Key Engineering Materials, Volume 403 : SiAlONs and
Non-oxidesSiAlONs and Non-oxides
Selected, peer reviewed papers from the 2nd International Symposium
on SiAlONs and Non-oxides,
December 2nd – 5th, 2007 Ise-shima Royal Hotel, Mie, Japan
Edited by
Katsutoshi Komeya
Yi-Bing Cheng
Junichi Tatami
Mamoru Mitomo
Copyright 2009 Trans Tech Publications Ltd, Switzerland
All rights reserved. No part of the contents of this publication
may be reproduced or
transmitted in any form or by any means without the written
permission of the publisher.
Trans Tech Publications Ltd
Switzerland
http://www.ttp.net
Volume 403 of Key Engineering Materials ISSN 1013-9826 Full text
available online at http://www.scientific.net
Distributed worldwide by and in the Americas by
Trans Tech Publications Ltd. Trans Tech Publications Inc.
Laubisrutistr. 24 PO Box 699, May Street
CH-8712 Stafa-Zurich Enfield, NH 03748
Switzerland USA
Fax: +41 (44) 922 10 33 Fax: +1 (603) 632-5611
e-mail:
[email protected] e-mail:
[email protected]
Sponsors
Foreword
Nitrides
Preparation and Characterization of MM’Si4N6C Ceramics D.P.
Thompson and Y. Zhang 3
The Phase Evolution in the Si3N4-AlN System after High-Energy
Mechanical Treatment of the Precursor Powder M. Sopicka-Lizer, T.
Pawlik, T. Wodek and M. Tacula 7
New Green Phosphor Ba3Si6O12N2:Eu for White LED: Crystal Structure
and Optical Properties M. Mikami, S. Shimooka, K. Uheda, H. Imura
and N. Kijima 11
Application of Nitride and Oxynitride Compounds to Various
Phosphors for White LED K. Uheda 15
Fabrication of Electrically Conductive Si3N4 Ceramics by Dispersion
of Carbon Nanotubes S. Yoshio, J. Tatami, T. Wakihara, K. Komeya
and T. Meguro 19
Low Temperature Sintering of Si3N4 Ceramics and its Applicability
as an Inert Matrix of the Transuranium Elements for Transmutation
of Minor Actinides T. Yano, J. Yamane and K. Yoshida 23
2.45 GHz Microwave Sintering of Silicon Nitride S. Chockalingam,
J.P. Kelly, V.R.W. Amarakoon and J.R. Varner 27
Sintering Shrinkage Behavior of Si3N4 Ceramics Prepared by a
Post-Reaction Sintering Technique H. Yabuki, T. Wakihara, J.
Tatami, K. Komeya, T. Meguro, H. Kita, N. Kondo and K. Hirao
31
Sintering Shrinkage Behavior and Mechanical Properties of
HfO2-Added Si3N4 Ceramics D. Horikawa, J. Tatami, T. Wakihara, K.
Komeya and T. Meguro 35
Fabrication and Evaluation of AlN–SiC Solid Solutions with p-Type
Electrical Conduction R. Kobayashi, J. Tatami, T. Wakihara, K.
Komeya, T. Meguro, R. Tu and T. Goto 39
Atomic Resolution and In Situ Characterization of Structural
Ceramics Y. Ikuhara 43
Non-Oxide Ceramic Nanocomposites with Multifunctionality T.
Kusunose and T. Sekino 45
Electrical Resistivity Control of Hot-Pressed Aluminum Nitride
Ceramics N. Yamada, J. Yoshikawa, Y. Katsuda and H. Sakai 49
Fracture Resistance and Wear Properties of Silicon Nitride Ceramics
H. Miyazaki, H. Hyuga, Y. Yoshizawa, K. Hirao and T. Ohji 53
Oxidation of Rare Earth Silicon Oxynitride J-Phases J. Takahashi
and T. Suehiro 57
Effect of Second Phase After-Heat Treatment on the Thermal
Conductivity of AlN Ceramics H.K. Lee and D.K. Kim 61
Thermal Conductivity Measurement of the AlN Ceramics at the Grain
Scale Using Thermoreflectance Technique S.K. Lee, I. Yamada, S.
Kume, H. Nakano and K. Watari 65
Viscosity Measurement of Molten RE-Mg-Si-O-N (RE=Y, Gd, Nd and La)
Glasses N. Saito, D. Nakata, S. Sukenaga and K. Nakashima 69
b SiAlONs and Non-oxides
First Principles Calculations of Advanced Nitrides, Oxides and
Alloys I. Tanaka, A. Kuwabara, K. Yuge, A. Seko, F. Oba and K.
Matsunaga 73
Advances in Computation of Temperature-Pressure Phase Diagrams of
High-Pressure Nitrides P. Kroll 77
SiAlONs
Luminescence Properties of α-SiAlONs and Related Compounds R.J.
Xie, M. Mitomo and N. Hirosaki 83
Developments in SiAlON Glasses and their Derivatives: Effects of
Chemistry on Properties S. Hampshire and M.J. Pomeroy 87
Controlled Crystallisation of a Y-Si-Al-O-N Glass Typical of Grain
Boundary Glasses Formed in Silicon Nitride-Based Ceramics M.J.
Pomeroy and S. Hampshire 91
Synthesis and Refinement of β-SiAlON by Nitriding and
Post-Sintering of Si Mixture Y.J. Park, E.A. Noh, J.H. Ahn and H.D.
Kim 95
Interactions between AlN and SiAlON Ceramics A. Kalemtas, N.C.
Acikbas, F. Kara, H. Mandal, K. Krnel and T. Kosma 97
Surface Structuring of α/β-Sialon Ceramics by Plasma-Etching M.
Riva, R. Oberacker, M.J. Hoffmann and C. Ziebert 99
SiAlON B-Phase Glass-Ceramic Microstructures L.K.L. Falk, Y. Menke
and S. Hampshire 103
Development of α-β SiAlON Ceramics from Different Si3N4 Starting
Powders N.C. Acikbas, F. Kara and H. Mandal 107
Effects of Process on Optical Transmittance of Dy-α-Sialon Sintered
at Lower Temperatures J.M. Xue, Q. Liu, M. Fang, L.L. Ma, T.P. Xiu
and L.H. Gui 109
Mechanical Properties of α- and β-SiAlON Composite Ceramics Using
β-SiAlON Powder K. Asakoshi, J. Tatami, K. Komeya, T. Meguro and M.
Yokouchi 111
Tribological Performance of Translucent Dy-α-Sialon Ceramics Q.
Liu, L.H. Gui, J.H. Meng and Z.F. Li 115
High-Temperature Compressive Deformation of SiAlON Polycrystals
Prepared without Additives K. Chihara, D. Hiratsuka, J. Tatami, F.
Wakai and K. Komeya 117
Dielectric Properties of β-SiAlON at High Temperature Using
Perturbation Method Y.H. Seong, H.N. Kim and D.K. Kim 121
Dielectric Properties of SiAlON Ceramics D.K. Kim, H.N. Kim, Y.H.
Seong, S.S. Baek, E.S. Kang and Y.G. Baek 125
Subcritical Crack Growth of α/β-Sialon Ceramics in Distilled Water
M. Riva, R. Oberacker, M.J. Hoffmann and T. Fett 129
Corrosion of β-SiAlON in Molten Aluminium, Cryolite and NaCl-KCl
Mixture T. Plachký, J. Kesan, M. Korenko, Z. Lenéš and P. Šajgalík
133
Preparation and Corrosion of Mullite Thin Film on ß-SiAlON Ceramics
Y. Noritake, H. Kiyono and S. Shimada 135
Thermal Conductivities of β-SiAlONs by Mechanically Activated
Combustion Synthesis R. Sivakumar, K. Aoyagi, T. Watanabe and T.
Akiyama 139
Synthesis and Characterization of β-SiAlON Phosphor Powder Prepared
by Reduction Nitridation of a Zeolite T. Wakihara, Y. Saito, J.
Tatami, K. Komeya, T. Meguro, Y. Fukuda, N. Matsuda and H. Asai
141
Carbides
New Ceramic Phases in the Ternary Si-C-N System R. Riedel, E.
Horvath-Bordon, H.J. Kleebe, P. Kroll, G. Miehe, P.A. van Aken and
S. Lauterbach 147
A New Route of Forming Silicon Carbide Nanostructures with
Controlled Morphologies Y.B. Cheng, K. Wang and H.T. Wang 149
Key Engineering Materials Vol. 403 c
TEM Study of SiCN Glasses; Polymer Architecture versus Ceramic
Microstructure H.J. Kleebe and H. Störmer 153
Aqueous Processing of Boron Carbide Powders J.X. Zhang, Q.L. Lin
and D.L. Jiang 157
Synthesis of Single-Phase, Hexagonal Plate-Like Al4SiC4 Powder J.S.
Lee, T. Nishimura, H. Tanaka and S.H. Lee 159
Gelcasting of Carbide Ceramics D.L. Jiang 163
Processing of High Performance Silicon Carbide Y. Hirata, N.
Matsunaga, N. Hidaka, S. Tabata and S. Sameshima 165
New Theory of Transformation Induced Grain Growth in Porous SiC H.
Tanaka 169
Evaluation of Microstructure of High-Strength Reaction-Sintered
Silicon Carbide S. Suyama and Y. Itoh 173
Nano-Grained Microstructure Design of Silicon Carbide Ceramics by
SPS Process J. Hojo, H. Matsuura and M. Hotta 177
Macro- and Micro-Scale Thermal Conductivities of SiC Single Crystal
and Ceramic I. Yamada, S. Kume, H. Nakano and K. Watari 179
Influence of Additives on Mechanical Properties in Liquid-Phase
Sintered Silicon Carbide Ceramics Y.W. Kim, T. Nishimura and M.
Mitomo 185
Microstructural and Mechanical Properties of Ti3SiC2 Composites
J.L. Huang and H.H. Lu 189
Effect of HfO2 Coating Films on Oxidation Resistance of SiC
Ceramics M. Kasajima, T. Akashi and S. Shimada 193
High Temperature Oxidation of SiC Powder in Oxidizing Atmosphere
Containing Water Vapor T. Akashi, M. Kasajima, C. Muraoka and H.
Kiyono 197
Applying SiC Nanoparticles to Functional Ceramics for Semiconductor
Manufacturing Process M. Konishi 201
First-Principles Study of Ceramic Interfaces: Structures and
Electronic and Mechanical Properties M. Kohyama and S. Tanaka
205
A New Technology with Porous Materials; Progress in the Development
of the Diesel Vehicle Business K. Ohno 207
Synthesis of Nano-Sized SiC Powders by Carbothermal Reduction Y.
Yoshioka, H. Tanaka, M. Konishi and T. Nishimura 211
Composites
Passive Oxidation Behavior of ZrB2-SiC Eutectic Composite Prepared
by Arc Melting R. Tu, H. Hirayama and T. Goto 217
Fabrication and Mechanical Properties of TiN Nanoparticle-Dispersed
Si3N4 Ceramics from Si3N4-Nano TiO2 Composite Particles Obtained by
Mechanical Treatment E. Kodama, J. Tatami, T. Wakihara, T. Meguro,
K. Komeya and H. Nakano 221
Spark Plasma Sintering of Silicon Nitride-Boron Carbide Composites
E. Ayas, A. Kalemtas, G. Arslan, A. Kara and F. Kara 225
In Situ Synthesis and Mechanical Properties of TiN-Si2N2O-Si3N4
Composites H. Kiyono and S. Shimada 227
Dispersion of Carbon Nanotubes in Polysilazanes for the Preparation
of Reinforced Si-C-N Composites L. Fernandez, Y. Li, M. Burghard,
Z. Burghard, P. Gerstel, J. Bill and F. Aldinger 231
Reaction Bonded Silicon Nitride - Silicon Carbide and SiAlON -
Silicon Carbide Refractories for Aluminium Smelting M.I. Jones, R.
Etzion, J. Metson, Y. Zhou, H. Hyuga, Y. Yoshizawa and K. Hirao
235
d SiAlONs and Non-oxides
Tribological Characteristics of Carbon Nano-Fiber Dispersed Silicon
Nitride Based Composites in High-Temperature Fuel M. Wada, K.
Kashiwagi, S. Kitaoka and Y. Fuwa 239
Preparation of β SiAlON-cBN Composites by Spark Plasma Sintering M.
Hotta and T. Goto 241
A Novel Approach for Preparing Electrically Conductive SiAlON-TiN
Composites by Spark Plasma Sintering E. Ayas, A. Kara and F. Kara
243
Fabrication of AlN Ceramics Using AlN and Nano-Y2O3 Composite
Particles Prepared by Mechanical Treatment D. Hiratsuka, J. Tatami,
T. Wakihara, K. Komeya and T. Meguro 245
Preparation of Precursors for Aluminum Nitride-Based Ceramic
Composites from Cage- Type and Cyclic Building Blocks Y. Sugahara,
H. Nakashima, S. Koyama and Y. Mori 249
Pressureless Melt Infiltrated Non-Oxide Ceramic-Metal Composites A.
Kalemtas, G. Arslan and F. Kara 251
The Oxidation Behavior of ZrB2-Based Mixed Boride S.J. Lee and D.K.
Kim 253
Mechanical and Thermal Properties of Silicon Carbide Composites
with Chopped Si-Al-C Fiber Addition K. Itatani, I.J. Davies and H.
Suemasu 257
Exergy Analysis on the Life Cycle of Ceramic Parts H. Kita, H.
Hyuga, N. Kondo and T. Ohji 261
SiAlON Microstructures L.K.L. Falk 265
Synthesis of Non-Oxide Ceramic Fine-Powders from Organic Precursors
T. Nishimura, S. Ishihara, Y. Yoshioka and H. Tanaka 269
Sponsored by The Japan Society for Promotion of Science
(JSPS)
Yokohama National University JSPS 124th Committee
Hosokawa Powder Technology Foundation
COMMITTEE MEMBERS
International Advisory Committee
A. Bellosi (CNR-IRTEC, Italy) G. Fantozzi (GEMPPM, France) L. Gao
(Shanghai Institute of Ceramics, China) H. T. Hintzen (Eindhoven
University of Technology, The Netherlands) J. Hojo (Kyushu
University, Japan) M. Ibukiyama (Denki Kagaku Kogyo, Japan) Y. Ito
(Toshiba Co., Japan) S. J. Kang (Korea Advanced Institute of
Science and Technology, Korea) D. K. Kim (Korea Advanced Institute
of Science and Technology, Korea) H. D. Kim (Korea Institute of
Machinery & Materials, Korea) H. T. Lin (Oak Ridge National
Laboratory, USA) K. MacKenzie (Victoria University of Wellington,
New Zealand) Y. Matsuo (Tokyo Institute of Technology, Japan) K.
Niihara (Nagaoka University of Technology, Japan) R. Riedel
(Technical University Darmstadt, Germany) T. Rouxel (University of
Rennes, France) P. Sajgalik (Slovak Academy of Science, Slovakia)
H. Sakai (NGK Insulator Ltd, Japan) Z. -J. Shen (Stockholm
University, Sweden) M. Singh (NASA Glenn Research Center, U.S.A.)
K. Uematsu (Nagaoka University Technology, Japan) F. Wakai (Tokyo
Institute of Technology, Japan) A. Yamaguchi (JUTEM, Japan) G.
Zhang (Shanghai Institute of Ceramics, China)
Sponsored by The Japan Society for Promotion of Science
(JSPS)
Yokohama National University JSPS 124th Committee
Hosokawa Powder Technology Foundation
COMMITTEE MEMBERS
International Advisory Committee
A. Bellosi (CNR-IRTEC, Italy) G. Fantozzi (GEMPPM, France) L. Gao
(Shanghai Institute of Ceramics, China) H. T. Hintzen (Eindhoven
University of Technology, The Netherlands) J. Hojo (Kyushu
University, Japan) M. Ibukiyama (Denki Kagaku Kogyo, Japan) Y. Ito
(Toshiba Co., Japan) S. J. Kang (Korea Advanced Institute of
Science and Technology, Korea) D. K. Kim (Korea Advanced Institute
of Science and Technology, Korea) H. D. Kim (Korea Institute of
Machinery & Materials, Korea) H. T. Lin (Oak Ridge National
Laboratory, USA) K. MacKenzie (Victoria University of Wellington,
New Zealand) Y. Matsuo (Tokyo Institute of Technology, Japan) K.
Niihara (Nagaoka University of Technology, Japan) R. Riedel
(Technical University Darmstadt, Germany) T. Rouxel (University of
Rennes, France) P. Sajgalik (Slovak Academy of Science, Slovakia)
H. Sakai (NGK Insulator Ltd, Japan) Z. -J. Shen (Stockholm
University, Sweden) M. Singh (NASA Glenn Research Center, U.S.A.)
K. Uematsu (Nagaoka University Technology, Japan) F. Wakai (Tokyo
Institute of Technology, Japan) A. Yamaguchi (JUTEM, Japan) G.
Zhang (Shanghai Institute of Ceramics, China)
Organising Committee
Chairman: K. Komeya (Yokohama National University, Japan)
Vice-Chairman: D. P. Thompson (University of Newcastle-upon-Tyne,
UK) M. Mitomo (National Institute for Materials Science, Japan)
Members: Y. -B. Cheng (Monash University, Australia) T. Goto
(Tohoku University, Japan) S. Hampshire (University of Limerick,
Ireland) M. J. Hoffmann (University of Karlsruhe, Germany) I. -W.
Chen (University of Pennsylvania, U.S.A.) S. Kanzaki (National
Institute of Advanced Industrial Science and Technology, Japan) H.
Mandal (Anadolu University, Turkey) M. Naito (Osaka University,
Japan) S. Shimada (Hokkaido University, Japan) H. Tanaka (National
Institute for Materials Science, Japan)
Programme Committee
Chairman: Y. -B. Cheng (Monash University, Australia) Members: Y.
Ikuhara (University of Tokyo, Japan) S. Itatani (Sophia University,
Japan) K. Hirao (National Institute of Advanced Industrial Science
and Technology, Japan) N. Hirosaki (National Institute for
Materials Science, Japan) T. Ohji (National Institute of Advanced
Industrial Science and Technology, Japan) Y. Sugawara (Waseda
University, Japan) J. Tatami (Yokohama National University, Japan)
Y. Ukyo (Toyota Central R&D Labs., Inc., Japan) T. Yano (Tokyo
Institute of Technology, Japan)
Local Committee
Chairman: J. Tatami (Yokohama National Univ., Japan) Members: T.
Akashi (Hokkaido University, Japan) N. Hotta (Niigata University,
Japan) H. Kita (National Institute of Advanced Industrial Science
and Technology, Japan) T. Nishimura (National Institute for
Materials Science, Japan) M. Ooyanagi (Ryukoku University, Japan)
J. Takahashi (Tohoku University, Japan) H. Tanaka (National
Institute for Materials Science, Japan) P. Xin (Covalent Materials,
Japan) H. Wada (Bridgistone Co., Japan) K. Watari (National
Institute of Advanced Industrial Science and Technology, Japan) T.
Wakihara (Yokohama National University, Japan) N. Yamada (NGK
Insulators Ltd., Japan)
Organising Committee
Chairman: K. Komeya (Yokohama National University, Japan)
Vice-Chairman: D. P. Thompson (University of Newcastle-upon-Tyne,
UK) M. Mitomo (National Institute for Materials Science, Japan)
Members: Y. -B. Cheng (Monash University, Australia) T. Goto
(Tohoku University, Japan) S. Hampshire (University of Limerick,
Ireland) M. J. Hoffmann (University of Karlsruhe, Germany) I. -W.
Chen (University of Pennsylvania, U.S.A.) S. Kanzaki (National
Institute of Advanced Industrial Science and Technology, Japan) H.
Mandal (Anadolu University, Turkey) M. Naito (Osaka University,
Japan) S. Shimada (Hokkaido University, Japan) H. Tanaka (National
Institute for Materials Science, Japan)
Programme Committee
Chairman: Y. -B. Cheng (Monash University, Australia) Members: Y.
Ikuhara (University of Tokyo, Japan) S. Itatani (Sophia University,
Japan) K. Hirao (National Institute of Advanced Industrial Science
and Technology, Japan) N. Hirosaki (National Institute for
Materials Science, Japan) T. Ohji (National Institute of Advanced
Industrial Science and Technology, Japan) Y. Sugawara (Waseda
University, Japan) J. Tatami (Yokohama National University, Japan)
Y. Ukyo (Toyota Central R&D Labs., Inc., Japan) T. Yano (Tokyo
Institute of Technology, Japan)
Local Committee
Chairman: J. Tatami (Yokohama National Univ., Japan) Members: T.
Akashi (Hokkaido University, Japan) N. Hotta (Niigata University,
Japan) H. Kita (National Institute of Advanced Industrial Science
and Technology, Japan) T. Nishimura (National Institute for
Materials Science, Japan) M. Ooyanagi (Ryukoku University, Japan)
J. Takahashi (Tohoku University, Japan) H. Tanaka (National
Institute for Materials Science, Japan) P. Xin (Covalent Materials,
Japan) H. Wada (Bridgistone Co., Japan) K. Watari (National
Institute of Advanced Industrial Science and Technology, Japan) T.
Wakihara (Yokohama National University, Japan) N. Yamada (NGK
Insulators Ltd., Japan)
FOREWORD This Volume contains the papers presented at the
International Symposium on SiAlONS and Non-oxides, held at
Ise-Shima, Mie, Japan on December 2-5, 2007. In these fifty years,
significant progress has been achieved in SiAlONs and non-oxides.
In particular, advances in research and development in the last
thirty years have made SiAlONs and non-oxides as important
engineering ceramics and new functional ceramics of today.
Following the successful 1st International Symposium on SiAlONs in
2001, the 2nd International Symposium on SiAlONs and Non-oxides was
held as a broader forum to discuss the most recent developments in
research and applications of SiAlONs and non-oxides at Ise-Shima,
Mie, Japan, on December 2-5, 2007. Papers presented in this volume
are authored by a group of international leading experts in SiAlON
and non-oxide materials and give an excellent indication of the
current and future directions in the field. All of the papers have
been peer-reviewed prior to publication. The Symposium was
sponsored by the Japan Society for the Promotion of Science (JSPS),
Yokohama National University, Japan, JSPS 124th Committee and
Hosokawa Powder Technology Foundation, Japan, and was cooperated by
the Ceramic Society of Japan. Supports by the members of the
International Advisory Committee, Drs. A. Bellosi, G. Fantozzi, L.
Gao, H. T. Hintzen, J. Hojo, M. Ibukiyama, Y. Ito, S. J. Kang, D.
K. Kim, H. D. Kim, H. T. Lin, K. MacKenzie, Y. Matsuo, K. Niihara,
R. Riedel, T. Rouxel, P. Sajgalik, H. Sakai, Z.-J. Shen, M. Singh,
K. Uematsu, F. Wakai, A. Yamaguchi, G. Zhang, and by the members of
the Local Organizing Committee and the Program Committee are
gratefully acknowledged by the conference organizers. Finally, a
special acknowledgement is due to the students and stuff of
Yokohama National University and researchers of AIST, Nagoya, and
NIMS, Tsukuba for their sustained assistance to the symposium.
Katsutoshi Komeya, Yi-Bing Cheng, Junichi Tatami and Mamoru Mitomo
Yokohama, Melbourne and Tsukuba, 2008
FOREWORD This Volume contains the papers presented at the
International Symposium on SiAlONS and Non-oxides, held at
Ise-Shima, Mie, Japan on December 2-5, 2007. In these fifty years,
significant progress has been achieved in SiAlONs and non-oxides.
In particular, advances in research and development in the last
thirty years have made SiAlONs and non-oxides as important
engineering ceramics and new functional ceramics of today.
Following the successful 1st International Symposium on SiAlONs in
2001, the 2nd International Symposium on SiAlONs and Non-oxides was
held as a broader forum to discuss the most recent developments in
research and applications of SiAlONs and non-oxides at Ise-Shima,
Mie, Japan, on December 2-5, 2007. Papers presented in this volume
are authored by a group of international leading experts in SiAlON
and non-oxide materials and give an excellent indication of the
current and future directions in the field. All of the papers have
been peer-reviewed prior to publication. The Symposium was
sponsored by the Japan Society for the Promotion of Science (JSPS),
Yokohama National University, Japan, JSPS 124th Committee and
Hosokawa Powder Technology Foundation, Japan, and was cooperated by
the Ceramic Society of Japan. Supports by the members of the
International Advisory Committee, Drs. A. Bellosi, G. Fantozzi, L.
Gao, H. T. Hintzen, J. Hojo, M. Ibukiyama, Y. Ito, S. J. Kang, D.
K. Kim, H. D. Kim, H. T. Lin, K. MacKenzie, Y. Matsuo, K. Niihara,
R. Riedel, T. Rouxel, P. Sajgalik, H. Sakai, Z.-J. Shen, M. Singh,
K. Uematsu, F. Wakai, A. Yamaguchi, G. Zhang, and by the members of
the Local Organizing Committee and the Program Committee are
gratefully acknowledged by the conference organizers. Finally, a
special acknowledgement is due to the students and stuff of
Yokohama National University and researchers of AIST, Nagoya, and
NIMS, Tsukuba for their sustained assistance to the symposium.
Katsutoshi Komeya, Yi-Bing Cheng, Junichi Tatami and Mamoru Mitomo
Yokohama, Melbourne and Tsukuba, 2008
Nitrides Nitrides
D.P. Thompsona and Yue Zhang
Advanced Materials Group, School of Chemical Engineering &
Advanced Materials,
University of Newcastle, Newcastle upon Tyne NE1 7RU,UK
a
[email protected]
Abstract. The preparation of high temperature ceramics
simultaneously containing silicon, nitrogen
and carbon has only relatively recently become an area of interest
for inorganic crystal chemists,
and the recent discovery of a new series of carbonitrides with the
general formula MM’Si4N6C is of
interest because of the good high temperature properties they
appear to display. On the one hand, M
and M’ can be the same trivalent metal - either rare earth or
yttrium; in this case, the resulting
compounds display orthorhombic (pseudo-hexagonal) structures.
Alternatively the metals may be a
mix of di- (Ca,Sr, Ba) and tri-valent (Y,Ln) cations, in which case
the carbon is replaced by
nitrogen, and the overall symmetry is hexagonal. Other quaternary
nitrides of a similar type can be
produced if the two metal cations remain trivalent and one of the
silicon atoms is replaced by
aluminium.
The present study describes the preparation of powder samples of
Y2Si4N6C and LaYSi4N6C
starting from YH2, La, Si3N4 and carbon precursors, and summarises
attempts to achieve a dense
product by hot-pressing at 1700-1800 o C. Some preliminary
mechanical property measurements are
included.
Introduction
In the first symposium [1], it was argued that the exploration of
new metal sialon compounds had
been very much neglected because of the impressive success of α-
and β- Si3N4 and their sialon
equivalents as high-strength engineering materials. As far as
α-sialon is concerned, there have been
an increasing range of nitrogen-rich oxynitrides reported in recent
years which are likely to have
similar properties as judged by the facts that (a) they consist of
a 3D linkage of [SiN4] tetrahedra,
(b) the proportion of large additional metal (M) cations is small
relative to Si+Al, and (c) Si+N by
Al+O replacement can generally occur, thereby making preparation/
densification easier and in
some cases allowing different M cations to be incorporated.
However, whereas most of these phases
are characterised, with unit cell (and sometimes full crystal
structure) information available, very
few attempts have been made to explore ceramic properties. During
the last 5 years, Lewis and
coworkers [2] carried out such a study on the sialon S-phase
(BaSi5Al2N8O2), and showed that this
was relatively easy to prepare in pure form starting with mixtures
of BaCO3, Al2O3, Si3N4 and AlN,
and could be readily densified, but careful TEM work showed that
the morphology was needle-
shaped, and therefore unlikely to display high values of fracture
toughness. Nevertheless, the
exercise was useful in showing that these compounds could be made
into dense materials with
acceptable properties.
The present paper describes a similar exercise applied to the
MM’Si4N6C group of ceramics
where M and M’ are typically Y or the rare earths. The first
published work on these compounds
was reported by Höppe et al. [3] on the compound LaYbSi4N6C, who
showed this to be iso-
structural with the group of compounds of the type ABSi4N7 (A =
Ca,Sr, Ba; B = Y or rare earth),
being built up of structural units of the type [Si4N12C],
consisting of four tetrahedra meeting at a
point, and stacked on top of the another (but with a 60 o rotation)
in the crystallographic z direction.
In these units, a central non-metal atom is linked to four
tetrahedra and is therefore an ideal site for
carbon to occupy in the carbonitride derivatives. Concurrent
Newcastle work [4,5] showed that
Preparation and Characterization of MM’Si4N6C Ceramics
D.P. Thompsona and Yue Zhang
Advanced Materials Group, School of Chemical Engineering &
Advanced Materials,
University of Newcastle, Newcastle upon Tyne NE1 7RU,UK
a
[email protected]
Abstract. The preparation of high temperature ceramics
simultaneously containing silicon, nitrogen
and carbon has only relatively recently become an area of interest
for inorganic crystal chemists,
and the recent discovery of a new series of carbonitrides with the
general formula MM’Si4N6C is of
interest because of the good high temperature properties they
appear to display. On the one hand, M
and M’ can be the same trivalent metal - either rare earth or
yttrium; in this case, the resulting
compounds display orthorhombic (pseudo-hexagonal) structures.
Alternatively the metals may be a
mix of di- (Ca,Sr, Ba) and tri-valent (Y,Ln) cations, in which case
the carbon is replaced by
nitrogen, and the overall symmetry is hexagonal. Other quaternary
nitrides of a similar type can be
produced if the two metal cations remain trivalent and one of the
silicon atoms is replaced by
aluminium.
The present study describes the preparation of powder samples of
Y2Si4N6C and LaYSi4N6C
starting from YH2, La, Si3N4 and carbon precursors, and summarises
attempts to achieve a dense
product by hot-pressing at 1700-1800 o C. Some preliminary
mechanical property measurements are
included.
Introduction
In the first symposium [1], it was argued that the exploration of
new metal sialon compounds had
been very much neglected because of the impressive success of α-
and β- Si3N4 and their sialon
equivalents as high-strength engineering materials. As far as
α-sialon is concerned, there have been
an increasing range of nitrogen-rich oxynitrides reported in recent
years which are likely to have
similar properties as judged by the facts that (a) they consist of
a 3D linkage of [SiN4] tetrahedra,
(b) the proportion of large additional metal (M) cations is small
relative to Si+Al, and (c) Si+N by
Al+O replacement can generally occur, thereby making preparation/
densification easier and in
some cases allowing different M cations to be incorporated.
However, whereas most of these phases
are characterised, with unit cell (and sometimes full crystal
structure) information available, very
few attempts have been made to explore ceramic properties. During
the last 5 years, Lewis and
coworkers [2] carried out such a study on the sialon S-phase
(BaSi5Al2N8O2), and showed that this
was relatively easy to prepare in pure form starting with mixtures
of BaCO3, Al2O3, Si3N4 and AlN,
and could be readily densified, but careful TEM work showed that
the morphology was needle-
shaped, and therefore unlikely to display high values of fracture
toughness. Nevertheless, the
exercise was useful in showing that these compounds could be made
into dense materials with
acceptable properties.
The present paper describes a similar exercise applied to the
MM’Si4N6C group of ceramics
where M and M’ are typically Y or the rare earths. The first
published work on these compounds
was reported by Höppe et al. [3] on the compound LaYbSi4N6C, who
showed this to be iso-
structural with the group of compounds of the type ABSi4N7 (A =
Ca,Sr, Ba; B = Y or rare earth),
being built up of structural units of the type [Si4N12C],
consisting of four tetrahedra meeting at a
point, and stacked on top of the another (but with a 60 o rotation)
in the crystallographic z direction.
In these units, a central non-metal atom is linked to four
tetrahedra and is therefore an ideal site for
carbon to occupy in the carbonitride derivatives. Concurrent
Newcastle work [4,5] showed that
Key Engineering Materials Vol. 403 (2009) pp 3-6 © (2009) Trans
Tech Publications, Switzerland
doi:10.4028/www.scientific.net/KEM.403.3
when M and M’ were different cations in the carbonitride
structures, the resulting compounds were
hexagonal, whereas when they were the same, this resulted in a
slight distortion to orthorhombic
(pseudo-hexagonal), the distortion being caused by alternate
columns of [Si4N12C] being anti-
parallel, in contrast to perfect parallel alignment when the two
cations were different.
Experimental
Preparation of oxygen-free nitrides and carbonitrides is difficult,
because during initial processing it
is almost impossible to avoid oxygen pick-up especially during
addition of the ionic metals used
jointly as sintering aids and as a source of the third metal cation
in the formulation (as for example
in α-sialons). Recently Esmaeilzadeh and colleagues at University
of Stockholm [6] have shown
that providing the third metal as either a hydride or as the metal
powders is an excellent way of
minimising oxygen pick-up. In the present study, it was decided to
compare one orthorhombic and
one hexagonal carbonitride; Y2Si4N6C was selected as the
orthorhombic derivative and LaYSi4N6C
as the hexagonal variant. YH2 (Aldrich) was used as the source of
yttrium, and lanthanum powder
(supplied immersed in oil - Aldrich) as the source of lanthanum.
Previous work had shown that
when firing samples in a carbon resistance furnace in a carbon
crucible, there was more than
sufficient carbon pick-up to provide the carbon required by the
formula; traces of oil present in the
La starting powder probably evaporated from the sample rather than
being pyrolysed to carbon.
Si3N4 powder was used as the source of silicon. The starting
powders were mixed in isopropanol,
and in the first experiments were compacted and fired in a carbon
resistance furnace at 1750 o C in a
nitrogen atmosphere. After only a few experiments it was possible
to produce >95% pure samples
of both the Y2 and LaY compounds (see Figure 1). The main impurity
was J-phase (Y4Si2O7N2),
stabilised by oxygen introduced either during processing or from
impurity in the nitriding gas.
Attempts were then made to prepare dense samples by hot-pressing in
graphite dies, and these were
prepared in two ways. In the first case, the already pre-prepared
carbonitride powders were finely
ground, and hot-pressed initially without additive at 1750 o C, and
then with extra J-phase (to provide
additional liquid phase) at 1800 o C. However, it was found (Figure
2(a)), that this resulted merely in
squeezing the grains together, rather than modifying grain
morphology and eliminating pores.
Alternatively, the original powder mixes were used, with the
assumption that the oxygen in the air
pockets between the starting powder grains would be removed by the
carbon of the die at low
temperatures, leaving the residual nitrogen to be available to
nitride the YH2 and La. In fact this
assumption was successful, and the final products were
substantially the required phases, even
though in every case there was a slight increase in the percentage
of J-phase. However, even
though this second method resulted in slightly better
densification, it was still impossible to achieve
full density, and the maximum values achieved were 93% for
Y2Si4N6C, and 91% for LaYSi4N6C.
Figure 2(b) shows a sample hot-pressed from the original starting
powders, showing regions of J-
phase and even some silicon carbide, in addition to the predominant
LaYSi4N6C phase. Regions of
black porosity are also clearly visible. Further work is needed to
identify ways of achieving 100%
densification in both the Y2 and the LaY samples
Preliminary measurements of hardness and fracture toughness made on
the best hot-pressed
Y2Si4N6C sample yielded values of 14.9GPa and 3.6 MPam ½
respectively, and on similar hot-
pressed samples of LaYSi4N6C of 15GPa and 4.0 MPam ½ . These are
typical values for a nitrogen
ceramic, and even though promising, should not be given too much
credibility, because the samples
have not yet fully densified, and there is not significant
morphological development of product
grains. Further work on microstructure is needed, which in turn
will emerge in response to better
densification procedures being developed.
when M and M’ were different cations in the carbonitride
structures, the resulting compounds were
hexagonal, whereas when they were the same, this resulted in a
slight distortion to orthorhombic
(pseudo-hexagonal), the distortion being caused by alternate
columns of [Si4N12C] being anti-
parallel, in contrast to perfect parallel alignment when the two
cations were different.
Experimental
Preparation of oxygen-free nitrides and carbonitrides is difficult,
because during initial processing it
is almost impossible to avoid oxygen pick-up especially during
addition of the ionic metals used
jointly as sintering aids and as a source of the third metal cation
in the formulation (as for example
in α-sialons). Recently Esmaeilzadeh and colleagues at University
of Stockholm [6] have shown
that providing the third metal as either a hydride or as the metal
powders is an excellent way of
minimising oxygen pick-up. In the present study, it was decided to
compare one orthorhombic and
one hexagonal carbonitride; Y2Si4N6C was selected as the
orthorhombic derivative and LaYSi4N6C
as the hexagonal variant. YH2 (Aldrich) was used as the source of
yttrium, and lanthanum powder
(supplied immersed in oil - Aldrich) as the source of lanthanum.
Previous work had shown that
when firing samples in a carbon resistance furnace in a carbon
crucible, there was more than
sufficient carbon pick-up to provide the carbon required by the
formula; traces of oil present in the
La starting powder probably evaporated from the sample rather than
being pyrolysed to carbon.
Si3N4 powder was used as the source of silicon. The starting
powders were mixed in isopropanol,
and in the first experiments were compacted and fired in a carbon
resistance furnace at 1750 o C in a
nitrogen atmosphere. After only a few experiments it was possible
to produce >95% pure samples
of both the Y2 and LaY compounds (see Figure 1). The main impurity
was J-phase (Y4Si2O7N2),
stabilised by oxygen introduced either during processing or from
impurity in the nitriding gas.
Attempts were then made to prepare dense samples by hot-pressing in
graphite dies, and these were
prepared in two ways. In the first case, the already pre-prepared
carbonitride powders were finely
ground, and hot-pressed initially without additive at 1750 o C, and
then with extra J-phase (to provide
additional liquid phase) at 1800 o C. However, it was found (Figure
2(a)), that this resulted merely in
squeezing the grains together, rather than modifying grain
morphology and eliminating pores.
Alternatively, the original powder mixes were used, with the
assumption that the oxygen in the air
pockets between the starting powder grains would be removed by the
carbon of the die at low
temperatures, leaving the residual nitrogen to be available to
nitride the YH2 and La. In fact this
assumption was successful, and the final products were
substantially the required phases, even
though in every case there was a slight increase in the percentage
of J-phase. However, even
though this second method resulted in slightly better
densification, it was still impossible to achieve
full density, and the maximum values achieved were 93% for
Y2Si4N6C, and 91% for LaYSi4N6C.
Figure 2(b) shows a sample hot-pressed from the original starting
powders, showing regions of J-
phase and even some silicon carbide, in addition to the predominant
LaYSi4N6C phase. Regions of
black porosity are also clearly visible. Further work is needed to
identify ways of achieving 100%
densification in both the Y2 and the LaY samples
Preliminary measurements of hardness and fracture toughness made on
the best hot-pressed
Y2Si4N6C sample yielded values of 14.9GPa and 3.6 MPam ½
respectively, and on similar hot-
pressed samples of LaYSi4N6C of 15GPa and 4.0 MPam ½ . These are
typical values for a nitrogen
ceramic, and even though promising, should not be given too much
credibility, because the samples
have not yet fully densified, and there is not significant
morphological development of product
grains. Further work on microstructure is needed, which in turn
will emerge in response to better
densification procedures being developed.
4 SiAlONs and Non-oxides
(a)
(b)
Figure 1. XRD patterns of (a) Y2Si4N6C and (b) LaYSi4N6C prepared
by nitriding YH2/Si3N4 and
La/YH2/Si3N4 mixtures respectively in a carbon furnace at 1750 o
C.
(a)
(b)
Figure 1. XRD patterns of (a) Y2Si4N6C and (b) LaYSi4N6C prepared
by nitriding YH2/Si3N4 and
La/YH2/Si3N4 mixtures respectively in a carbon furnace at 1750 o
C.
Key Engineering Materials Vol. 403 5
(a) (b)
Figure 2. Microstructures of hot-pressed (a) pre-prepared
LaYSi4N6C, and (b) powder mixes of
La/YH2/Si3N4 to produce LaYSi4N6C, both at 1750 o C.
Conclusions
Powder samples of MM’Si4N6C ceramics (M = yttrium; M’ = yttrium or
lanthanum) were relatively
easily prepared by reacting YH2/Si3N4 and La/YH2/Si3N4 mixes in
nitrogen in a carbon element
furnace at 1750 o C. The main impurity was J-phase (Y4Si2O7N2),
present because of the small levels
of oxygen impurity incorporated during powder processing. Attempts
to densify these materials
using either the pre-prepared powders, or the original powder
starting mix were not fully successful,
probably because the solubility of carbon in the nitride/oxynitride
liquids present at high-
temperature in these systems is very low, and even though some
rearrangement no doubt occurs
assisted by the applied pressure, there is almost no densification
from solution/reprecipitation.
Further work is continuing on related pure nitrides of the type
MM’Si4N7, where M is divalent
(Ca,Sr,Ba) and M’ is Y or La, to establish whether these materials
densify by traditional liquid
phase sintering methods.
References [1] K. Liddell and D.P. Thompson, Key. Engineering
Materials, 237, (2003), 1-10.
[2] M.H. Lewis, B. Basu, M.E. Smith, M. Bunyard and T. Kemp,
Silicates Industrielles, Special
Issue, 69(7-8), (2004), 225-32.
[3] H.A. Höppe, G. Kotzyba, R. Pöttgen and W. Schnick, J. Mater.
Chem., 11, (2001), 3300-06.
[4] K. Liddell, D.P. Thompson, T. Bräuniger and R.K. Harris, J.
Eur. Ceram. Soc., 25, (2005), 37-
47.
[5] K.Liddell, D.P. Thompson and S.J. Teat, J. Eur. Ceram. Soc.,
25, (2005), 49-54.
[6] A.S. Hakeem, J. Grins and S. Esmaeilzadeh, J. Eur. Ceram. Oc.,
27, (2007), 4783-87.
(a) (b)
Figure 2. Microstructures of hot-pressed (a) pre-prepared
LaYSi4N6C, and (b) powder mixes of
La/YH2/Si3N4 to produce LaYSi4N6C, both at 1750 o C.
Conclusions
Powder samples of MM’Si4N6C ceramics (M = yttrium; M’ = yttrium or
lanthanum) were relatively
easily prepared by reacting YH2/Si3N4 and La/YH2/Si3N4 mixes in
nitrogen in a carbon element
furnace at 1750 o C. The main impurity was J-phase (Y4Si2O7N2),
present because of the small levels
of oxygen impurity incorporated during powder processing. Attempts
to densify these materials
using either the pre-prepared powders, or the original powder
starting mix were not fully successful,
probably because the solubility of carbon in the nitride/oxynitride
liquids present at high-
temperature in these systems is very low, and even though some
rearrangement no doubt occurs
assisted by the applied pressure, there is almost no densification
from solution/reprecipitation.
Further work is continuing on related pure nitrides of the type
MM’Si4N7, where M is divalent
(Ca,Sr,Ba) and M’ is Y or La, to establish whether these materials
densify by traditional liquid
phase sintering methods.
References [1] K. Liddell and D.P. Thompson, Key. Engineering
Materials, 237, (2003), 1-10.
[2] M.H. Lewis, B. Basu, M.E. Smith, M. Bunyard and T. Kemp,
Silicates Industrielles, Special
Issue, 69(7-8), (2004), 225-32.
[3] H.A. Höppe, G. Kotzyba, R. Pöttgen and W. Schnick, J. Mater.
Chem., 11, (2001), 3300-06.
[4] K. Liddell, D.P. Thompson, T. Bräuniger and R.K. Harris, J.
Eur. Ceram. Soc., 25, (2005), 37-
47.
[5] K.Liddell, D.P. Thompson and S.J. Teat, J. Eur. Ceram. Soc.,
25, (2005), 49-54.
[6] A.S. Hakeem, J. Grins and S. Esmaeilzadeh, J. Eur. Ceram. Oc.,
27, (2007), 4783-87.
6 SiAlONs and Non-oxides
The phase evolution in the Si3N4-AlN system after high-energy
mechanical treatment of the precursor powder
Magorzata Sopicka-Lizer, Tomasz Pawlik, Tomasz Wodek, Marta Tacula
Silesian University of Technology, 40-019 Katowice, Krasiskiego 8,
Poland
[email protected],
[email protected],
[email protected],
[email protected]
Keywords: β-sialon, mechanical activation, precursor,
nanostructured powder Abstract. The high-energy milling uses the
mechanical energy to activate chemical reactions by developing
structural changes in the powder particles. High-energy milling
with an acceleration of 28g was applied for the mechanical
activation of the aluminium and silicon nitrides mixture with
yttria additive. The activated powders showed the significant
damage of the crystal structure and limited formation of a solid
solution. Sintering of the activated precursor demonstrated higher
ability for densification and started at 300 ºC lower temperature
in comparison to the standard mixture. The phase evolution during
sintering was dependent on the starting composition and degree of
powder activation.
Introduction
High-energy milling of β-SiAlON precursor has been reported lately
as a new method of the nitride-based powder preparation for
subsequent densification at lower temperature and/or nanoceramic
manufacturing [1- 4]. The nanostructured precursor powder offers
new possibilities in tailoring the microstructure and properties of
silicon nitride ceramics and pressureless sintering techniques
could be applied for manufacturing of the fully dense silicon
nitride ceramics. However, little is known about the phase
transformation and densification in the Si-Al-O-N system with the
highly defective crystal structure. The previous studies on
densification of nanosized β-Si3N4 showed dependence of α→β
transformation and the features of the resultant microstructure on
the amount and size of the β-nuclei [5]. The aim of this work is to
compare the low-temperature behavior and the phase evolution of the
Si3N4-AlN mixture after high-energy activation in a planetary
mill.
Materials and methods
The initial powders were α-Si3N4 (H.C.Starck-B7), 10-15 wt% of
β-Si3N4), β-Si3N4 (Aldrich, -325 mesh, 5 wt% of Si2N2O
contamination), AlN (H.C.Starck-C), Y2O3 (H.C.Starck-C). The
composition of the batches was chosen to be close to z=0.4 in the
β-SiAlON solid solution: 89,3 wt% of Si3N4, 5.7 wt% of AlN and 5
wt% of Y2O3. Alumina was eliminated from the precursor mixture
because of the expected oxidation during milling. There were
prepared two batches for activation milling: with α-Si3N4 (A) and
β-Si3N4 (B) as a source of silicon nitride. Both batches were
activated in a MPP-1 (TTD, Russia) planetary mill for 30 min or 60
min with silicon nitride balls. The ball-to-powder ratio was 6:1
and the acceleration of the centrifugal field attained the value of
28g (g is the gravitational acceleration). For the comparison
reason the reference A-RB and B-RB batches were prepared by mixing
the components with isopropanol on the roller bench for 24 hours.
The powders after activation or mixing were dispersed in an
aezotropic MEK+EtOH solvent with hypermer KD1 dispersant.
Subsequently the suspensions were dried at room temperature, the
powders were cold uni-axially pressed into form of tablets and
isostatically pressed with 250 MPa pressure. The resultant tablets
were placed in a BN crucible filled with Si3N4/BN powder bed.
Sintering was performed in a graphite
The phase evolution in the Si3N4-AlN system after high-energy
mechanical treatment of the precursor powder
Magorzata Sopicka-Lizer, Tomasz Pawlik, Tomasz Wodek, Marta Tacula
Silesian University of Technology, 40-019 Katowice, Krasiskiego 8,
Poland
[email protected],
[email protected],
[email protected],
[email protected]
Keywords: β-sialon, mechanical activation, precursor,
nanostructured powder Abstract. The high-energy milling uses the
mechanical energy to activate chemical reactions by developing
structural changes in the powder particles. High-energy milling
with an acceleration of 28g was applied for the mechanical
activation of the aluminium and silicon nitrides mixture with
yttria additive. The activated powders showed the significant
damage of the crystal structure and limited formation of a solid
solution. Sintering of the activated precursor demonstrated higher
ability for densification and started at 300 ºC lower temperature
in comparison to the standard mixture. The phase evolution during
sintering was dependent on the starting composition and degree of
powder activation.
Introduction
High-energy milling of β-SiAlON precursor has been reported lately
as a new method of the nitride-based powder preparation for
subsequent densification at lower temperature and/or nanoceramic
manufacturing [1- 4]. The nanostructured precursor powder offers
new possibilities in tailoring the microstructure and properties of
silicon nitride ceramics and pressureless sintering techniques
could be applied for manufacturing of the fully dense silicon
nitride ceramics. However, little is known about the phase
transformation and densification in the Si-Al-O-N system with the
highly defective crystal structure. The previous studies on
densification of nanosized β-Si3N4 showed dependence of α→β
transformation and the features of the resultant microstructure on
the amount and size of the β-nuclei [5]. The aim of this work is to
compare the low-temperature behavior and the phase evolution of the
Si3N4-AlN mixture after high-energy activation in a planetary
mill.
Materials and methods
The initial powders were α-Si3N4 (H.C.Starck-B7), 10-15 wt% of
β-Si3N4), β-Si3N4 (Aldrich, -325 mesh, 5 wt% of Si2N2O
contamination), AlN (H.C.Starck-C), Y2O3 (H.C.Starck-C). The
composition of the batches was chosen to be close to z=0.4 in the
β-SiAlON solid solution: 89,3 wt% of Si3N4, 5.7 wt% of AlN and 5
wt% of Y2O3. Alumina was eliminated from the precursor mixture
because of the expected oxidation during milling. There were
prepared two batches for activation milling: with α-Si3N4 (A) and
β-Si3N4 (B) as a source of silicon nitride. Both batches were
activated in a MPP-1 (TTD, Russia) planetary mill for 30 min or 60
min with silicon nitride balls. The ball-to-powder ratio was 6:1
and the acceleration of the centrifugal field attained the value of
28g (g is the gravitational acceleration). For the comparison
reason the reference A-RB and B-RB batches were prepared by mixing
the components with isopropanol on the roller bench for 24 hours.
The powders after activation or mixing were dispersed in an
aezotropic MEK+EtOH solvent with hypermer KD1 dispersant.
Subsequently the suspensions were dried at room temperature, the
powders were cold uni-axially pressed into form of tablets and
isostatically pressed with 250 MPa pressure. The resultant tablets
were placed in a BN crucible filled with Si3N4/BN powder bed.
Sintering was performed in a graphite
Key Engineering Materials Vol. 403 (2009) pp 7-10 © (2009) Trans
Tech Publications, Switzerland
doi:10.4028/www.scientific.net/KEM.403.7
furnace (Thermal Technology) in a nitrogen flow of 3l/h in the
temperature range of 1200-1600 ºC for 60 min. One tablet of the
A-30 batch was hot-pressed with 250 bar pressure at 1500 ºC for 60
min. The activated powders were measured for their specific surface
area by BET method (ASAP 2010) and observed in scanning electron
microscope (Hitachi S-4200) as well as transmission microscope
(TEM). The oxygen level was controlled (ELTRA ON) and XRD studies
(X’PERT) with Rietveld refinement were applied for measurement of
the phase composition after activation.
The density of sintered tablets was determined by the Archimedes
method. XRD studies were performed on the polished cross section of
the tablets. The phase composition was calculated and lattice
constants for silicon nitride phases were measured.
Results and discussion
The specific surface area of the initial mixtures was 5.5 m2g-1.
The activation milling for 30 min increased the specific surface
area (SSA) of the precursor to 18,7 m2g-1. Prolongation of the
activation milling to 60 min enlarged SSA to 21,9 m2g-1. 2.4 wt% of
oxygen was measured in the non-activated A-type mixture while
activation milling for 30 min led to a slight increase of oxygen
content (3,9 wt%). Estimation of the mean particle size from BET
measurement showed the value of 80 nm. However, the microscopic
studies revealed rather nanostructured particles than the nanosized
ones (Fig.1). On the other hand, XRD results demonstrate
substantial changes of the phase composition after mechanical
treatment: decrease or decay of yttria and aluminum nitride; thus
increase of silicon nitride content after milling is apparent.
Decline of Y2O3 and AlN component in the activated batches is
apparent as well since they are not identified by XRD. It is
assumed that their crystal lattice has been destroyed to such
extent they were not able to produce the well defined diffraction
lines. The several crystal defects (dislocations, plastic
deformation) were observed under TEM studies.
Thermal treatment of the specimens showed a different behavior of
the activated batches derived samples in comparison to the standard
precursor: full density (98,7%) was obtained for the sintered
tablets if the precursor was activated for 60 min. The similar
results (relative density=97.2 %) were obtained for the A-type
batch but activated for 30 min. Increase of the density was
observed after sintering the activated precursor at 1200 ºC whereas
densification start of the reference tablets from non-activated
powders occurred at 1500 ºC (Fig.2). Densification was accompanied
by the changes of the phase composition of the thermal treated
samples. The sequence of the phase evolution was dependent on the
starting composition and time of the precursor
Figure 1. The powder of the B-type precursor after activation with
28g for 30 min, a-SEM, the microstructure of the green tablet;
b-TEM
50
60
70
80
90
100
Sintering temperature [oC]
activated
RB
Figure 2. The effect of sintering temperature on a relative density
of the B-type mixture. Homogenized (RB) and activated for 60 min.
precursors are compared
furnace (Thermal Technology) in a nitrogen flow of 3l/h in the
temperature range of 1200-1600 ºC for 60 min. One tablet of the
A-30 batch was hot-pressed with 250 bar pressure at 1500 ºC for 60
min. The activated powders were measured for their specific surface
area by BET method (ASAP 2010) and observed in scanning electron
microscope (Hitachi S-4200) as well as transmission microscope
(TEM). The oxygen level was controlled (ELTRA ON) and XRD studies
(X’PERT) with Rietveld refinement were applied for measurement of
the phase composition after activation.
The density of sintered tablets was determined by the Archimedes
method. XRD studies were performed on the polished cross section of
the tablets. The phase composition was calculated and lattice
constants for silicon nitride phases were measured.
Results and discussion
The specific surface area of the initial mixtures was 5.5 m2g-1.
The activation milling for 30 min increased the specific surface
area (SSA) of the precursor to 18,7 m2g-1. Prolongation of the
activation milling to 60 min enlarged SSA to 21,9 m2g-1. 2.4 wt% of
oxygen was measured in the non-activated A-type mixture while
activation milling for 30 min led to a slight increase of oxygen
content (3,9 wt%). Estimation of the mean particle size from BET
measurement showed the value of 80 nm. However, the microscopic
studies revealed rather nanostructured particles than the nanosized
ones (Fig.1). On the other hand, XRD results demonstrate
substantial changes of the phase composition after mechanical
treatment: decrease or decay of yttria and aluminum nitride; thus
increase of silicon nitride content after milling is apparent.
Decline of Y2O3 and AlN component in the activated batches is
apparent as well since they are not identified by XRD. It is
assumed that their crystal lattice has been destroyed to such
extent they were not able to produce the well defined diffraction
lines. The several crystal defects (dislocations, plastic
deformation) were observed under TEM studies.
Thermal treatment of the specimens showed a different behavior of
the activated batches derived samples in comparison to the standard
precursor: full density (98,7%) was obtained for the sintered
tablets if the precursor was activated for 60 min. The similar
results (relative density=97.2 %) were obtained for the A-type
batch but activated for 30 min. Increase of the density was
observed after sintering the activated precursor at 1200 ºC whereas
densification start of the reference tablets from non-activated
powders occurred at 1500 ºC (Fig.2). Densification was accompanied
by the changes of the phase composition of the thermal treated
samples. The sequence of the phase evolution was dependent on the
starting composition and time of the precursor
Figure 1. The powder of the B-type precursor after activation with
28g for 30 min, a-SEM, the microstructure of the green tablet;
b-TEM
50
60
70
80
90
100
Sintering temperature [oC]
activated
RB
Figure 2. The effect of sintering temperature on a relative density
of the B-type mixture. Homogenized (RB) and activated for 60 min.
precursors are compared
8 SiAlONs and Non-oxides
activation but was independent on the technique of sintering (Table
2). α→β-Si3N4 transformation is believed to occur in the presence
of the liquid phase. If there is a sufficient amount of the
Y-Si-Al-O-N eutectic liquid phase then β-phase
crystallization/precipitation follows. The amount of the eutectic
liquid phase depends on temperature if the same chemical
composition is considered, but could also be affected by
amorphization of one or more components. The amount and composition
of the liquid phase at 1500 ºC in the non-activated precursor must
have been insufficient for β-Si3N4 crystallization because only
less than 1/4 of the initial silicon nitride was transformed to
β-phase and YAG presence shows silicon lacking in the liquid at
1500 ºC. On the other hand, significantly more liquid phase was
present at 1500 ºC in the activated precursor: the amount of
β-phase was two times higher in comparison to the non-activated
precursor. Application of the hot-pressing technique can not change
the amount of the liquid phase but can accelerate densification:
thus the activated specimens were fully dense (3,20 gcm-3) after
hot pressing at 1500 ºC but degree of α→β-Si3N4 transformation was
similar to that from the powder bed sintering method (Table
2).
The phase evolution of β-Si3N4-based precursor after heat treatment
in powder bed was different from α-Si3N4-based ones, because the
initial β-Si3N4 powder was contaminated by Si2N2O and their
susceptibility for activation was different. Moreover, increase of
Si2N2O was observed after heat treatment of the activated precursor
and/or mullite formation. The unexpected deviation and mullite or
Si2N2O presence must be due to the oxidation or crystallization of
the highly defective Si-Al-O-N phase. The final amount of β-Si3N4
in the resultant ceramic was comparable after heat treatment at
1600ºC despite the starting composition. However, β-SiAlON
formation was found if measured by changes of the β-Si3N4 unit cell
parameters (Fig. 3). That
2.906
2.908
2.91
2.912
2.914
Temperature [oC]
RB
Activated
Figure 3. Development of c-parameter in β-Si3N4 unit cell vs
sintering temperature Table 1 The phase composition of the initial
A and B mixtures as batched and measured after activation
milling
Phase composition of β-sialon precursor [wt%] Precursor α-Si3N4
β-Si3N4 AlN Y2O3 Si2N2O A-initial 73 16 6 5 0
A-30 min 81,6 14,4 3,2 0,8 0 B-initial 0 84 6 5 5
B-60 min 0 96,4 0 0 3,6 Table 2 Phase assemblage [wt%] of β-sialon
precursor after sintering at 1500 ºC and 1600 ºC in powder bed. The
HP denoted specimen was hot pressed.
Temperature of sintering:1500oC α-Si3N4 β-Si3N4 Si2N2O YAG mullite
A-RB 64,2 30,4 0 5,4 0 A-30 min 35,9 64,1 0 0 0 A-30 min, HP 38,2
61,8 0 0 0 B-RB 1,3 89,7 0 9 0 B-30 min 1 94,1 1,1 0,3 3,4 B-60 min
0 91,9 8,1 0 0
Temperature of sintering: 1600oC α-Si3N4 β-Si3N4 Si2N2O YAG mullite
A-30 min 11,4 88,9 0 0 0 B-RB 0 92,4 0 7,6 0 B-30 min 0 96,5 0 0
3,5 B-60 min 0 90,8 9,2 0 0
activation but was independent on the technique of sintering (Table
2). α→β-Si3N4 transformation is believed to occur in the presence
of the liquid phase. If there is a sufficient amount of the
Y-Si-Al-O-N eutectic liquid phase then β-phase
crystallization/precipitation follows. The amount of the eutectic
liquid phase depends on temperature if the same chemical
composition is considered, but could also be affected by
amorphization of one or more components. The amount and composition
of the liquid phase at 1500 ºC in the non-activated precursor must
have been insufficient for β-Si3N4 crystallization because only
less than 1/4 of the initial silicon nitride was transformed to
β-phase and YAG presence shows silicon lacking in the liquid at
1500 ºC. On the other hand, significantly more liquid phase was
present at 1500 ºC in the activated precursor: the amount of
β-phase was two times higher in comparison to the non-activated
precursor. Application of the hot-pressing technique can not change
the amount of the liquid phase but can accelerate densification:
thus the activated specimens were fully dense (3,20 gcm-3) after
hot pressing at 1500 ºC but degree of α→β-Si3N4 transformation was
similar to that from the powder bed sintering method (Table
2).
The phase evolution of β-Si3N4-based precursor after heat treatment
in powder bed was different from α-Si3N4-based ones, because the
initial β-Si3N4 powder was contaminated by Si2N2O and their
susceptibility for activation was different. Moreover, increase of
Si2N2O was observed after heat treatment of the activated precursor
and/or mullite formation. The unexpected deviation and mullite or
Si2N2O presence must be due to the oxidation or crystallization of
the highly defective Si-Al-O-N phase. The final amount of β-Si3N4
in the resultant ceramic was comparable after heat treatment at
1600ºC despite the starting composition. However, β-SiAlON
formation was found if measured by changes of the β-Si3N4 unit cell
parameters (Fig. 3). That
2.906
2.908
2.91
2.912
2.914
Temperature [oC]
RB
Activated
Figure 3. Development of c-parameter in β-Si3N4 unit cell vs
sintering temperature Table 1 The phase composition of the initial
A and B mixtures as batched and measured after activation
milling
Phase composition of β-sialon precursor [wt%] Precursor α-Si3N4
β-Si3N4 AlN Y2O3 Si2N2O A-initial 73 16 6 5 0
A-30 min 81,6 14,4 3,2 0,8 0 B-initial 0 84 6 5 5
B-60 min 0 96,4 0 0 3,6 Table 2 Phase assemblage [wt%] of β-sialon
precursor after sintering at 1500 ºC and 1600 ºC in powder bed. The
HP denoted specimen was hot pressed.
Temperature of sintering:1500oC α-Si3N4 β-Si3N4 Si2N2O YAG mullite
A-RB 64,2 30,4 0 5,4 0 A-30 min 35,9 64,1 0 0 0 A-30 min, HP 38,2
61,8 0 0 0 B-RB 1,3 89,7 0 9 0 B-30 min 1 94,1 1,1 0,3 3,4 B-60 min
0 91,9 8,1 0 0
Temperature of sintering: 1600oC α-Si3N4 β-Si3N4 Si2N2O YAG mullite
A-30 min 11,4 88,9 0 0 0 B-RB 0 92,4 0 7,6 0 B-30 min 0 96,5 0 0
3,5 B-60 min 0 90,8 9,2 0 0
Key Engineering Materials Vol. 403 9
behavior is in contrary to the non-activated batches as any changes
of the β-Si3N4 unit cell parameters were observed after heat
treatment at the tested temperature range. It is interesting to
note that formation of β-SiAlON solid solution and degree of
substitution in the specimens from the activated powders was
closely related to temperature of the heat treatment and grew
smoothly. The final z-value was close to 0.3 which is slightly
lower than the designed 0.4 value.
Summary
Nanustructured silicon and aluminum nitrides mixture can be
successfully produced by high-energy planetary milling. XRD studies
showed substantial changes of activated precursor’s diffraction
picture. The resultant activated powder could be sintered to full
density at 1600 ºC without external nitrogen (gas) pressure. The
densification and phase transformation occurred via transient
liquid phase and was related to the extend of the precursor
amorphization during the activation milling. Silicon oxynitride
contamination should be avoided in the initial mixture.
References
[1] X. Xu, T. Nishimura, N. Hirosaki, R-J. Xie, Y. Yamamoto, H.
Tanaka : Nanotechnology, Vol. 16 (2005), p. 1569
[2] X. Xu, T. Nishimura, T. Hirosaki, R.-J. Xie, Y. Yamamoto, H.
Tanaka, J. Am. Ceram. Soc.,Vol. 88 (2005), p. 934
[3] M. Sopicka-Lizer, M. Tacula, T. Wodek, K. Rodak, M. Hüller, V.
Kochnev, E. Fokina, K. MacKenzie, J. Eur. Ceram. Soc., Vol. 28
(2008), p. 279
[4] M. Sopicka-Lizer, M. Tacula, T. Pawlik, V. Kochnev, E. Fokina,
Mat. Sci. Forum, Vol. 554 (2007), p. 59
[5] M. Herrmann, I. Schulz, I. Zalite, J. Eur. Ceram. Soc., Vol. 24
(2004), p. 3327
behavior is in contrary to the non-activated batches as any changes
of the β-Si3N4 unit cell parameters were observed after heat
treatment at the tested temperature range. It is interesting to
note that formation of β-SiAlON solid solution and degree of
substitution in the specimens from the activated powders was
closely related to temperature of the heat treatment and grew
smoothly. The final z-value was close to 0.3 which is slightly
lower than the designed 0.4 value.
Summary
Nanustructured silicon and aluminum nitrides mixture can be
successfully produced by high-energy planetary milling. XRD studies
showed substantial changes of activated precursor’s diffraction
picture. The resultant activated powder could be sintered to full
density at 1600 ºC without external nitrogen (gas) pressure. The
densification and phase transformation occurred via transient
liquid phase and was related to the extend of the precursor
amorphization during the activation milling. Silicon oxynitride
contamination should be avoided in the initial mixture.
References
[1] X. Xu, T. Nishimura, N. Hirosaki, R-J. Xie, Y. Yamamoto, H.
Tanaka : Nanotechnology, Vol. 16 (2005), p. 1569
[2] X. Xu, T. Nishimura, T. Hirosaki, R.-J. Xie, Y. Yamamoto, H.
Tanaka, J. Am. Ceram. Soc.,Vol. 88 (2005), p. 934
[3] M. Sopicka-Lizer, M. Tacula, T. Wodek, K. Rodak, M. Hüller, V.
Kochnev, E. Fokina, K. MacKenzie, J. Eur. Ceram. Soc., Vol. 28
(2008), p. 279
[4] M. Sopicka-Lizer, M. Tacula, T. Pawlik, V. Kochnev, E. Fokina,
Mat. Sci. Forum, Vol. 554 (2007), p. 59
[5] M. Herrmann, I. Schulz, I. Zalite, J. Eur. Ceram. Soc., Vol. 24
(2004), p. 3327
10 SiAlONs and Non-oxides
New Green Phosphor Ba3Si6O12N2:Eu for White LED: Crystal Structure
and Optical Properties
Masayoshi Mikami1,a, Satoshi Shimooka1, Kyota Uheda1, Hiroyuki
Imura1 and Naoto Kijima1
1 Mitsubishi Chemical Group Science and Technology Research Center,
Inc. 1000, Kamoshida-cho, Aoba-ku, Yokohama, 227-8502, Japan
[email protected]
Abstract. A new oxynitride, Ba3Si6O12N2, has been synthesized. The
crystal structure has been successfully determined by close
collaboration between experiment and first-principles calculation.
This compound doped with Eu exhibits intense green
photoluminescence with high color purity under near-ultraviolet to
blue light excitation; in particular, it has much less thermal
quenching than (Ba,Sr,Eu)2SiO4. Thus (Ba,Eu)3Si6O12N2 appears
promising green phosphor for white LED backlight for display. The
atomic/electronic structure is discussed in comparison with
Ba3Si6O9N4, which could not become efficient phosphor by doping Eu
due to strong thermal quenching at room temperature.
Introduction
In the white LED market, the share of green and red phosphors has
been gradually increasing to achieve good color reproducibility for
liquid crystal display (LCD) backlight. Promising candidates that
satisfy high color purity are, e.g., (M,Eu)2SiO4[1] and β
-sialon:Eu[2] for green phosphor, and (M,Eu)AlSiN3[3] for red
phosphor (M: alkaline-earth element). The green phosphors still
confront some difficulties; (M,Eu)2SiO4 has strong thermal
quenching[1], whereas the synthesis of efficient β -sialon:Eu is
not facile. We have thus explored M-Si-O-N system for such green
phosphors. Although MSi2O2N2[4] and Ba3Si6O9N4[5] have been known,
it has not been reported that the both compounds doped with
rare-earth element work as efficient green phosphors with high
color purity.
In the present work, we have successfully synthesized a new green
phosphor, (Ba,Eu)3Si6O12N2[6]. We have also identified the crystal
structure by a new protocol combining X-ray/neutron powder
diffraction analysis with first-principles study. Although
Ba3Si6O9N4 looks similar to Ba3Si6O12N2 from the viewpoint of
crystal structure and chemical formula, their optical properties
are quite different; (Ba,Eu)3Si6O9N4 exhibits blue-green
photoluminescence (PL) only at low temperatures (i.e. little
luminescence at room temperature (R.T.)), whereas (Ba,Eu)3Si6O12N2
exhibits intense green PL at R.T. with thermal quenching smaller
than (Ba,Sr,Eu)2SiO4. In this paper, the crystal structure and
optical properties of the new green phosphor are described. The
interpretation of the origin of the different optical properties
between (Ba,Eu)3Si6O12N2 and (Ba,Eu)3Si6O9N4 are briefly
explained.
Experimental and theoretical procedure
A single phase of the new compound was prepared by conventional
method [6]. The X-ray diffraction (XRD) indicated that the compound
should be a new crystal phase. The crystal symmetry (trigonal) and
approximate lattice constants (a=7.48(1)Å, c=6.47(1)Å) were
obtained by TEM. From the chemical analysis (O-rich) and weight
density (= 4.13 g/cm3) indicating Z=1, we assumed that the compound
has crystal structure close to Ba3Si6O9N4 (P3, a = 7.249(1)Å, c =
6.784(2)Å, Z=1). Among possible compositions, Ba3Si6OxNy (x > y)
with (x, y)=(6,6),(9,4),(12,2), only Ba3Si6O12N2 had not been
reported. Thus the crystal structure of Ba3Si6O12N2 was
approximately solved as P3 by a direct method from the XRD. The
geometry was then optimized by ab initio calculation based on
density
New Green Phosphor Ba3Si6O12N2:Eu for White LED: Crystal Structure
and Optical Properties
Masayoshi Mikami1,a, Satoshi Shimooka1, Kyota Uheda1, Hiroyuki
Imura1 and Naoto Kijima1
1 Mitsubishi Chemical Group Science and Technology Research Center,
Inc. 1000, Kamoshida-cho, Aoba-ku, Yokohama, 227-8502, Japan
[email protected]
Abstract. A new oxynitride, Ba3Si6O12N2, has been synthesized. The
crystal structure has been successfully determined by close
collaboration between experiment and first-principles calculation.
This compound doped with Eu exhibits intense green
photoluminescence with high color purity under near-ultraviolet to
blue light excitation; in particular, it has much less thermal
quenching than (Ba,Sr,Eu)2SiO4. Thus (Ba,Eu)3Si6O12N2 appears
promising green phosphor for white LED backlight for display. The
atomic/electronic structure is discussed in comparison with
Ba3Si6O9N4, which could not become efficient phosphor by doping Eu
due to strong thermal quenching at room temperature.
Introduction
In the white LED market, the share of green and red phosphors has
been gradually increasing to achieve good color reproducibility for
liquid crystal display (LCD) backlight. Promising candidates that
satisfy high color purity are, e.g., (M,Eu)2SiO4[1] and β
-sialon:Eu[2] for green phosphor, and (M,Eu)AlSiN3[3] for red
phosphor (M: alkaline-earth element). The green phosphors still
confront some difficulties; (M,Eu)2SiO4 has strong thermal
quenching[1], whereas the synthesis of efficient β -sialon:Eu is
not facile. We have thus explored M-Si-O-N system for such green
phosphors. Although MSi2O2N2[4] and Ba3Si6O9N4[5] have been known,
it has not been reported that the both compounds doped with
rare-earth element work as efficient green phosphors with high
color purity.
In the present work, we have successfully synthesized a new green
phosphor, (Ba,Eu)3Si6O12N2[6]. We have also identified the crystal
structure by a new protocol combining X-ray/neutron powder
diffraction analysis with first-principles study. Although
Ba3Si6O9N4 looks similar to Ba3Si6O12N2 from the viewpoint of
crystal structure and chemical formula, their optical properties
are quite different; (Ba,Eu)3Si6O9N4 exhibits blue-green
photoluminescence (PL) only at low temperatures (i.e. little
luminescence at room temperature (R.T.)), whereas (Ba,Eu)3Si6O12N2
exhibits intense green PL at R.T. with thermal quenching smaller
than (Ba,Sr,Eu)2SiO4. In this paper, the crystal structure and
optical properties of the new green phosphor are described. The
interpretation of the origin of the different optical properties
between (Ba,Eu)3Si6O12N2 and (Ba,Eu)3Si6O9N4 are briefly
explained.
Experimental and theoretical procedure
A single phase of the new compound was prepared by conventional
method [6]. The X-ray diffraction (XRD) indicated that the compound
should be a new crystal phase. The crystal symmetry (trigonal) and
approximate lattice constants (a=7.48(1)Å, c=6.47(1)Å) were
obtained by TEM. From the chemical analysis (O-rich) and weight
density (= 4.13 g/cm3) indicating Z=1, we assumed that the compound
has crystal structure close to Ba3Si6O9N4 (P3, a = 7.249(1)Å, c =
6.784(2)Å, Z=1). Among possible compositions, Ba3Si6OxNy (x > y)
with (x, y)=(6,6),(9,4),(12,2), only Ba3Si6O12N2 had not been
reported. Thus the crystal structure of Ba3Si6O12N2 was
approximately solved as P3 by a direct method from the XRD. The
geometry was then optimized by ab initio calculation based on
density
Key Engineering Materials Vol. 403 (2009) pp 11-14 © (2009) Trans
Tech Publications, Switzerland
doi:10.4028/www.scientific.net/KEM.403.11
functional theory(DFT)[7]. It turned out that the calculated
geometry has inversion symmetry, which could not be detected by the
XRD analysis. The optimized crystal structure was used for initial
inputs for Rietveld analysis of the X-ray/neutron diffraction. The
structure was finally obtained as P3 .
Results and Discussions
The crystal structure of Ba3Si6O12N2 is illustrated in Fig. 1. The
crystallographic data are summarized in Table 1. The calculated
lattice constants overestimate the experimental data due to the
approximation, generalized-gradient correction (GGA), in the DFT.
It is noted that the calculated parameters were used for the
Rieltveld analysis of XRD/neutron diffraction (not given
here).
Table 1: Crystallographic data of Ba3Si6O12N2 (experimental and
theoretical data) System (Space group, No.) Trigonal (P 3 , No.
147)
Experiment (from XRD) Theory Lattice parameters / Å a=7.5046(8),
c=6.4703(5) a=7.59684, c=6.57487
Atomic coordinates Label Wyckoff-position x y z x y z Ba1 1a 0 0 0
0 0 0 Ba2 2d 1/3 2/3 0.1039(2) 1/3 2/3 0.10060 Si1 6g 0.2366(6)
0.8310(6) 0.6212(8) 0.23594 0.82847 0.60978 N1 2d 1/3 2/3 0.568(3)
1/3 2/3 0.56311 O1 6g 0.356(2) 0.295(2) 0.173(1) 0.36096 0.29646
0.17190 O2 6g 0.000(1) 0.681(1) 0.589(2) -0.01462 0.68138
0.58846
The fused rings-sheet, 2 ∞ [(Si ]4[
6 O ]2[ 6 N ]3[
2 )O ]1[ 6 ] −6 , is composed of 8-membered Si-(O,N) and
12-membered Si-O rings. The compound is built up of corner sharing
SiO3N tetrahedra forming corrugated layers between which the Ba2+
ions are located. The Ba2+ ions occupy two different
crystallographic sites; one is trigonal anti-prism (distorted
octahedron) with six oxygen atoms, and the other is trigonal
anti-prism with six oxygen atoms, further capped with a nitrogen
atom (Fig.1(c)).
The crystal structure and chemical formula of Ba3Si6O12N2 appears
close to Ba3Si6O9N4. We review the crystal structure of Ba3Si6O9N4
by following Ref. 5. The compound is composed by corner sharing
SiO2N2 tetrahedra forming corrugated layers between which the Ba2+
ions are located. The Ba2+ ions occupy three different
crystallographic sites; two of them are trigonal anti-prisms with
six oxygen atoms, and the other is trigonal anti-prism with six
oxygen atoms, further capped with a nitrogen atom (Fig.2). The Ba
clusters in Ba3Si6O9N4 looks similar to those in Ba3Si6O12N2.
Still, a main difference is the Ba-N distance: about 3.2 Å in
Ba3Si6O9N4 whereas about 3.0 Å in Ba3Si6O12N2. It is also
noteworthy that the host absorption band of the Ba3Si6O9N4 appears
lower in energy than that of Ba3Si6O12N2, although the both
absorption were observed below 300 nm from the diffuse reflectance
spectra. This is also implied by our band structure calculation;
the computed band gap of Ba3Si6O12N2 is 4.63 eV whereas that of
Ba3Si6O9N4
is 4.37 eV. Although it is known that band gap calculated within
GGA will be underestimated, computed band gaps can be qualitatively
compared.
Figure 1: Projections of the unit cell of Ba3Si6O12N2 viewed along
the c axis (a) and b axis (b), and two coordination environments
around Ba2+ ion (c); the clusters are defined within 3.2 Å.
functional theory(DFT)[7]. It turned out that the calculated
geometry has inversion symmetry, which could not be detected by the
XRD analysis. The optimized crystal structure was used for initial
inputs for Rietveld analysis of the X-ray/neutron diffraction. The
structure was finally obtained as P3 .
Results and Discussions
The crystal structure of Ba3Si6O12N2 is illustrated in Fig. 1. The
crystallographic data are summarized in Table 1. The calculated
lattice constants overestimate the experimental data due to the
approximation, generalized-gradient correction (GGA), in the DFT.
It is noted that the calculated parameters were used for the
Rieltveld analysis of XRD/neutron diffraction (not given
here).
Table 1: Crystallographic data of Ba3Si6O12N2 (experimental and
theoretical data) System (Space group, No.) Trigonal (P 3 , No.
147)
Experiment (from XRD) Theory Lattice parameters / Å a=7.5046(8),
c=6.4703(5) a=7.59684, c=6.57487
Atomic coordinates Label Wyckoff-position x y z x y z Ba1 1a 0 0 0
0 0 0 Ba2 2d 1/3 2/3 0.1039(2) 1/3 2/3 0.10060 Si1 6g 0.2366(6)
0.8310(6) 0.6212(8) 0.23594 0.82847 0.60978 N1 2d 1/3 2/3 0.568(3)
1/3 2/3 0.56311 O1 6g 0.356(2) 0.295(2) 0.173(1) 0.36096 0.29646
0.17190 O2 6g 0.000(1) 0.681(1) 0.589(2) -0.01462 0.68138
0.58846
The fused rings-sheet, 2 ∞ [(Si ]4[
6 O ]2[ 6 N ]3[
2 )O ]1[ 6 ] −6 , is composed of 8-membered Si-(O,N) and
12-membered Si-O rings. The compound is built up of corner sharing
SiO3N tetrahedra forming corrugated layers between which the Ba2+
ions are located. The Ba2+ ions occupy two different
crystallographic sites; one is trigonal anti-prism (distorted
octahedron) with six oxygen atoms, and the other is trigonal
anti-prism with six oxygen atoms, further capped with a nitrogen
atom (Fig.1(c)).
The crystal structure and chemical formula of Ba3Si6O12N2 appears
close to Ba3Si6O9N4. We review the crystal structure of Ba3Si6O9N4
by following Ref. 5. The compound is composed by corner sharing
SiO2N2 tetrahedra forming corrugated layers between which the Ba2+
ions are located. The Ba2+ ions occupy three different
crystallographic sites; two of them are trigonal anti-prisms with
six oxygen atoms, and the other is trigonal anti-prism with six
oxygen atoms, further capped with a nitrogen atom (Fig.2). The Ba
clusters in Ba3Si6O9N4 looks similar to those in Ba3Si6O12N2.
Still, a main difference is the Ba-N distance: about 3.2 Å in
Ba3Si6O9N4 whereas about 3.0 Å in Ba3Si6O12N2. It is also
noteworthy that the host absorption band of the Ba3Si6O9N4 appears
lower in energy than that of Ba3Si6O12N2, although the both
absorption were observed below 300 nm from the diffuse reflectance
spectra. This is also implied by our band structure calculation;
the computed band gap of Ba3Si6O12N2 is 4.63 eV whereas that of
Ba3Si6O9N4
is 4.37 eV. Although it is known that band gap calculated within
GGA will be underestimated, computed band gaps can be qualitatively
compared.
Figure 1: Projections of the unit cell of Ba3Si6O12N2 viewed along
the c axis (a) and b axis (b), and two coordination environments
around Ba2+ ion (c); the clusters are defined within 3.2 Å.
12 SiAlONs and Non-oxides
Next, we have examined the optical properties of (Ba,Eu)3Si6O12N2
and (Ba,Eu)3Si6O9N4. In the
both compounds, Eu2+ ions are supposed to occupy the Ba-sites
(Fig.1(c) and Fig.2). The spectra of PL and PL excitation (PLE)
have been measured from liquid-He temperature to R.T. The details
will be discussed elsewhere [8], so that we briefly introduce the
results here. Under the near-UV to blue light irradiation,
(Ba,Eu)3Si6O12N2 exhibited intense broad green emission spectrum
(PL peak at about 530 nm, with full width at half maximum of 68nm)
at R.T. As for the PLE of (Ba,Eu)3Si6O12N2, the broad excitation
bands were observed in wavelengths ranging from 200 to 500 nm. The
PL/PLE should originate from the allowed transition from 4f7 grand
state to 5d state of Eu2+, because the host absorption is below 300
nm. It is underscored that the phosphor will be suitable for LED
backlight in LCD owing to its high green color purity with the CIE
color coordinates (x,y)=(0.274, 0.644) similar to those of
(Ba,Sr,Eu)2SiO4. It was confirmed that thermal quenching of
(Ba,Eu)3Si6O12N2 is much smaller than (Ba,Sr,Eu)2SiO4; the emission
intensity of (Ba,Eu)3Si6O12N2 at 100ºC was about 90% of that
measured at R.T., whereas the emission intensity of (Ba,Sr,Eu)2SiO4
at 100ºC was about 75% of that at R.T. Thus (Ba,Eu)3Si6O12N2
appears a promising green phosphor for LCD backlight use.
On the other hand, the PL of (Ba,Eu)3Si6O9N4, broad blue-green
emission spectra, was observed only at low temperature (lower than
RT) due to its strong thermal quenching. The PL peak was about 480
nm, shorter than the PL peak of (Ba,Eu)3Si6O12N2 (about 530 nm).
The excitation bands ranging from 200 to 440 nm was narrower than
that of (Ba,Eu)3Si6O12N2. The PL and PLE also originate from the
allowed transition from 4f7 grand state to 5d state of Eu2+. We did
not see large difference in Stokes shift for the both compounds.
The question is why the PL/PLE spectra and the thermal quenching
behaviors were quite different for (Ba,Eu)3Si6O12N2 and
(Ba,Eu)3Si6O9N4, even with the similar structure of the Ba-clusters
(i.e. the trigonal symmetry at the Ba centers, Fig.1(c) and
Fig.2).
Figure 2: Coordination environments around Ba2+ ion (defined within
3.2 Å) in Ba3Si6O9N4 [5].
Figure 3: Schematic illustration of level energies in Eu2+doped
barium silicon oxynitrides [conduction band (C.B.) , valence band
(V.B.) and photoluminescence (PL)] . Dotted lines and broken lines
denote crystal field splitting due to Eu-O ligand and Eu-N ligand,
respectively. The autoionization process (Eu2+àEu3++e-) is denoted
in Eu-doped Ba3Si6O9N4 (right). Stokes shifts and details of the
trigonal distortion splits of Eu-d states are not drawn for
simplification.
Next, we have examined the optical properties of (Ba,Eu)3Si6O12N2
and (Ba,Eu)3Si6O9N4. In the
both compounds, Eu2+ ions are supposed to occupy the Ba-sites
(Fig.1(c) and Fig.2). The spectra of PL and PL excitation (PLE)
have been measured from liquid-He temperature to R.T. The details
will be discussed elsewhere [8], so that we briefly introduce the
results here. Under the near-UV to blue light irradiation,
(Ba,Eu)3Si6O12N2 exhibited intense broad green emission spectrum
(PL peak at about 530 nm, with full width at half maximum of 68nm)
at R.T. As for the PLE of (Ba,Eu)3Si6O12N2, the broad excitation
bands were observed in wavelengths ranging from 200 to 500 nm. The
PL/PLE should originate from the allowed transition from 4f7 grand
state to 5d state of Eu2+, because the host absorption is below 300
nm. It is underscored that the phosphor will be suitable for LED
backlight in LCD owing to its high green color purity with the CIE
color coordinates (x,y)=(0.274, 0.644) similar to those of
(Ba,Sr,Eu)2SiO4. It was confirmed that thermal quenching of
(Ba,Eu)3Si6O12N2 is much smaller than (Ba,Sr,Eu)2SiO4; the emission
intensity of (Ba,Eu)3Si6O12N2 at 100ºC was about 90% of that
measured at R.T., whereas the emission intensity of (Ba,Sr,Eu)2SiO4
at 100ºC was about 75% of that at R.T. Thus (Ba,Eu)3Si6O12N2
appears a promising green phosphor for LCD backlight use.
On the other hand, the PL of (Ba,Eu)3Si6O9N4, broad blue-green
emission spectra, was observed only at low temperature (lower than
RT) due to its strong thermal quenching. The PL peak was about 480
nm, shorter than the PL peak of (Ba,Eu)3Si6O12N2 (about 530 nm).
The excitation bands ranging from 200 to 440 nm was narrower than
that of (Ba,Eu)3Si6O12N2. The PL and PLE also originate from the
allowed transition from 4f7 grand state to 5d state of Eu2+. We did
not see large difference in Stokes shift for the both compounds.
The question is why the PL/PLE spectra and the thermal quenching
behaviors were quite different for (Ba,Eu)3Si6O12N2 and
(Ba,Eu)3Si6O9N4, even with the similar structure of the Ba-clusters
(i.e. the trigonal symmetry at the Ba centers, Fig.1(c) and
Fig.2).
Figure 2: Coordination environments around Ba2+ ion (defined within
3.2 Å) in Ba3Si6O9N4 [5].
Figure 3: Schematic illustration of level energies in Eu2+doped
barium silicon oxynitrides [conduction band (C.B.) , valence band
(V.B.) and photoluminescence (PL)] . Dotted lines and broken lines
denote crystal field splitting due to Eu-O ligand and Eu-N ligand,
respectively. The autoionization process (Eu2+àEu3++e-) is denoted
in Eu-doped Ba3Si6O9N4 (right). Stokes shifts and details of the
trigonal distortion splits of Eu-d states are not drawn for
simplification.
Key Engineering Materials Vol. 403 13
From the observation of the Ba-cluster models, we may suppose that
the excited states of Eu is composed as the superposition of the
Eu-O and the Eu-N crystal field splitting; if E-N ligand would not
affect PL/PLE, (Ba,Eu)3Si6O12N2 and (Ba,Eu)3Si6O9N4 should have
exhibited similar PL/PLE properties due to the same local symmetry
with the similar Ba(Eu)-O lengths. Since we see the difference in
the above, the Eu-N ligand should be effective. Thus we focus on
the trigonal anti-prisms with six O atoms plus one N atom (Fig.3).
Our PL/PLE results suggest that the Eu d-states from Eu-N ligand
determine the lowest excited state, because of the nephelauxetic
effect originating from anion polarizability; note that N3- is more
polarizable/covalent than O2-. Since the Ba-N length in
(Ba,Eu)3Si6O12N2 (about 3.0Å) is shorter than that in
(Ba,Eu)3Si6O9N4 (about 3.2Å), the Eu-N crystal field splitting in
(Ba,Eu)3Si6O12N2 is expected to be wider than that in
(Ba,Eu)3Si6O9N4.
With the above assumption in mind, we are ready to explain our
experimental results. The blue shift of the PL peak of 530 nm
((Ba,Eu)3Si6O12N2) to 480 nm ((Ba,Eu)3Si6O9N4) originates from the
higher Eu-d states of (Ba,Eu)3Si6O9N4 than that of
(Ba,Eu)3Si6O12N2. Remembering the smaller band gap of
(Ba,Eu)3Si6O9N4, we suppose that the lowest Eu-d states are so
close to the conduction bands in (Ba,Eu)3Si6O9N4 that the
photo-excited 5d-electrons of Eu2+ can be thermally ionized to the
conduction bands at R.T.[9] The broadness of the excitation bands
may be related to the superposition of the Eu-O and the Eu-N
crystal field splittings. Incidentally, the narrowness of the PL
may be reflected by the anisotropic structure of the trigonal
anti-prisms with six O atoms plus one atom.
Summary
A new oxynitride, Ba3Si6O12N2, has been synthesized and its crystal
structure has been determined. It means that we have a series of
composition, Ba3Si6O6N6(=BaSi2O2N2), Ba3Si6O9N4, Ba3Si6O12N2, and
Ba3Si6O15(=BaSi2O5), by substituting N2 with O3 formally.
(Ba,Eu)3Si6O12N2 exhibits efficient green photoluminescence with
high color purity under InGaN diode irradiation; in particular, it
has much less thermal quenching than the other green phosphor,
(Ba,Sr,Eu)2SiO4. Although the crystal structure and chemical
formula appears close to Ba3Si6O9N4, their optical properties and
thermal quenching behaviors are quite different. Stronger thermal
quenching in (Ba,Eu)3Si6O9N4 may be ascribed to smaller band gap
and longer Ba-N distance (i.e. smaller crystal field
splitting).
Acknowledgment: We wish to thank Dr. M. Takashima for the TEM
analysis, Dr. Y. Sasaki for the PL/PLE measurement (liquid-He
temperature to room temperature), and Mr. K. Horibe for LED
fabrication. The neutron diffraction experiment was performed at
Japan Atomic Energy Agency.
References
[1] T. L. Barry : J. Electrochem. Soc. Vol. 115 (1968), p.
1181.
[2] N. Hirosaki et al., Appl. Phys. Lett.86 (2005), p.211905.
[3] K. Uheda et al., Electrochem. Solid State Lett. Vol. 9, (2006),
p. H22.
[4] R. Müller-Mach et al., phys. stat. sol. (a) Vol. 202 (2005), p.
1727.
[5] F. Stadler and W. Schnick, Z. Anorg. Allg. Chem. Vol. 632
(2006), p. 949.
[6] S. Shimooka, K. Uheda, M. Mikami, N. Kijima, H. Imura, and K.
Horibe, PCT WO2007/088966.
[7] The ABINIT code is a common project of the Université
Catholique de Louvain, Corning Incorporated, and other contributors
(URL http://www.abinit.org).