- Email: [email protected]
Problems of YAG nanopowders compaction for laser ceramics
Optical Materials 33 (2011) 702–705
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Problems of YAG nanopowders compaction for laser ceramics S.N. Bagayev a, A.A. Kaminskii b, Yu.L. Kopylov c, V.B. Kravchenko c,⇑ a
Institute of Laser Physics, Russian Academy of Sciences, 630090 Novosibirsk, Russia Institute of Crystallography, Russian Academy of Sciences, 119333 Moscow, Russia c Institute of Radioengineering and Electronics named after V.A. Kotelnikov, Russian Academy of Sciences, Fryazino Branch, 141190 Fryazino, Moscow Region, Russia b
a r t i c l e
i n f o
Article history: Received 13 March 2010 Accepted 3 November 2010 Available online 15 December 2010 Keywords: Transparent ceramics YAG ceramics Nano-powder compaction Slip casting
a b s t r a c t Slip casting and colloidal slip casting at high pressure of yttrium aluminum garnet powders were investigated. It was found that the presence of residual pores in laser oxide ceramics was determined mainly by big size pores in the compact. The size of pore in compact is critical when it is greater than the mean size of initial particles. It was shown that formation of pores’ structure in compact was controlled by appearance of quasi-particles in heavy loaded slurry. Pores concentration is critical for ceramics optical transmittance. Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction One of the most critical stages in transparent oxide ceramics processing is compaction of initial powders. The general goal of compaction is conversion of ensemble of free spherical or quasispherical particles to the structure with one of the known dense packing. Due to permanent aspiration of nano-sized particles for agglomeration the most preferable method for compaction in this case among well known ones is method which gives possibility for self-organization during the packing process. One of these methods is slip casting. Application of this method determined impressing success in laser ceramics processing [1,2]. In spite of rich history of this technology its application for nano-sized powder was not yet investigated in detail. Recently the role of particles’ size [3,4], deflocculant type [5] and slurry rheology [6,7] for compact formation by slip casting was investigated, and general requirements for compaction conditions in transparent ceramics processing are understood better now. But laser application requires ceramics with residual porosity less than 10 3 vol.%, so more detailed investigation of slip casting and related processes seems necessary. 2. Experiment Powders of undoped and Nd doped yttrium aluminum garnet (YAG) for experiments were produced by co-precipitation reverse
⇑ Corresponding author. Address: FIRE RAS, Vvedenskogo squ. 1, Lab. 219, 141190 Fryazino, Moscow Region, Russia. Tel.: +7 903 220 51 18; fax: +7 495 702 95 72. E-mail address: [email protected] (V.B. Kravchenko). 0925-3467/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.optmat.2010.11.009
type method. As source of Y and Nd the high purity Y and Nd nitrates water solutions and as sources of aluminum the water solutions of aluminum nitrate or (NH4)Al(SO4)212H2O (Donetsk’s Chemical Plant production) were used. Y and Nd nitrates were prepared by dissolving of high purity Y (99.95%) and Nd (99.99%) oxides (Giridmet’s Plant manufactured) in high purity HNO3 acid. As precipitant the water solutions of ammonium hydroxycarbonate (reagent grade Panreac) were used. Obtained powders of precursors were transformed to YAG by calcination in the temperature range 900–1300 °C. Powder after calcination was ball milled in water suspension. Shape and size of particles were calculated from SEM and surface specific area (SSA) measurements by BET method. Manufacturing of YAG powders by solvothermal method [8] was also used. The slurry for casting was prepared in ball mill with addition of dispersant–deflocculants. As deflocculants the water solutions of ammonium salts of polyacrilic or polymetacrilic acids with molecular weight in the range 300–16,000 were used. Ammonia water with variable pH index in the range 9.5–11.5 was used as dispersing media. Viscosity of slurry was measured by Brookfield DV-II+ viscosimeter with coaxial cylinders. The size of particles in slurry was measured by laser light scattering method (LLS) with ‘‘22 MicroTec’’ Fritsch’s instrument. Compaction was produced by high pressure (200 MPa) colloidal slip casting method [9–11] and by conventional slip casting. Sintering of samples was made in vacuum furnace with carbon heater at 1500–1750 °C with keeping time at the highest temperature during 2–20 h. Pressure in vacuum furnace during sintering cycle was in the range 10 4– 10 5 Pa. Apparent density of compact and sintered samples was measured by Archimedes’s method. Relative porosity of samples sintered at different temperatures was investigated by SEM, AFM and optical microscopy.
703
S.N. Bagayev et al. / Optical Materials 33 (2011) 702–705
Fig. 1. Nano-sized particles of YAG. 1 and 2 particles produced by co-precipitation method, 1 – hardly and 2 – weekly agglomerated particles, 3 – non-agglomerated particles produced by solvothermal method. Length of bar is 200 nm.
700 600 500
viscosity. cP
reduced by ball milling to the shape and size of initial particles as it is demonstrated in Fig. 1. It is shown in this figure that particles of powders produced by solvothermal method practically are not agglomerated. The value of agglomeration was measured as ratio of mean particles size estimated by LLS to the size of initial particles estimated from SSA and SEM measurements. The average value of this ratio for powders produced by co-precipitation method was 70 and can be reduced (de-agglomerated) to 1–2 by ball milling if co-precipitation and calcinations procedures were optimized. After such de-agglomeration the powder was used for subsequent compaction. Deflocculant was added to suspension for slurry preparation as weight amount in mg per square meter of total surface square of powder. The optimum quantity of deflocculants was estimated by minimum of viscosity in dependence on deflocculant concentration as shown in Fig. 2. Value of pH index of slurry for all experiments was chosen as 10.5 because the minimum of zeta potential was found at this pH value. Results of viscosity measurements in dependence on deflocculant molecular weight and powder loading are presented in Fig. 3. As it is shown in insert of Fig. 3 the powder loading of slurry with viscosity level suitable for slip casting is increased when powder SSA and molecular weight of deflocculant decrease. That means it is preferable to have bigger particle size and smaller molecular weight of deflocculant for heavy loaded slurry. Viscosity of slurry can be considered as measure of interaction between particles. We divide all possible range of particles
SSA - 9,7 m 2/g pH - 10 shear rate 93 s-1
400 27 vol.% of dry
300 24 vol.% of dry
200 100
18 vol.% of dry
0 0.4
0.6
0.8
1.0
1.2
deflocculant concentration. mg/m
1.4 2
Fig. 2. Viscosity of YAG slurry vs deflocculant content with molecular mass 4000.
3. Results and discussion Generally powder particles produced by co-precipitation method are strongly agglomerated. The value of agglomeration estimated by SEM depends on precipitation and calcination conditions. At the optimum conditions it is possible to produce powder in which shape and mean dimension of agglomerates can be
1,2 - МW 16000 1200
2
1-SSA 9.7m /g 2 2-SSA 20 m /g
Viscosity level for slip casting 140
4
2
800
120
3 1
600 400
3,4 МW 300
200
3-SSA 9.7m /g 2 4-SSA 20 m /g
2
viscosity, cP
viscosity, cP
1000
100 80 60 40 20 0
15
20
25
30
35
dry powder fraction, vol.%
40
45
18
20 22
24
26
28
30
32 34
36
38
dry powder fraction, vol.%
Fig. 3. Viscosity of YAG slurry vs powder fraction for different deflocculants and powders. In the right side of the figure the level of viscosity acceptable for slip casting is fixed and acceptable values of slurry loading are determined.
704
S.N. Bagayev et al. / Optical Materials 33 (2011) 702–705
Quasi-particle
3
2
1
Fig. 4. Conditional groups of slurry states. 1 – Free particles, 2 – flocculated particles, 3 – coagulated particles.
interaction and consequently all possible range of slurry viscosities in three conditional groups. These groups are shown in Fig. 4. During casting process when liquid leaving out of slurry the state of slurry is changed from well de-flocculated to coagulated one. Some spatial network is formed in the middle state due to interaction between particles. It was proposed that this spatial network could be unstable. In this case spatial network may be divided to some parts – quasi-particles. The presence of such quasi-particles was confirmed by direct measurements of particle size distributions in the slurry. The powder used for these measurements had mean particle size 80 nm but quasi-particles with size about 500 nm for flocculated and 100 and 400 lm for coagulated states were found in dependence on slurry loading. Results of these measurements are shown in Fig. 5. We proposed according to these results that while liquid is leaving out of the slurry its state is changed from well de-flocculated to first partially flocculated then to completely flocculated and at last to coagulated one. It is found when slurry loading is too high quasi-
particles appear immediately after slurry preparation. This can be explained by liquid evaporation or by insufficient slurry stabilization. In any case some additional investigations of heavy loaded slurry are necessary. Packing of particles in compacts depends on sequence in change of state of slurry. The presence of quasi-particles can be considered as analogue of granulated powder in dry pressing process when there is no strong dependence between packing of initial particles inside quasi-particles and quasi-particles themselves in the compact. That means it is possible to create two kinds of packing pores. Smaller pores appear as result of packing of initial particles inside quasi-particles and bigger pores as result of packing of quasiparticles. It was found that ability to eliminate of pores during sintering process depended on size of packing pores and on ratio between small and big pores. Slurry showing particle size distribution with the presence of both initial particles and quasi-particles was prepared for experiments. The mean size of initial particles for these measurements was 230 nm. After casting and sintering at different temperatures pore size distributions in these ceramic samples were investigated. Results of these measurements for compacts having pores of two types, ‘‘small’’ and ‘‘big’’ ones, as function of ratio pores’ size to particles’ size are shown in Fig. 6. The size of small pores is smaller than the size of initial particles. While the sintering temperature is increased quantity of small pores is decreased and there is a temperature at which these pores will be eliminated. The size of big pores is about one order magnitude greater than initial particles’ size. Quantity of these pores is not decreased really during heating to the temperatures preferable for sintering of YAG ceramics and such pores are not eliminated at normal sintering conditions. Smooth approximation of obtained distribution is
Fig. 5. The particle size distribution in slurry. Mean size of initial particles is 80 nm. 1 – Completely flocculated slurry, 2 – partially flocculated slurry, 3 – coagulated slurry.
S.N. Bagayev et al. / Optical Materials 33 (2011) 702–705
pores owing to deformation of quasi-particles. This method increases additionally the possible thickness of compact as shown in [13,14]. In this work we used this method for compact formation. Samples of YAG:Nd ceramics when compacts were produced by HPCSC method had different transmission spectra depending on residual pores’ concentration (Fig. 7). Sample 2 has 82% transmittance at 1.06 mcm as compared with 84% for single crystal YAG of the same thickness. Great importance of pores concentration for optical transmittance is obvious from this figure.
80 1
number of pores
70
705
2
60 3
50 40 4
30
4. Conclusion
20
It is necessary to produce compact with pores’ size less than size of particles of initial powder to obtain pore – free ceramics. In the case of slip casting process quasi-particles appear in slurry while liquid is leaving out. Packing of quasi-particles can create pores with size bigger than size of initial particles especially when slurry loading is too high. Casting at high pressure increases compact possible thickness and uniformity in comparison with traditional slip casting. Pores concentration is critical for ceramics optical transmittance.
0.1
1
10
ratio of pore's diameter to particle's diameter Fig. 6. Pores’ size distribution in samples sintered 1 h at different temperatures. 1 – 1700 K, 2 – 1720 K, 3 – 1730 K, 4 – 1750 K. Number of pores is given for 10 10 microns area of SEM picture.
90 Acknowledgments
transmittance (%)
2
This work have been supported by Program of Presidium of Russian Academy of Sciences, grants of RFBR 07-02-00057 and 08-02-12143-ofi and ISTC project 3719. The authors thank Dr. L.V. Lyapin for help with measurements of particles’ size distribution in slurries and Dr. O.V. Karban for help with AFM measurements.
60
30
1
References
0
400
600
800
wavelength (nm) Fig. 7. Transmittance of YAG:Nd ceramics samples (1 mm thick) with different quantity of residual pores. 1 – Pores content is about 0.2 vol.%, 2 – pores content is about 0.02 vol.%.
very close to results obtained in [12] for pressed compacts and sintered ceramics on the base of alumina powder. Probably existence of quasi-particles in slurry does not limit production of pore free ceramics themselves if packing of these quasi-particles does not create pores of greater size than the initial particles. High pressure colloidal slip casting (HPCSC) described in [9] is interesting as method to decrease the quantity of big size
[1] H. Yagi, T. Yanagitani, K. Takaichi, K. Ueda, A. Kaminskii, Opt. Mater. 29 (2007) 1258. [2] H. Yagi, 3rd Laser Ceramics Symp. Paris, October 8–10, 2007, p. O-S-8. [3] A. Krell, J. Klimke, J. Am. Ceram. Soc. 80 (6) (2006) 1985. [4] A. Krell, T. Hutzler, J. Klimke, J. Eur. Ceram. Soc. 29 (2009) 207. [5] K. Lu, C. Kessler, J. Mater. Sci. 41 (2006) 5613. [6] K.A. Appiagyei, G.L. Messing, J. Dumm, Ceram. Intern. 34 (2008) 1309. [7] X. Li, Q. Li, Ceram. Intern. 34 (2008) 397. [8] M.A. Korjinsky, Yu.L. Kopylov, II Russian-French Seminar ‘‘Nanotechnology, Energy, Plasma, Lasers 9NEPL-2008’’. Abstracts. Tomsk Polytechnic University, Tomsk, 2008, p. 46. [9] Yu.L. Kopylov, V.B. Kravchenko, S.N. Bagaev, V.V. Shemet, A.A. Komarov, O.V. Karban, A.A. Kaminskii, Opt. Mater. 31 (5) (2009) 707. [10] A.A. Kaminskii, V.B. Kravchenko, Yu.L. Kopylov, S.N. Bagayev, V.V. Shemet, A.A. Komarov, F. Kallmeyer, H.J. Eichler, Phys. Stat. Solidi (a) 204 (2007) 2411. [11] V.B. Kravchenko, Yu.L. Kopylov, S.N. Bagayev, V.V. Shemet, A.A. Komarov, A.A. Kaminskii, 3rd Laser Ceramics Symp. Paris, October 8–10, 2007, p. O-S-9. [12] J. Zheng, J.S. Reed, Am. Ceram. Soc. Bull. 71 (9) (1992) 1410. [13] B.C. Yu, P.M. Biesheuvel, F.F. Lange, J. Am. Ceram. Soc. 85 (6) (2002) 1456. [14] P.M. Biesheuvel, H. Verweij, J. Eur. Ceram. Soc. 20 (2000) 835.
Contents lists available at ScienceDirect
Optical Materials journal homepage: www.elsevier.com/locate/optmat
Problems of YAG nanopowders compaction for laser ceramics S.N. Bagayev a, A.A. Kaminskii b, Yu.L. Kopylov c, V.B. Kravchenko c,⇑ a
Institute of Laser Physics, Russian Academy of Sciences, 630090 Novosibirsk, Russia Institute of Crystallography, Russian Academy of Sciences, 119333 Moscow, Russia c Institute of Radioengineering and Electronics named after V.A. Kotelnikov, Russian Academy of Sciences, Fryazino Branch, 141190 Fryazino, Moscow Region, Russia b
a r t i c l e
i n f o
Article history: Received 13 March 2010 Accepted 3 November 2010 Available online 15 December 2010 Keywords: Transparent ceramics YAG ceramics Nano-powder compaction Slip casting
a b s t r a c t Slip casting and colloidal slip casting at high pressure of yttrium aluminum garnet powders were investigated. It was found that the presence of residual pores in laser oxide ceramics was determined mainly by big size pores in the compact. The size of pore in compact is critical when it is greater than the mean size of initial particles. It was shown that formation of pores’ structure in compact was controlled by appearance of quasi-particles in heavy loaded slurry. Pores concentration is critical for ceramics optical transmittance. Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction One of the most critical stages in transparent oxide ceramics processing is compaction of initial powders. The general goal of compaction is conversion of ensemble of free spherical or quasispherical particles to the structure with one of the known dense packing. Due to permanent aspiration of nano-sized particles for agglomeration the most preferable method for compaction in this case among well known ones is method which gives possibility for self-organization during the packing process. One of these methods is slip casting. Application of this method determined impressing success in laser ceramics processing [1,2]. In spite of rich history of this technology its application for nano-sized powder was not yet investigated in detail. Recently the role of particles’ size [3,4], deflocculant type [5] and slurry rheology [6,7] for compact formation by slip casting was investigated, and general requirements for compaction conditions in transparent ceramics processing are understood better now. But laser application requires ceramics with residual porosity less than 10 3 vol.%, so more detailed investigation of slip casting and related processes seems necessary. 2. Experiment Powders of undoped and Nd doped yttrium aluminum garnet (YAG) for experiments were produced by co-precipitation reverse
⇑ Corresponding author. Address: FIRE RAS, Vvedenskogo squ. 1, Lab. 219, 141190 Fryazino, Moscow Region, Russia. Tel.: +7 903 220 51 18; fax: +7 495 702 95 72. E-mail address: [email protected] (V.B. Kravchenko). 0925-3467/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.optmat.2010.11.009
type method. As source of Y and Nd the high purity Y and Nd nitrates water solutions and as sources of aluminum the water solutions of aluminum nitrate or (NH4)Al(SO4)212H2O (Donetsk’s Chemical Plant production) were used. Y and Nd nitrates were prepared by dissolving of high purity Y (99.95%) and Nd (99.99%) oxides (Giridmet’s Plant manufactured) in high purity HNO3 acid. As precipitant the water solutions of ammonium hydroxycarbonate (reagent grade Panreac) were used. Obtained powders of precursors were transformed to YAG by calcination in the temperature range 900–1300 °C. Powder after calcination was ball milled in water suspension. Shape and size of particles were calculated from SEM and surface specific area (SSA) measurements by BET method. Manufacturing of YAG powders by solvothermal method [8] was also used. The slurry for casting was prepared in ball mill with addition of dispersant–deflocculants. As deflocculants the water solutions of ammonium salts of polyacrilic or polymetacrilic acids with molecular weight in the range 300–16,000 were used. Ammonia water with variable pH index in the range 9.5–11.5 was used as dispersing media. Viscosity of slurry was measured by Brookfield DV-II+ viscosimeter with coaxial cylinders. The size of particles in slurry was measured by laser light scattering method (LLS) with ‘‘22 MicroTec’’ Fritsch’s instrument. Compaction was produced by high pressure (200 MPa) colloidal slip casting method [9–11] and by conventional slip casting. Sintering of samples was made in vacuum furnace with carbon heater at 1500–1750 °C with keeping time at the highest temperature during 2–20 h. Pressure in vacuum furnace during sintering cycle was in the range 10 4– 10 5 Pa. Apparent density of compact and sintered samples was measured by Archimedes’s method. Relative porosity of samples sintered at different temperatures was investigated by SEM, AFM and optical microscopy.
703
S.N. Bagayev et al. / Optical Materials 33 (2011) 702–705
Fig. 1. Nano-sized particles of YAG. 1 and 2 particles produced by co-precipitation method, 1 – hardly and 2 – weekly agglomerated particles, 3 – non-agglomerated particles produced by solvothermal method. Length of bar is 200 nm.
700 600 500
viscosity. cP
reduced by ball milling to the shape and size of initial particles as it is demonstrated in Fig. 1. It is shown in this figure that particles of powders produced by solvothermal method practically are not agglomerated. The value of agglomeration was measured as ratio of mean particles size estimated by LLS to the size of initial particles estimated from SSA and SEM measurements. The average value of this ratio for powders produced by co-precipitation method was 70 and can be reduced (de-agglomerated) to 1–2 by ball milling if co-precipitation and calcinations procedures were optimized. After such de-agglomeration the powder was used for subsequent compaction. Deflocculant was added to suspension for slurry preparation as weight amount in mg per square meter of total surface square of powder. The optimum quantity of deflocculants was estimated by minimum of viscosity in dependence on deflocculant concentration as shown in Fig. 2. Value of pH index of slurry for all experiments was chosen as 10.5 because the minimum of zeta potential was found at this pH value. Results of viscosity measurements in dependence on deflocculant molecular weight and powder loading are presented in Fig. 3. As it is shown in insert of Fig. 3 the powder loading of slurry with viscosity level suitable for slip casting is increased when powder SSA and molecular weight of deflocculant decrease. That means it is preferable to have bigger particle size and smaller molecular weight of deflocculant for heavy loaded slurry. Viscosity of slurry can be considered as measure of interaction between particles. We divide all possible range of particles
SSA - 9,7 m 2/g pH - 10 shear rate 93 s-1
400 27 vol.% of dry
300 24 vol.% of dry
200 100
18 vol.% of dry
0 0.4
0.6
0.8
1.0
1.2
deflocculant concentration. mg/m
1.4 2
Fig. 2. Viscosity of YAG slurry vs deflocculant content with molecular mass 4000.
3. Results and discussion Generally powder particles produced by co-precipitation method are strongly agglomerated. The value of agglomeration estimated by SEM depends on precipitation and calcination conditions. At the optimum conditions it is possible to produce powder in which shape and mean dimension of agglomerates can be
1,2 - МW 16000 1200
2
1-SSA 9.7m /g 2 2-SSA 20 m /g
Viscosity level for slip casting 140
4
2
800
120
3 1
600 400
3,4 МW 300
200
3-SSA 9.7m /g 2 4-SSA 20 m /g
2
viscosity, cP
viscosity, cP
1000
100 80 60 40 20 0
15
20
25
30
35
dry powder fraction, vol.%
40
45
18
20 22
24
26
28
30
32 34
36
38
dry powder fraction, vol.%
Fig. 3. Viscosity of YAG slurry vs powder fraction for different deflocculants and powders. In the right side of the figure the level of viscosity acceptable for slip casting is fixed and acceptable values of slurry loading are determined.
704
S.N. Bagayev et al. / Optical Materials 33 (2011) 702–705
Quasi-particle
3
2
1
Fig. 4. Conditional groups of slurry states. 1 – Free particles, 2 – flocculated particles, 3 – coagulated particles.
interaction and consequently all possible range of slurry viscosities in three conditional groups. These groups are shown in Fig. 4. During casting process when liquid leaving out of slurry the state of slurry is changed from well de-flocculated to coagulated one. Some spatial network is formed in the middle state due to interaction between particles. It was proposed that this spatial network could be unstable. In this case spatial network may be divided to some parts – quasi-particles. The presence of such quasi-particles was confirmed by direct measurements of particle size distributions in the slurry. The powder used for these measurements had mean particle size 80 nm but quasi-particles with size about 500 nm for flocculated and 100 and 400 lm for coagulated states were found in dependence on slurry loading. Results of these measurements are shown in Fig. 5. We proposed according to these results that while liquid is leaving out of the slurry its state is changed from well de-flocculated to first partially flocculated then to completely flocculated and at last to coagulated one. It is found when slurry loading is too high quasi-
particles appear immediately after slurry preparation. This can be explained by liquid evaporation or by insufficient slurry stabilization. In any case some additional investigations of heavy loaded slurry are necessary. Packing of particles in compacts depends on sequence in change of state of slurry. The presence of quasi-particles can be considered as analogue of granulated powder in dry pressing process when there is no strong dependence between packing of initial particles inside quasi-particles and quasi-particles themselves in the compact. That means it is possible to create two kinds of packing pores. Smaller pores appear as result of packing of initial particles inside quasi-particles and bigger pores as result of packing of quasiparticles. It was found that ability to eliminate of pores during sintering process depended on size of packing pores and on ratio between small and big pores. Slurry showing particle size distribution with the presence of both initial particles and quasi-particles was prepared for experiments. The mean size of initial particles for these measurements was 230 nm. After casting and sintering at different temperatures pore size distributions in these ceramic samples were investigated. Results of these measurements for compacts having pores of two types, ‘‘small’’ and ‘‘big’’ ones, as function of ratio pores’ size to particles’ size are shown in Fig. 6. The size of small pores is smaller than the size of initial particles. While the sintering temperature is increased quantity of small pores is decreased and there is a temperature at which these pores will be eliminated. The size of big pores is about one order magnitude greater than initial particles’ size. Quantity of these pores is not decreased really during heating to the temperatures preferable for sintering of YAG ceramics and such pores are not eliminated at normal sintering conditions. Smooth approximation of obtained distribution is
Fig. 5. The particle size distribution in slurry. Mean size of initial particles is 80 nm. 1 – Completely flocculated slurry, 2 – partially flocculated slurry, 3 – coagulated slurry.
S.N. Bagayev et al. / Optical Materials 33 (2011) 702–705
pores owing to deformation of quasi-particles. This method increases additionally the possible thickness of compact as shown in [13,14]. In this work we used this method for compact formation. Samples of YAG:Nd ceramics when compacts were produced by HPCSC method had different transmission spectra depending on residual pores’ concentration (Fig. 7). Sample 2 has 82% transmittance at 1.06 mcm as compared with 84% for single crystal YAG of the same thickness. Great importance of pores concentration for optical transmittance is obvious from this figure.
80 1
number of pores
70
705
2
60 3
50 40 4
30
4. Conclusion
20
It is necessary to produce compact with pores’ size less than size of particles of initial powder to obtain pore – free ceramics. In the case of slip casting process quasi-particles appear in slurry while liquid is leaving out. Packing of quasi-particles can create pores with size bigger than size of initial particles especially when slurry loading is too high. Casting at high pressure increases compact possible thickness and uniformity in comparison with traditional slip casting. Pores concentration is critical for ceramics optical transmittance.
0.1
1
10
ratio of pore's diameter to particle's diameter Fig. 6. Pores’ size distribution in samples sintered 1 h at different temperatures. 1 – 1700 K, 2 – 1720 K, 3 – 1730 K, 4 – 1750 K. Number of pores is given for 10 10 microns area of SEM picture.
90 Acknowledgments
transmittance (%)
2
This work have been supported by Program of Presidium of Russian Academy of Sciences, grants of RFBR 07-02-00057 and 08-02-12143-ofi and ISTC project 3719. The authors thank Dr. L.V. Lyapin for help with measurements of particles’ size distribution in slurries and Dr. O.V. Karban for help with AFM measurements.
60
30
1
References
0
400
600
800
wavelength (nm) Fig. 7. Transmittance of YAG:Nd ceramics samples (1 mm thick) with different quantity of residual pores. 1 – Pores content is about 0.2 vol.%, 2 – pores content is about 0.02 vol.%.
very close to results obtained in [12] for pressed compacts and sintered ceramics on the base of alumina powder. Probably existence of quasi-particles in slurry does not limit production of pore free ceramics themselves if packing of these quasi-particles does not create pores of greater size than the initial particles. High pressure colloidal slip casting (HPCSC) described in [9] is interesting as method to decrease the quantity of big size
[1] H. Yagi, T. Yanagitani, K. Takaichi, K. Ueda, A. Kaminskii, Opt. Mater. 29 (2007) 1258. [2] H. Yagi, 3rd Laser Ceramics Symp. Paris, October 8–10, 2007, p. O-S-8. [3] A. Krell, J. Klimke, J. Am. Ceram. Soc. 80 (6) (2006) 1985. [4] A. Krell, T. Hutzler, J. Klimke, J. Eur. Ceram. Soc. 29 (2009) 207. [5] K. Lu, C. Kessler, J. Mater. Sci. 41 (2006) 5613. [6] K.A. Appiagyei, G.L. Messing, J. Dumm, Ceram. Intern. 34 (2008) 1309. [7] X. Li, Q. Li, Ceram. Intern. 34 (2008) 397. [8] M.A. Korjinsky, Yu.L. Kopylov, II Russian-French Seminar ‘‘Nanotechnology, Energy, Plasma, Lasers 9NEPL-2008’’. Abstracts. Tomsk Polytechnic University, Tomsk, 2008, p. 46. [9] Yu.L. Kopylov, V.B. Kravchenko, S.N. Bagaev, V.V. Shemet, A.A. Komarov, O.V. Karban, A.A. Kaminskii, Opt. Mater. 31 (5) (2009) 707. [10] A.A. Kaminskii, V.B. Kravchenko, Yu.L. Kopylov, S.N. Bagayev, V.V. Shemet, A.A. Komarov, F. Kallmeyer, H.J. Eichler, Phys. Stat. Solidi (a) 204 (2007) 2411. [11] V.B. Kravchenko, Yu.L. Kopylov, S.N. Bagayev, V.V. Shemet, A.A. Komarov, A.A. Kaminskii, 3rd Laser Ceramics Symp. Paris, October 8–10, 2007, p. O-S-9. [12] J. Zheng, J.S. Reed, Am. Ceram. Soc. Bull. 71 (9) (1992) 1410. [13] B.C. Yu, P.M. Biesheuvel, F.F. Lange, J. Am. Ceram. Soc. 85 (6) (2002) 1456. [14] P.M. Biesheuvel, H. Verweij, J. Eur. Ceram. Soc. 20 (2000) 835.