrificial drying layer, as was used for printing on glass substrates. (Bottom) A dried green body flipped over, showing the sacrificial drying layer.
Various doping profiles were achieved using the printing patterns illustrated in Fig. 2(c), where the ink filaments were deposited in two different orientations. Samples printed on the porous PTFE substrate were printed with a flat, solid bottom since the solvent could evaporate through the porous substrate. However, when printing on a glass substrate, a sacrificial ladder-like pattern was printed as a base (sacrificial drying layer), shown in Fig. 2(d), and the sample was then printed on top of this layer to allow solvent evaporation through the bottom.
The printing pattern had a major impact on the final doping profile achieved in the finished laser rod. To produce rods with endcaps, “Pattern A” in Fig. 2(c) is used to make the interface smooth and avoid mixing of the inks where the nozzle turns around (as turnarounds typically cause slight over-extrusion due to limited directional acceleration). In addition, voids must be avoided between filaments in the printed green body, as these would result in voids in the final ceramic, causing large defects. Therefore, the extrusion rate is tuned to slight excess during direct ink writing of transparent ceramic green bodies. The core/side-clad structure can be printed either with the Er-doped core parallel (horizontally) or perpendicular (vertically) to the substrate, as pictured in Fig. 2(a), recalling that the side-cladding is only on two of the sides, to accommodate the zig-zag optical geometry. Cross-sections of the cores of these printed structures are shown in Fig. 3. To print the core vertically, “Pattern B” is used. The Lu-doped cladding is printed first, leaving the middle empty, which is then filled in with the Er-doped ink. This pattern is repeated for each layer, resulting in four interfaces between the inks in each layer, along the sides and endcaps. Due to the necessary over-extrusion to completely fill the space between the filaments and error in the tip spacing between the separate ink nozzles, there is some penetration of the Lu-doped ink in between the layers of Er-doped ink, leaving the core inhomogeneous and “striped”. This problem could be mitigated by more precisely determining and programming the nozzle tip spacing between the two ink compositions to avoid overlap when switching from one ink to the other. Also, the extrusion rate should be tuned to the minimum allowable while still completely space filling so excess Lu:YAG ink does not penetrate the Er:YAG core. A more viscous ink composition would likely lessen this issue since there would be less “leveling” of the ink but could result in more defects from voids due to the ink not completely filling the space between filaments. In contrast, printing the core horizontally provides a smooth, homogeneously doped core. In this case, the Lu:YAG ink is first printed along the whole length of the rod for several layers to create one side of the cladding. Then “Pattern A” in Fig. 2(c) is used to print several layers, with the number of layers determining the thickness of the Er:YAG core. Finally, several more layers of pure Lu:YAG ink are printed on top to create the other side of the cladding. This printing geometry eliminates the issue of Lu-doped ink penetrating between the layers of Er-doped ink. Figure 3(a) shows the core printed vertically, in which there is significant mixing of the inks, versus the core printed horizontally in Fig. 3(b), which results in a contiguously doped core.
Fig. 3. Cross-sections of fully consolidated transparent ceramic rods under UV excitation (254 nm) with various doping profiles illustrated in Fig. 2. (a) A vertically printed core, in which significant mixing between the doped and undoped inks is revealed, (b) a horizontally printed core where the core appears as a uniform contiguous section at the interior of the optic, and (c) a nominally cylindrical core.
The cylindrical core rod is also printed using “Pattern B” for the core region after printing several layers of solely Lu:YAG ink for the bottom cladding. An extra filament of Er:YAG ink is added as each subsequent layer is printed until halfway through the core, then the same pattern is repeated in reverse, reducing the core in width by one filament as each new layer is printed. This approximates a cylinder, but due to the resolution limit of approximately 400 µm, produces an effectively hexagonal core. This reveals a disadvantage of printing the rods horizontally along the length rather than vertically, for the purpose of producing a perfectly cylindrical core structure. The individual layers can be discerned for the cylindrical core sample in Fig. 3(c), as well as for the vertically printed core, due to the ink mixing issue, reinforcing the importance of refractive index matching.
Warping is also a concern for large-aspect-ratio green bodies and can occur during post-processing. Samples printed on the porous PTFE substrate consistently warped parallel to the z-axis, such that the bottom of the sample which was in contact with the porous PTFE substrate became convex and the top became concave, as in Fig. 4(a). This warping is likely due to solvent wicking into the porous substrate during drying, resulting in a lower particle packing density on the bottom. This warping provided a motivation for printing the core vertically, despite the irregularity of the core, since the rods always warped parallel to the core material in this orientation. Thus, the top and bottom of the rod can be ground away to yield a rod with a straight core and equal cladding thickness on either side. When the core was printed horizontally on porous PTFE, the warping occurred perpendicular to the core, thus resulting in an undesirable bend that cannot be fixed by grinding. The warping can be reduced by placing the concave side to face downwards during the HIP processing step. This allowed them to flatten against the yttria plate and removed some of the warping.
Fig. 4. (Left) Side view of as-sintered rods, before HIP’ing, printed on a (a) porous PTFE substrate, (b) glass substrate, and (c) sacrificial drying layer on glass substrate. (Right) Portion of bottom of respective rods on the left. Red circles indicate voids.
A solid glass substrate significantly mitigated the warping issue but resulted in the formation of large voids on the bottom of the samples, shown in Fig. 4(b). These defects may be the result of entrapped gas from the evaporation of the ink solvent underneath the sample, thus producing bubbles. Both the warping and the voids were eliminated by using a PTFE-coated glass substrate and including a sacrificial drying layer on the bottom of the sample as pictured in Fig. 2(d). The openness of the drying layer allowed for solvent to evaporate from the bottom of the sample and was subsequently ground away after densification. A sample produced in this manner is pictured in Fig. 4(c).
Optically finished laser rods with the doping profiles illustrated in Fig. 2, except for the horizontally printed core, are shown in Fig. 5. Sample #3 was ground and polished to specifications for the intended laser cavity. The Er-doped region has a faint red appearance under room lighting and luminesces light blue under 254 nm UV light. Higher optical scatter was measured in the Er-doped regions (up to 2–3 times greater) than in the Lu-doped regions for all samples. Significant variability in the amount of scatter was observed among the samples, with samples 1, 2 and 3 achieving scatter levels of 0.5 to 0.6%/cm at 543 nm while samples 4 and 5 had 3%/cm and 6%/cm scatter through the core, respectively. These last two samples were made with a batch of likely somewhat off-stoichiometry Er:YAG nanopowder feedstock that produced small secondary phase precipitates at grain boundaries. In addition to sensitivity to stoichiometry, particle size or morphological differences batch-to-batch can degrade the sinterability to full density and leave small pores. SEM images of fracture surfaces of both regions exhibited cleaner grain boundaries for the Lu:YAG, while the Er:YAG had some small particles stuck to the surfaces of grains, indicating secondary phase. The grain size of the Lu:YAG regions was larger than the Er:YAG regions: 16 µm and 12 µm respectively, which also may indicate grain boundary pinning by secondary phase in the Er:YAG, leading to smaller grain size. Pores were not observed in the SEM images, indicating that the scatter discrepancy is more likely due to secondary phase. Despite these differences in quality, the DIW method has been demonstrated here to achieve scatter levels as low as 0.5%/cm at 543 nm. This scattering loss may be lower at the lasing wavelength of 1.6 µm since the defects are less than 1 µm in size, as seen through optical microscopy. Er:YAG laser ceramics with similar optical scatter levels have been lased with efficiency comparable to single crystal Er:YAG [24].
Fig. 5. Various Er:YAG transparent ceramic laser rod samples. (Left) Laser rods with doping profiles described in Fig. 2, under room lights. (Right) The same rods are shown under 254 nm UV excitation with the Er-doped regions luminescing light blue. While sample #3 has been ground and optically polished to its final form, the others await grind/polish to the final form consistent with the intended application.
An electron microprobe was used to map and quantitatively measure the Lu and Er doping concentrations across the interface of an endcap (Fig. 6). The elemental maps [Figs. 6(c), (d)] show that the interface is smooth. The yttrium concentration is practically homogeneous, as expected. The dopant concentrations were measured across a linear portion of the interface, graphed in Fig. 7. This measurement verifies that the expected dopant concentrations were achieved, with average values of 1.02 at% Lu and 0.50 at% Er in the Lu-doped and Er-doped regions respectively. Both dopants diffused across a 200 µm-wide area at the interface of the two regions, with the bulk of diffusion occurring within the 100 µm in the middle. These concentration profiles were used to calculate the expected refractive index change (green line) from the Lu:YAG region to the Er:YAG region using the equation below.
Fig. 6. (a) Backscatter electron image at Er-Lu interface. The black spots are surface defects such as grain pullouts or surface contamination. The grains are visible with varying contrast due to their random orientations affecting the degree of electron backscatter. Electron probe maps of: (b) Y, (c) Lu, and (d) Er concentrations at the interface. The left half of the region imaged is Lu-doped while the right half is Er-doped. Lighter color indicates higher concentrations. Sample shown in inset of Fig. 7.
Fig. 7. Dopant concentration profiles across the Lu/Er interface in a Lu:YAG/Er:YAG transparent ceramic along with the calculated Δn (averaged across five data points) and the experimental Δn. The measured sample under UV (365 nm) is shown in the inset, with the Er:YAG region luminescing green.
The absorption spectrum of one of the 0.5% Er:YAG samples from 200 to 1700nm is shown in Fig. 8, along with an inset from 1400 to 1700nm for the IR region crucial to laser operation. The absorption cross section at 1532 nm, the typical pump wavelength for 1.6 µm operation, is found to be 2.64 × 10−20 cm2, comparable to values found in the literature [27–29], further confirming that the expected doping concentration of Er was achieved and the material is suitable for laser operation.
Fig. 8. Absorption spectrum of a 0.5% Er:YAG transparent ceramic sample. The absorption cross section is given on the righthand axis.
4. Conclusions
We fabricated transparent Er:YAG ceramic laser gain media with engineered dopant profiles by additive manufacturing via the direct ink writing technique. Laser rods with three different dopant profiles were printed: an Er-doped core with only endcaps, a rectangular prism core with endcaps and side cladding, and a nominally cylindrical core with endcaps and cladding. To produce larger rods up to 11 cm in length, several aspects of the DIW process were improved. Printing the rods with a horizontal orientation allowed for the fabrication of longer rods but introduced new challenges. Ink penetration and mixing between print layers was observed for the prints with side cladding and cylindrical doping profiles but could be mitigated in the side clad profile by changing the orientation of the core during printing. PTFE-coated glass with a sacrificial printed drying layer was the best substrate, as it prevented the green body from sticking, allowed solvent evaporation from the bottom of the green body, and did not induce warping during the sintering step. Optical scatter levels as low as 0.5%/cm at 543 nm were achieved, a value similar to Er:YAG laser ceramics that have been lased with good performance [24]. The interface between the Lu- and Er-doped regions was found to have a smooth diffusion profile across which there was a small refractive index change of 5.7 ppm, which is relatively small and not expected to have a major impact on laser performance. In addition, this difference can be eliminated in the future by increasing the relative amount of Lu to Er. The absorption spectrum shows strong absorption at the intended pumping wavelength of 1532 nm and is consistent with other reported values. Future work will compare the performance of samples with these custom doping profiles with that of a conventional homogeneously doped Er:YAG rod.
Funding
Lawrence Livermore National Laboratory (DE-AC52-07NA27344); Joint Directed Energy Transition Office (DE-JTO) (17-S&A-0579).
Acknowledgments
The authors would like to acknowledge Oscar Herrara for the interferometry measurements, Fredrick Ryerson for the electron probe measurements, Colby McNamee, Eric Strang, and Onyx Optics, Inc. for optical polishing.
This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344 and was sponsored by the Joint Directed Energy Transition Office (DE-JTO) under Award 17-S&A-0579. The release is number is LLNL- JRNL-841438
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. T.H. Maiman, “Stimulated optical radiation in ruby,” Nature 189, 493–494 (1960). [CrossRef]
2. A. Ikesue and Y. L. Aung, “Ceramic laser materials,” Nat. Photonics 2(12), 721–727 (2008). [CrossRef]
3. A. Ikesue, T. Kinoshita, K. Kamata, and K. Yoshida, “Fabrication and optical properties of high-performance polycrystalline Nd: YAG ceramics for solid-state lasers,” J. Am. Ceram. Soc. 78(4), 1033–1040 (1995).
4. J. Li, Y. Pan, Y. Zeng, W. Liu, B. Jiang, and J. Guo, “The history, development, and future prospects for laser ceramics: A review,” Int. J. Refract. Hard Met. 39, 44–52 (2013). [CrossRef]
5. C. Bader and I. Krejci, “Indications and limitations of Er: YAG laser applications in dentistry,” American Journal of Dentistry 19(3), 178 (2006).
6. F. Sgolastra, A. Petrucci, R. Gatto, and A. Monaco, “Efficacy of Er: YAG laser in the treatment of chronic periodontitis: systematic review and meta-analysis,” Lasers Med. Sci. 27(3), 661–673 (2012). [CrossRef]
7. C.B. Zachary, “Modulating the Er: YAG laser,” Lasers in Surgery and Medicine 26(2), 223–226 (2000).
8. X. Yang, D. Shen, T. Zhao, H. Chen, J. Zhou, J. Li, H. Kou, and Y. Pan, “In-band pumped Er: YAG ceramic laser with 11 W of output power at 1645 nm,” Laser Phys. 21(6), 1013–1016 (2011). [CrossRef]
9. R. C. Stoneman, R. Hartman, A. I. Malm, and P. Gatt, “Coherent laser radar using eyesafe YAG laser transmitters, Laser Radar Technology and Applications X,” SPIE167–174 (2005).
10. W. Koechner, Solid-state Laser Engineering (Springer-Verlag, 1988).
11. E. C. Honea, R. J. Beach, S. B. Sutton, J. A. Speth, S. C. Mitchell, J. A. Skidmore, M. A. Emanuel, and S. A. Payne, “115-W Tm: YAG diode-pumped solid-state laser,” IEEE J. Quantum Electron. 33(9), 1592–1600 (1997). [CrossRef]
12. M. Tsunekane, N. Taguchi, T. Kasamatsu, and H. Inaba, “Analytical and experimental studies on the characteristics of composite solid-state laser rods in diode-end-pumped geometry,” IEEE J. Select. Topics Quantum Electron. 3(1), 9–18 (1997). [CrossRef]
13. C. Bibeau, R. Beach, C. Ebbers, M. Emanuel, and J. Skidmore, “CW and Q-switched performance of a diode end-pumped Yb: YAG laser, Advanced Solid State Lasers,” Optical Society of America9 (1997).
14. M. Tsunekane, N. Taguchi, and H. Inaba, “Improvement of thermal effects in a diode-end-pumped, composite Tm: YAG rod with undoped ends,” Appl. Opt. 38(9), 1788–1791 (1999). [CrossRef]
15. M. Tsunekane, N. Taguchi, and H. Inaba, “Reduction of thermal effects in a diode-end-pumped, composite Nd: YAG rod with a sapphire end,” Appl. Opt. 37(15), 3290–3294 (1998). [CrossRef]
16. K. Hasegawa, T. Ichikawa, Y. Takeda, A. Ikesue, H. Ito, and T. Motohiro, “Lasing characteristics of refractive-index-matched composite Y3Al5O12 rods employing transparent ceramics for solar-pumped lasers,” Jpn. J. Appl. Phys. 57(4), 042701 (2018). [CrossRef]
17. A. Sträßer and M. Ostermeyer, “Improving the brightness of side pumped power amplifiers by using core doped ceramic rods,” Opt. Express 14(15), 6687–6693 (2006). [CrossRef]
18. I. K. Jones, Z. M. Seeley, N. J. Cherepy, E. B. Duoss, and S. A. Payne, “Direct ink write fabrication of transparent ceramic gain media,” Opt. Mater. (Amsterdam, Neth.) 75, 19–25 (2018). [CrossRef]
19. Z. Seeley, T. Yee, N. Cherepy, A. Drobshoff, O. Herrera, R. Ryerson, and S. A. Payne, “3D printed transparent ceramic YAG laser rods: Matching the core-clad refractive index,” Opt. Mater. (Amsterdam, Neth.) 107, 110121 (2020). [CrossRef]
20. J. Chen, H. Ji, J. Zhang, S. Wang, and Y. Liu, “Fabrication of YAG ceramic tube by UV-assisted direct ink writing,” Ceram. Int. 48(14), 19703–19708 (2022). [CrossRef]
21. H. Ji, J. Zhao, J. Chen, S. Shimai, H. Chen, G. Zhou, Y. Liu, J. Zhang, S. Wang, and D. Yang, “Direct ink writing of cellulose-plasticized aqueous ceramic slurry for YAG transparent ceramics,” MRS Commun. 12(2), 206–212 (2022). [CrossRef]
22. J. A. Lewis, J. E. Smay, J. Stuecker, and J. Cesarano, “Direct ink writing of three-dimensional ceramic structures,” J. Am. Ceram. Soc. 89(12), 3599–3609 (2006). [CrossRef]
23. X. Mu, H. Meissner, H.-C. Lee, and M. Dubinskii, “True crystalline fibers: double-clad LMA design concept of Tm: YAG-core fiber and its mode simulation,” Fiber Lasers IX: Technology, Systems, and Applications, SPIE, 2012, pp. 572–576.
24. E. R. Kupp, G. L. Messing, J. M. Anderson, V. Gopalan, J. Q. Dumm, C. Kraisinger, N. Ter-Gabrielyan, L. D. Merkle, M. Dubinskii, and V. K. Simonaitis-Castillo, “Co-casting and optical characteristics of transparent segmented composite Er: YAG laser ceramics,” J. Mater. Res. 25(3), 476–483 (2010). [CrossRef]
25. J. T. Rudzik, M. S. Zachary, J. R. Frederick, J. C. Nerine, and A. P. Stephen, “(INVITED)Counter-ion effect on the diffusion behavior of Yb, Lu, and Nd ions in YAG transparent ceramics,” Opt. Mater.: X 13, 100132 (2022). [CrossRef]
26. Y. Kuwano, K. Suda, N. Ishizawa, and T. Yamada, “Crystal growth and properties of (Lu, Y)3Al5O12,” Journal of Crystal Growth 260(1-2), 159–165 (2004).
27. N. Ter-Gabrielyan, V. Fromzel, X. Mu, H. Meissner, and M. Dubinskii, “High efficiency, resonantly diode pumped, double-clad, Er:YAG-core, waveguide laser,” Opt. Express 20(23), 25554–25561 (2012). [CrossRef]
28. S.A. Payne, L. Chase, L.K. Smith, W.L. Kway, and W.F. Krupke, “Infrared cross-section measurements for crystals doped with Er/sup 3+/, Tm/sup 3+/, and Ho/sup 3+,” IEEE J. Quantum Electron. 28(11), 2619–2630 (1992). [CrossRef]
29. F. Cornacchia, M. Alshourbagy, A. Toncelli, M. Tonelli, H. Ogino, A. Yoshikawa, and T. Fukuda, “Growth and spectroscopic properties of Er: YAG crystalline fibers,” J. Cryst. Growth 275(3-4), 534–540 (2005). [CrossRef]
