Open Access
Add to BibSonomyAbstract
This paper demonstrates that we can further eliminate the remaining micro pores by hot isostatic press (HIP) approach. The porosity of Nd:LuAG laser ceramics sharply declined from 7 × 10−6 to 5.60 × 10−7 after additional HIP. Annealing is one of the key steps to optimize the quality of vacuum sintered ceramics, such as make the transmittance rise from 79.98% to 81.02% at 1200nm. However, the trend is in the opposite direction for the HIPed Nd:LuAG: fallen from 82.65% to 25.50% at 1200nm. The microstructure of Nd:LuAG studied by microscope, SEM and TEM shows that a large number of pores and a few second phases like Lu2SiO5 appeared along grain boundaries in post-annealed HIPed Nd:LuAG. The gas composition was confirmed as argon by GCMS analysis. By further optimizing the annealing process of HIPed ceramics we found that vacuum annealed at 1600 °C for 10h and oxygen re-annealed at 1200 °C for 30h can reduce pores and make the ceramic quality better, the transmittance can reach 83.05% at 1200nm, and the porosity dropped to 3.2 × 10−7.
© 2014 Optical Society of America
1. Introduction
Transparent ceramics have attracted great attentions since Ikesue et al. reported that the fabrication of Nd3+-doped YAG laser ceramics with excellent optical quality and laser output [1]. Nd doped Lutetium aluminum garnet (LuAG), another kind of garnet with the same structure, have obvious advantages as high thermal conductivity (9.6W/m·K), and moderate emission cross sections (~9 × 10−20 cm2). This makes it has potential applications in inertial confinement fusion (ICF) as gain materials [2]. It is critical to eliminate all the remaining porosity in order to get a transparent ceramics with good quality [3,4]. Hot isostatic pressing (HIP) is a very effective method in reaching full densification of ceramics [5,6]. The driving force for densification is ~50 times greater during HIP than during normal sintering [7]. A number of transparent ceramics such as Nd:YAG [8–10], MgAl2O4 [11–13], MgO [14], and Y2O3 [15], are produced by the sinter plus HIP approach. Unfortunately, the quality of the HIPed ceramics sometimes gets worse after annealing [16] and obviously deterioration occurred in HIPed Nd:LuAG ceramics after annealing. However, annealing is one of the key steps in preparing laser ceramics [17]. The oxygen vacancy can be significantly reduced by annealing, which will optimize the quality of ceramics. Therefore it is necessary to study the effect of annealing on the quality of laser ceramics comprehensively.
The objective of this study was to investigate the effect of annealing on the microstructure of HIPed Nd:LuAG laser ceramics. High quality Nd:LuAG laser ceramics were produced using either vacuum sintering or vacuum sintering with additional HIP treatment. The optical properties and changes of grain-boundary were studied to determine how annealing affects the microstructures and quality of Nd:LuAG laser ceramics.
2. Experimental
The procedure of Nd:LuAG transparent ceramics preparation (or production) by solid-state reactive sintering method is shown in Fig. 1. High-purity powders of Lu2O3, Nd2O3 and Al2O3 were used as starting materials, the tetraethyl orthosilicate (TEOS) and Sc2O3 were added as sintering aids. These raw materials were mixed by ball-milling in anhydrous alcohol for 12h, then dried, sieved, dry-pressed under 100 Mpa into Ø40 mm disks and cold-isostatic pressed under 210 MPa.
Fig. 1 Schematic diagram for the procedure of Nd:LuAG transparent ceramics by vacuum sintering and sinter plus HIP methods.
As for the vacuum sintering method, Nd:LuAG green body were sintered at 1780-1820 °C for 15-20 hours under a vacuum of 10−3 Pa. While for sinter plus HIP method, the Nd:LuAG green body were firstly sintered at 1750-1820 °C for only 2-5 hours under a vacuum of 10−3 Pa, and then sintered in 200 MPa hot isostatic pressing under argon atmosphere for 2 h.
After sintering, they were annealed at 1450 °C for 10–20 hours in air to remove any possible oxygen vacancies which formed during the sintering process.
The samples were mirror-polished on both surfaces, and then the optical transmittance and absorption spectra were measured. The microstructure of Nd:LuAG transparent ceramics were studied using transmission microscopy(Olympus), scanning electron microscope (SEM, JEOL, Japan) and transmission electron microscope (TEM, Hitachi, Japan). The phase properties and elemental chemical composition were characterized using Energy Dispersive Spectrometer (EDS, Hitachi, Japan) and X-ray diffraction (XRD, Rigaku Co., Japan).
3. Results and discussion
The XRD patterns of Nd:LuAG specimens which was sintered by HIP(A) and under vacuum(B) are shown in Fig. 2.We can see that the phase structures of all specimens are almost the same despite in different sintering methods. All the diffraction peaks of specimens can be well indexed as the cubic garnet structure of Lu3(ScAl)2Al3O12 (PDF#53-0272) which generated due to Sc2O3 were added as sintering aids, and no other phases are detected.
Fig. 2 X-ray diffraction pattern of transparent Nd:LuAG ceramics which was sintered by HIP(A) and under vacuum(B).
Figure 3 shows photograph of vacuum-sintered and HIP-sintered specimens before and after annealing by traditional annealing methods. It can be noticed that the color of both vacuum-sintered and HIP-sintered specimens before annealing is gray, which may originate from existence of some color centers. After annealing, the vacuum-sintered specimen become colorless and the letters under the ceramics can be seen distinctly. However, the trend is in the opposite direction for the HIP-sintered specimen which becomes almost non-transparent after annealing. The result indicates traditional annealing method is not suitable for HIP-sintered specimens.
Fig. 3 Photograph of vacuum-sintered and HIP-sintered specimens before and after annealing by traditional annealing methods. (the thickness of all the samples is 5mm).
The in-line transmittance can describe this phenomena more objectively, Fig. 4. Shows In-line transmittances spectra of vacuum-sintered and HIP-sintered specimens before and after annealing by traditional annealing methods. For vacuum-sintered specimen, traditional annealing method have a positive impact on it’s optical properties, especially in the visible light region. The transmittance of vacuum-sintered specimen increases from 79.98% to 81.02% at 1200nm, and from 53.30% to 78.39% at 400nm. For HIP-sintered specimen, traditional annealing method have a completely opposite impact on it’s optical properties. The transmittance of HIP-sintered specimen dropped dramatically, falling from 82.65% to 25.48% at 1200nm, and from 43.55% to 23.54% at 400nm.
Fig. 4 In-line transmittances spectra of vacuum-sintered and HIP-sintered specimens before and after annealing by traditional annealing methods. Inset: magnified transmittance spectrum between 950nm to 1200nm. (the thickness of all the samples is 5mm).
We have further studied the microstructure of post-annealed sample treated by HIP in order to get the reason that caused the transmission down. Figure 5(a) shows the transmission microscopy micrographs of the surface of the Nd:LuAG ceramic by sinter plus HIP approach after annealing in air. We can see that the average grain size of the sample is about 15um and large amount of pores presence in the sample, especially around the grain boundary. Visualized spatial distribution of pores can be seen in the Fig. 5(b), almost all the grains were surrounded by pores about several hundred nm.
Fig. 5 Transmission microscopy micrographs of the surface (A) and interior (B) of the Nd:LuAG ceramic sintered using HIP methods after annealing in air.
Figures 6(a) and 6(c) show the SEM micrographs of the fractures surface of vacuum-sintered specimens before and after annealing, respectively. We can see that no grain growth occurs when annealing at 1450 °C and the average grain size of the specimens remains about 15μm. There are no obvious pores and grain-boundary phase in or between the grains, and the image of a typical triple junction of grains (Inset upper right) gives a clearly demonstration. Figures 6(b) and 6(d) show the SEM micrographs of the fractures surface of HIP-sintered specimens before and after annealing, respectively. The average grain size of the specimens also remains about 15 μm after annealing. It is interesting to note that there is no amorphous phase and obvious pores observed at the grain boundary in the sample before annealing. However, a large amount of pores and grain-boundary phase appeared around boundaries. Figure 6(d) (Inset upper right) shows the magnified image of a typical triple junction of grains where a secondary phase has crystallized. Pores appeared along the grain-boundary can be seen in magnified boundary image Fig. 6(d) (Inset, lower right) .
Fig. 6 SEM micrographs of the fractured surface of the Nd:LuAG ceramics. (A) Vacuum-sintered specimen before annealing (B) HIP-sintered specimen before annealing (C) Vacuum-sintered specimen after annealing. Image of a typical triple junction of grains (Inset upper right) (D) HIP-sintered specimen after annealing, Image of a typical triple junction of grains where a secondary phase has crystallized (Inset upper right). Magnified image (Inset, lower right) shows pores appear along the grain-boundary.
Since one major goal of this work is to study the microstructure near grain boundaries and the constituent of grain-boundary phase, we analyzed the HIPed sample by TEM in depth. Figure 7(a) shows a high-resolution TEM image of grain boundary in a HIPed sample before annealing. We can see that the width of grain boundary is less than 1nm and there is no amorphous phase observed at the grain boundary. Electron diffraction pattern (Inset, upper left) showed the lattice plane of the grain is <011> and confirmed that the crystal structure was LuAG.
Fig. 7 Characterization of the Nd:LuAG ceramic sintered by sinter plus HIP approach. (A) High-resolution TEM image. Electron diffraction pattern (Inset, upper left) of Nd:LuAG sintered by HIP before annealing. (B) TEM image of the typical triple junction of grains where a secondary phase has crystallized. Electron diffraction pattern (Inset, upper left) of secondary phase (a).
Figure 7(b) shows a TEM micrograph of grains in a HIPed sample after annealing, a typical triple junction of grains where a second phase has crystallized can be seen in the iamge, the sizes of the second phase(a) is about 150nm, which shows large amount of second phase appears after annealing.
Compositional analysis of the zones (a) and (b) was carried out using energy dispersive spectroscopic (EDS) analysis, and the results are reported in Table 1.We speculate zone (a) made of Lu2SiO5 or Lu2Si2O7. According to the result of EDS analysis and electron diffraction pattern of secondary phase (Fig. 7(b), inset, upper left), we calculate the inter planar spacing, compare with PDF-#35-0326(Lu2Si2O7) and PDF-#-41-0239 (Lu2SiO5), and confirmed that the crystal structure was Lu2SiO5 which may originate from the enrichment of silicon around the grain boundary.
Table 1. Composition of different areas which shown in Fig. 7(B).
We also studied the component in pores by Gas Chromatography-mass Spectrometer (GCMS) which is shown Fig. 8. From it we can see that argon was detected in the HIPed-sample after heat at 200°C in air, The test specimen used here is a 1-mm cube which was cut from the center of HIPed-Nd:LuAG ceramic before annealing. which means Ar got into Nd:LuAG ceramic during the process of HIP,It can be explained that argon was dissolved in liquid-grain boundary phase(of silicate composition) by HIP treatment,and generated bubbles by releasing the pressure through annealing under normal pressure, in consideration of pore growth occurred along the grain boundary(of silicate composition) in the sample that were treated by HIP and annealed.
Ikesue showed that if the grain-boundary phases contained no argon, these phase could easily dissolve within the lattice, however, after argon is dissolved in the grain-boundary phase through the effect of high pressure during HIP, it was difficult to dissolve the grain-boundary-phase into the inner grain [16], and Kuklja predicted that the most likely defect mechanism [18] for Si4+(0.026nm) solubility in lattice is
To give off the Ar that got into Nd:LuAG ceramic during the process of HIP, vacuum annealing is added to anneal process, and then oxygen re-annealing is used to remove color centers and oxygen vacancy. For the samples, which were HIPed, annealed in vacuum, argon would exsolve from the grain-boundary-phase, so the grain-boundary phases could easily dissolve within the lattice, so the additional Lu2SiO5 phase didn’t appeared in the samples. The in-line transmittances of HIP-sintered specimens as-prepared, vacuum annealed at 1600 °C for 10h, vacuum annealed at 1600 °C for 10h and oxygen re-annealed at 1200 °C for 30h are shown in Fig. 9, and the pressure should under 1 × 10−3 Pa. The results of the present work indicate that new annealing method is an effective process to improve the optical properties of HIP-sintered Nd:LuAG laser ceramics. After vacuum annealing, the transmittance of HIP-sintered specimen increase from 82.65% to 82.85% at 1200nm, and from 43.55% to 79.01% at 400nm. After oxygen re-annealing, the transmittance of HIP-sintered specimen can achieve 83.05% and 80.62% at 1200nm and 400nm, respectively.
Fig. 9 In-line transmittances spectra of HIP-sintered specimens as-prepared, vacuum annealed at 1600 °C, vacuum annealed at 1600 °C and oxygen re-annealed at 1200 °C . Inset: magnified transmittance spectrum between 950nm to 1200nm.
4. Conclusions
High quality Nd:LuAG laser ceramics were produced using either vacuum sintering or vacuum sintering with additional HIP treatment. The differences in annealing effects on both vacuum and HIPed Nd:LuAG laser ceramics were studied. Electron diffraction pattern and EDS analysis at the TEM scale show that the secondary phase is made of Lu2SiO5. It also demonstrates that composition of the pore in the HIPed Nd:LuAG laser ceramics after annealing is argon, which get into during the process of HIP. The presence of second phase and pores is the reason why annealing makes the quality of HIP-sintered Nd:LuAG laser ceramics get worse. By vacuum annealed at 1600 °C for 10h and oxygen re-annealed at 1200°C for 30h we can reduce pores and make the ceramic quality better .
Acknowledgment
The work is financially supported by the National Nature Science Founds of China (No. 61378069, 51102257, 51302284), Shanghai City Star Program (No.14QB1400900, 14QB1402100).
References and links
1. A. Ikesue, I. Furusato, and K. Kamata, “Fabrication of polycrystalline, transparent YAG ceramics by a solid state reaction method,” J. Am. Ceram. Soc. 78(1), 225–228 (1995). [CrossRef]
2. H. L. Li, X. J. Liu, and L. P. Huang, “Fabrication of transparent cerium-doped lutetium aluminum garnet (LuAG:Ce) ceramics by a solid-state reaction method,” J. Am. Ceram. Soc. 88(11), 3226–3228 (2005). [CrossRef]
3. A. Ikesue, K. Yoshida, T. Yamamoto, and I. Yamaga, “Optical scattering centers in polycrystalline Nd:YAG laser,” J. Am. Ceram. Soc. 80(6), 1517–1522 (1997). [CrossRef]
4. Q. Yi, T. Tsuboi, S. M. Zhou, Y. Nakai, H. Lin, and H. Teng, “Investigation of emission properties of Tm3+:Y2O3 transparent ceramic,” Chin. Opt. Lett. 10(9), 091602 (2012). [CrossRef]
5. S. F. Wang, J. Zhang, D. W. Luo, F. Gu, D. Y. Tang, Z. L. Dong, G. E. B. Tan, W. X. Que, T. S. Zhang, S. Lie, and L. B. Kong, “Transparent ceramics: Processing, materials and applications,” Prog. Solid State Chem. 41(1–2), 20–54 (2013). [CrossRef]
6. A. J. Stevenson, X. Li, M. A. Martinez, J. M. Anderson, D. L. Suchy, E. R. Kupp, E. C. Dickey, K. T. Mueller, and G. L. Messing, “Effect of SiO2 on densification and microstructure development in Nd:YAG transparent ceramics,” J. Am. Ceram. Soc. 94(5), 1380–1387 (2011). [CrossRef]
7. S. H. Lee, E. R. Kupp, A. J. Stevenson, J. M. Anderson, G. L. Messing, X. Li, E. C. Dickey, J. Q. Dumm, V. K. Simonaitis-Castillo, and G. J. Quarles, “Hot isostatic pressing of transparent Nd:YAG ceramics,” J. Am. Ceram. Soc. 92(7), 1456–1463 (2009). [CrossRef]
8. A. Ikesue and K. Kamata, “Microstructure and optical properties of hot isostatically pressed Nd:YAG ceramics,” J. Am. Ceram. Soc. 79(7), 1927–1933 (1996). [CrossRef]
9. H. D. Lee, T. I. Mah, and T. A. Parthasarathy, “Low-cost processing of fine grained transparent yttrium aluminum garnet,” Ceram. Eng. Sci. Proc. 25(3), 147–152 (2004). [CrossRef]
10. Y. Rabinovitch, C. Bogicevic, F. Karolak, D. Tetard, and H. Dammak, “Freeze-dried nanometric neodymium-doped YAG powders for transparent ceramics,” j. mater. proc. tech. 199(1–3), 314–320 (2008). [CrossRef]
11. G. Gilde, P. Patel, P. Patterson, D. Blodgett, D. Duncan, and D. Hahn, “Evaluation of hot pressing and hot isostatic pressing parameters on the optical properties of spinel,” J. Am. Ceram. Soc. 88(10), 2747–2751 (2005). [CrossRef]
12. K. Tsukuma, “Transparent MgAl2O4 Spinel ceramics produced by HIP post-sintering,” J. Ceram. Soc. Jpn. 114(1334), 802–806 (2006). [CrossRef]
13. M. Shimada, T. Endo, T. Saito, and T. Sato, “Fabrication of transparent spinel polycrystalline materials,” Mater. Lett. 28(4–6), 413–415 (1996). [CrossRef]
14. K. Itatani, T. Tsujimoto, and A. Kishimoto, “Thermal and optical properties of transparent magnesium oxide ceramics fabricated by post hot-isostatic pressing,” J. Eur. Ceram. Soc. 26(4–5), 639–645 (2006). [CrossRef]
15. A. Ikesue, K. Kamata, and K. Yoshida, “Synthesis of transparent nd-doped HfO2–Y2O3 using HIP,” J. Am. Ceram. Soc. 79(2), 359–364 (1996). [CrossRef]
16. A. Ikesue and K. Kamata, “Microstructure and optical properties of hot isostatically pressed Nd:YAG ceramics,” J. Am. Ceram. Soc. 79(7), 1927–1933 (1996). [CrossRef]
17. W. Zhang, T. C. Lu, N. Wei, B. Y. Ma, F. Li, Z. W. Lu, and J. Q. Qi, “Effect of annealing on the optical properties of Nd:YAG transparent ceramics,” Opt. Mater. 34(4), 685–690 (2012). [CrossRef]
18. M. M. Kuklja, “Defects in yttrium aluminium perovskite and garnetcrystals: atomistic study,” J. Phys. Condens. Matter 12(13), 2953–2967 (2000). [CrossRef]
