Novel Routes for Ceramics Synthesis and Processing: Proceedings of the 12th International Ceramics Congress, Part B

Table of contents :
12th INTERNATIONAL CERAMICS CONGRESS PART B......Page 2
Preface......Page 5
Committees......Page 6
Table of Contents......Page 7
Spray Freeze Granulation of Nano Powders for Die Pressing......Page 11
New Synthesis Process of Li, Na and K Niobates from Metal Alkoxides......Page 17
Novel Sol-Gel Synthesis of LiMn2O4 and LiNixCo1-xO2 Powders......Page 24
Synthesis of Alumina and Aluminum Nitride Layers on a Graphitic Substrate via a Sol-Gel Route......Page 34
Synthesis of Monodispersed Plate-Like CeO2 Particles by Precipitation Process in Sodium Hydrogen Carbonate Solution......Page 40
Co-Doping Effect of Metal Ion on the Visible Light Responsive Photocatalytic Properties of Nitrogen Doped Titanium Dioxide......Page 46
Tailored Silica Based Xerogels and Aerogels for Insulation in Space Environments......Page 51
In Situ TEM Observation of Crystallization Process for LiNbO3 and NaNbO3......Page 57
Microwave Assisted Solvothermal Synthesis and Visible Light Photocatalytic Properties of Nb and N Co-Doped SrTiO3 Nanoparticles......Page 62
Numerical Simulation of Spark Plasma Sintering......Page 68
Densification Mechanism of MgAl2O4 Spinel during Spark-Plasma-Sintering......Page 72
Production and Characterization of Boron Carbide – Titanium Diboride Ceramics by Spark Plasma Sintering Method......Page 78
The Effects of Codoping Y2O3 on MgO Doped Spark Plasma Sintered Al2O3......Page 84
Spark Plasma Sintering of Boron Carbide and Effects of Various Additives on Sintering and Material Properties......Page 89
Microwave Absorbency Change of Zirconia Powder and Fiber during Vacuum Heating......Page 95
Microwave Assisted Reaction Sintering of ZrSiO4/α -Al2O3 Mixtures......Page 101
Hybrid Foams, Colloids and Beyond: Advanced Ceramics through Integrative Chemistry......Page 107
Panoscopic Assembling of Ceramic Materials for High Performance UV-Ray Shielding Application......Page 117
Lightweight Hybrid Foam with Dimensional Stability......Page 124
Ceramic/Polymeric Hybrids with Reduced Coefficients of Thermal Expansion......Page 130
Dimension- and Direction-Controlled Gold Nanorods Deposited in Ordered Mesoporous Silica......Page 136
Heterogeneous Sol-Gel Systems – Derived Ceramics......Page 141
Smart Processing for Ceramics Structure Tectonics: Fabrication of Dielectric Micro Patterns for Artificial Photosynthesis in Terahertz Wave Regions by Using Stereolithography......Page 151
Clay Aerogel Composite Materials......Page 157
Synthesis and Characterization of Mesoporous Hydroxyapatite......Page 162
New Methodology in Modeling Ceramics......Page 168
Influence of Binder on Porous Ceramic Properties Prepared by Polymeric Sponge Method......Page 174
Fabrication of Porous Silicon Nitride by Sacrificing Template Method......Page 180
Aluminum Oxide Ceramics Obtained by Commercial Starch Consolidation with Gradient Porosity......Page 185
Processing of Municipal Solid Waste (MSW) Fly Ash into an Environmentally Stable and Safe Material......Page 191
Solution Combustion as a Promising Method for the Synthesis of Nanomaterials......Page 197
Microwave Activated Combustion Synthesis and Compaction in Separate E and H Fields: Numerical Simulation and Experimental Results......Page 207
Use of Electrothermal Explosion and Electro-Thermal Analyser (ETA-100) for the Study the Kinetics of Fast High-Temperature Reactions in SHS-Ceramic Systems......Page 213
Simulation of Gasless Combustion of Mechanically Activated Solid Powder Mixtures......Page 223
Macrokinetics for Macrostructure Forming of a Product in Self-Propagating High-Temperature Synthesis......Page 232
Past and Current Accomplishments in Production of Ceramic Powders and Structures by Self-Propagating High-Temperature Synthesis Method......Page 238
Carbon Combustion Synthesis of Ceramic Oxide Nanopowders......Page 246
Double SHS of W2B5 Powder from CaWO4 and B2O3......Page 256
Production of Zirconium Diboride Powder by Self Propagating High Temperature Synthesis......Page 261
Utilization of NbC Nanoparticles Obtained by Reactive Milling in Production of Alumina Niobium Carbide Nanocomposites......Page 267
Composites Produced by SHS Method – Current Development and Future Trends......Page 273
Self-Propagating High-Temperature Synthesis of Iron- and Copper-Matrix Cermets......Page 283
Sintering and Hot-Pressing of Ti2AlC Obtained by SHS Process......Page 292
Catalytic Properties of the SHS Products - Review......Page 297
Porous SHS-Ceramics......Page 307
Mass-Forced SHS Technology of Ceramic Materials......Page 312
SHS Refractory Materials “Furnon” and their Practical Implementations in Kazakhstan and Russia......Page 322
The Potential of Spark Plasma Sintering (SPS) Method for the Fabrication on an Industrial Scale of Functionally Graded Materials......Page 332
Comparison of Microwave and Conventional Sintering of LHA Ceramics and Functionally Graded Alumina-LHA Ceramics......Page 342
Fabrication of Functionally Graded ZTA Ceramics Using a Novel Combination of Freeze Casting and Electrophoretic Deposition......Page 350
Effects of Strain-Graded Plastic Deformation on Mechanical Properties of Metals......Page 358
Defect Crystal Structure of Low Temperature Modifications of Li2MO3 (M=Ti, Sn) and Related Hydroxides......Page 362
Atomic and Electronic Structure of Zinc and Copper Pyrovanadates with Negative Thermal Expansion......Page 368
Layered Alumina Ceramics with Porosity Steps......Page 374
Fabrication of Porous Intermetallic Thick Films by Metallic Powder-Liquid Reaction......Page 380
High-Strength Reaction-Sintered Silicon Carbide for Large-Scale Mirrors - Effect of Surface Oxide Layer on Bending Strength......Page 384
Development of Functionally Graded Coating Based Plasma Facing Materials for Fusion Reactor......Page 393
Relationship Between Microstructure and Hardness of ZrN/TiN Multi-Layers with Various Bilayer Thicknesses......Page 402
Mechanical Properties of Si3N4-SiC Composites Sintered by HPHT Method......Page 406
Effect of the Hydrothermal Heat Treatment Conditions of Titanium Substrates on the Bio-Mimetically Grown “Bone-Like” Apatite Coatings......Page 412
Highly Porous Hydroxyapatite Ceramics for Engineering Applicatios......Page 418
Lithium Disilicate Glass-Ceramic Obtained from Rice Husk-Based Silica......Page 424
Reactive Milling and Mechanical Alloying in Electroceramics......Page 430
Crystal Growth of Calcite Nano-Plate by Alternate Soaking Method, Using CDS Crystals......Page 435
Keywords Index......Page 441
Authors Index......Page 445

Citation preview

12th INTERNATIONAL CERAMICS CONGRESS PART B

12th INTERNATIONAL CERAMICS CONGRESS Proceedings of the 12 th International Ceramics Congress, part of CIMTEC 201012 th International Ceramics Congress and 5th Forum on New Materials Montecatini Terme, Italy, June 6-11, 2010

PART B

including:

Symposium CB – Novel Routes for Ceramics Synthesis and Processing

Edited by Pietro VINCENZINI World Academy of Ceramics and National Research Council, Italy Co-edited by Ralf RIEDEL Technical University of Darmstadt, Germany Alexander G. MERZHANOV ISMAN, Russia Chang-Chun GE University of Science and Technology Beijing, China

TRANS TECH PUBLICATIONS LTD Switzerland • UK • USA on behalf of TECHNA GROUP Faenza • Italy

Copyright  2010 Trans Tech Publications Ltd, Switzerland Published by Trans Tech Publications Ltd., on behalf of Techna Group Srl, Italy All rights reserved. No part of this book may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, recording, photocopying or otherwise, without the prior written permission of the Publisher. No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein.

Trans Tech Publications Ltd Laubisrutistr. 24 CH-8712 Stafa-Zuerich Switzerland http://www.ttp.net EAN: 9783908158301

Volume 63 of Advances in Science and Technology ISSN 1661-819X

Full text available online at http://www.scientific.net The listing of the other Volumes (1-61) of the Series "Advances in Science and Technology" are available at TECHNA GROUP website: http://www.technagroup.it

Distributed worldwide by

and in the Americas by

Trans Tech Publications Ltd Laubisrutistr. 24 CH-8712 Stafa-Zuerich Switzerland

Trans Tech Publications Inc. PO Box 699, May Street Enfield, NH 03748 USA

Fax: +41 (44) 922 10 33 e-mail: [email protected]

Phone: +1 (603) 632-7377 Fax: +1 (603) 632-5611 e-mail: [email protected]

PREFACE CIMTEC 2010 was held in Montecatini Terme, Italy on June 6-18, 2010. This high qualitative and comprehensive congressional event, similarly to the previous editions, has been designed to encompass and derive synergism from a broad interdisciplinarity network capable of offering opportunities for identifying and exploring new directions for research and production. The above based on the view that ongoing and future innovations require at an ever increasing extent a complex array of interconnections among scientific research, innovating technology and industrial infrastructure. CIMTEC 2010 consisted of two major events: the 12th INTERNATIONAL CERAMICS CONGRESS (June 611, 2010) and the 5th FORUM ON NEW MATERIALS (June 13-18, 2010). The World Academy of Ceramics and the International Ceramic Federation (ICF) acted as principal endorsers for the first one, and the International Union of Materials Research Societies (IUMRS) for the FORUM. The 12th INTERNATIONAL CERAMICS CONGRESS included 12 International Symposia, two Focused Sessions and two Serial International Conferences (“Disclosing Materials at Nanoscale” and “Advanced Inorganic Fibre Composites for Structural and Thermal Management Applications”) which covered recent progress in almost all relevant fields of ceramics science and technology. The 5th FORUM ON NEW MATERIALS consisted of 11 International Symposia primarily concerned with energy technologies, one Focused Session and two Serial International Conferences (“Science and Engineering of Novel Superconductors” and “Medical Applications of Novel Biomaterials and Nano-biotechnology”). A balanced, high quality programme of invited and contributed papers resulted from the over one thousand and seven hundred scientific and technical contributions effectively presented during the working days to a large international audience coming from fifty-seven countries throughout the world. The 15 volumes which constitute the Official Proceedings of CIMTEC 2010 (10 for the Ceramics Congress, 5 for the Forum) include a selection of the papers presented. Having most of them been written by authors whose mother tongue is not English, considerable revision of the original texts was often required. The partial reworking of several papers and sometimes even complete rewriting was needed to make clear work valid as regards the technical content but difficult to understand because of lack of proficiency in the English language. Even so, in order to allow the scientific and technical community to have access to the proceedings volumes within a reasonable length of time, compromise was necessary in regard to the quality of writing, and papers containing language imperfections were considered acceptable provided that their technical content was adequate and easily understandable. The Editor, who also acted as the Chairman of CIMTEC 2010, would like to express his sincere appreciation to all the Institutions and Professional Organizations involved in the congress, to the members of the International Advisory Committees, the National Coordinating Committees, the Co-Chairs Prof. Akio Makishima (Japan) for the INTERNATIONAL CERAMICS CONGRESS and Prof. Robert P.H. Chang (USA) for the FORUM ON NEW MATERIALS, the Programme Chairs, the Lecturers, the technical staff of Techna Group, and to the many others who directly or indirectly contributed to the organization. Indeed it was mainly through the involvement of the above bodies and individuals, and the active participation of most internationally qualified experts from major academic and government research institutes and industrial R&D centers that a very valuable scientific programme could be arranged. It is therefore expected for the Proceedings of CIMTEC 2010-12th INTERNATIONAL CERAMICS CONGRESS & 5th FORUM ON NEW MATERIALS to constitute a further valuable contribution to the literature in the field. P. VINCENZINI World Academy of Ceramics Emeritus Research Manager National Research Council of Italy

12th INTERNATIONAL CERAMICS CONGRESS Chairman Pietro VINCENZINI, Italy Co-Chair Akio MAKISHIMA, Japan

Symposium CB – Novel Routes for Ceramics Synthesis and Processing Programme Chair Ralf RIEDEL, Germany Members Dinesh K. Agrawal, USA David Avnir, Israel Florence Babonneau, France Joachim Bill, Germany J.P.G. Binner, UK Christian Bonhomme, France Raj Bordia, USA Jin-Ho Choy, Korea Mark M. De Guire, USA Etienne Duguet, France Mohan Edirisinghe, UK Nahum Frage, Israel Takashi Goto, Japan Peter Greil, Germany Keijiro Hiraga, Japan Nicola Huesing, Germany Hubert Huppertz, Austria Leonard V. Interrante, USA Edwin Kroke, Germany Terence G. Langdon, USA Ya-Li Li, China Yoshitake Masuda, Japan Gary Messing, USA Philippe Miele, France Kazuki Nakanishi, Japan Takeshi Okutani, Japan Clément Sanchez, France Tsugio Sato, Japan Toshimori Sekine, Japan Yoshiyuki Sugahara, Japan Yoko Suyama, Japan Naresh N. Thadhani, USA Omer Van Der Biest, Belgium Monika Willert-Porada, Germany Masahiro Yoshimura, Japan

Focused Session CB-11 - Self-propagating High-temperature Synthesis of Ceramics Chair: Alexander G. MERZHANOV, Russia Coordinator: Hamazasp E. GRIGORYAN, Russia Members Frédéric Bernard, France Inna P. Borovinskaya, Russia Giacomo Cao, Italy Elazar Gutmanas, Israel Suren Kharatyan, Armenia Evgeny Levashov, Russia Jerzy Lis, Poland Alexander Mukasyan, USA Zuhair A. Munir, USA Manshi Ohyanagi, Japan Roman Pampuch, Poland Alexander E. Sytschev, Russia Galina Xanthopoulou, Greece

Focused Session CB-12 - Layered and Functionally Graded Materials Chair: Changchun GE, P.R. China Members Serkan Dag, Turkey Moshe P. Dariel, Israel Martin Friess, Germany Michael M. Gasik, Finland Jae-Ho Jeon, Korea Akira Kawasaki, Japan Jeong-Ho Kim, USA Kiyotaka Matsuura, Japan Luis Augusto Rocha, Portugal Alexander S. Rogachev, Russia

Table of Contents Preface Committees

SECTION I – SOFT SOLUTION PROCESSING Spray Freeze Granulation of Nano Powders for Die Pressing J. Binner, K. Annapoorani and B. Vaidhyanathan New Synthesis Process of Li, Na and K Niobates from Metal Alkoxides Y. Suyama, T. Yamada, Y. Hirano, K. Takamura and K. Takahashi Novel Sol-Gel Synthesis of LiMn2O4 and LiNixCo1-xO2 Powders A. Deptula, W. Łada, T. Olczak, D. Wawszczak, M. Brykala, F. Zaza and K.C. Goretta Synthesis of Alumina and Aluminum Nitride Layers on a Graphitic Substrate via a Sol-Gel Route F. Fontaine, R. Pailler and A. Guette Synthesis of Monodispersed Plate-Like CeO2 Particles by Precipitation Process in Sodium Hydrogen Carbonate Solution S. Yin, Y. Minamidate and T. Sato Co-Doping Effect of Metal Ion on the Visible Light Responsive Photocatalytic Properties of Nitrogen Doped Titanium Dioxide P.L. Zhang, S. Yin and T. Sato Tailored Silica Based Xerogels and Aerogels for Insulation in Space Environments L. Durães, M. Ochoa, A. Portugal, N. Duarte, J.P. Dias, N. Rocha and J. Hernandez In Situ TEM Observation of Crystallization Process for LiNbO3 and NaNbO3 H. Nakano and Y. Suyama Microwave Assisted Solvothermal Synthesis and Visible Light Photocatalytic Properties of Nb and N Co-Doped SrTiO3 Nanoparticles U. Sulaeman, S. Yin and T. Sato

1 7 14 24 30 36 41 47 52

SECTION II – SPARK PLASMA AND MICROWAVE SYNTHESIS AND SINTERING Numerical Simulation of Spark Plasma Sintering C. Garcia and E. Olevsky Densification Mechanism of MgAl2O4 Spinel during Spark-Plasma-Sintering K. Morita, B.N. Kim, H. Yoshida, K. Hiraga and Y. Sakka Production and Characterization of Boron Carbide – Titanium Diboride Ceramics by Spark Plasma Sintering Method B. Uygun, G. Göller, Y. Onüralp and F.Ç. Şahin The Effects of Codoping Y2O3 on MgO Doped Spark Plasma Sintered Al2O3 B. Apak, G. Göller, Y. Onüralp and F.Ç. Şahin Spark Plasma Sintering of Boron Carbide and Effects of Various Additives on Sintering and Material Properties Y. Çelik, G. Göller, Y. Onüralp and F.Ç. Şahin Microwave Absorbency Change of Zirconia Powder and Fiber during Vacuum Heating S. Sano, S. Kawakami, Y. Takao, S. Takayama and M. Sato Microwave Assisted Reaction Sintering of ZrSiO4/α -Al2O3 Mixtures O. Ertugrul, S. Akpinar, I.M. Kusoglu and K. Onel

58 62 68 74 79 85 91

SECTION III – HYBRID MATERIALS Hybrid Foams, Colloids and Beyond: Advanced Ceramics through Integrative Chemistry N. Brun, S. Ungureanu, F. Carn, B. Julián-López and R. Backov Panoscopic Assembling of Ceramic Materials for High Performance UV-Ray Shielding Application T. Sato, X.W. Liu and S. Yin

97 107

b

12th INTERNATIONAL CERAMICS CONGRESS PART B

Lightweight Hybrid Foam with Dimensional Stability M.Y. Chen and C.G. Chen Ceramic/Polymeric Hybrids with Reduced Coefficients of Thermal Expansion C.G. Chen, K.H. Hoos and M.Y. Chen Dimension- and Direction-Controlled Gold Nanorods Deposited in Ordered Mesoporous Silica G. Kawamura, I. Hayashi, R.A. Fitrah, H. Muto, J. Hamagami and A. Matsuda

114 120 126

SECTION IV – POROUS CERAMICS Heterogeneous Sol-Gel Systems – Derived Ceramics O.A. Shilova Smart Processing for Ceramics Structure Tectonics: Fabrication of Dielectric Micro Patterns for Artificial Photosynthesis in Terahertz Wave Regions by Using Stereolithography S. Kirihara, N. Komori and N. Ohta Clay Aerogel Composite Materials D.A. Schiraldi, M.D. Gawryla and S. Alhassan Synthesis and Characterization of Mesoporous Hydroxyapatite K.S. Lew, O. Radzali and F.Y. Yeoh New Methodology in Modeling Ceramics M.A. Algatti, E. Ferreira de Lucena, É. de Campos, R.P. Mota and J.G.A. Santana Influence of Binder on Porous Ceramic Properties Prepared by Polymeric Sponge Method K. Jach, K. Pietrzak, D. Kalinski and M. Chmielewski Fabrication of Porous Silicon Nitride by Sacrificing Template Method R.M. Mesquita and A.H.A. Bressiani Aluminum Oxide Ceramics Obtained by Commercial Starch Consolidation with Gradient Porosity R.P. Mota, R. Sampaio Fernandes, É. de Campos, E. Ferreira de Lucena and M.A. Algatti Processing of Municipal Solid Waste (MSW) Fly Ash into an Environmentally Stable and Safe Material M. Isac, Z. Ghouleh and R.I.L. Guthrie

131

141 147 152 158 164 170 175 181

SECTION V – SHS CERAMICS V-1 Fundamentals and New Methods for SHS Solution Combustion as a Promising Method for the Synthesis of Nanomaterials A.S. Mukasyan Microwave Activated Combustion Synthesis and Compaction in Separate E and H Fields: Numerical Simulation and Experimental Results R. Rosa, P. Veronesi, C. Leonelli, A. Bonamartini Corradi, M. Ferraris, V. Casalegno, M. Salvo and S.H. Han Use of Electrothermal Explosion and Electro-Thermal Analyser (ETA-100) for the Study the Kinetics of Fast High-Temperature Reactions in SHS-Ceramic Systems A.S. Shteinberg and A.A. Berlin Simulation of Gasless Combustion of Mechanically Activated Solid Powder Mixtures S. Rashkovskiy Macrokinetics for Macrostructure Forming of a Product in Self-Propagating HighTemperature Synthesis V. Prokofiev and V. Smolyakov

187

197 203 213 222

V-2 SHS of Ceramic Powders Past and Current Accomplishments in Production of Ceramic Powders and Structures by Self-Propagating High-Temperature Synthesis Method J.A. Puszynski and A. Degraw Carbon Combustion Synthesis of Ceramic Oxide Nanopowders K.S. Martirosyan

228 236

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE Double SHS of W2B5 Powder from CaWO4 and B2O3 S. Yazici and B. Derin Production of Zirconium Diboride Powder by Self Propagating High Temperature Synthesis B. Akkas, M. Alkan, B. Derin and Y. Onüralp Utilization of NbC Nanoparticles Obtained by Reactive Milling in Production of Alumina Niobium Carbide Nanocomposites V. Trombini, A.H.A. Bressiani, E.M.J. Agnolon Pallone and R. Tomasi

c

246 251 257

V-3 Fabrication of SHS Products and Application Issues Composites Produced by SHS Method – Current Development and Future Trends J. Lis, D. Kata, M.M. Bućko, L. Chlubny and D. Zientara Self-Propagating High-Temperature Synthesis of Iron- and Copper-Matrix Cermets A. Chrysanthou Sintering and Hot-Pressing of Ti2AlC Obtained by SHS Process L. Chlubny, J. Lis and M.M. Bućko Catalytic Properties of the SHS Products - Review G. Xanthopoulou Porous SHS-Ceramics Y.M. Maksimov, A.I. Kirdyashkin, V.K. Baev and A.N. Gushin Mass-Forced SHS Technology of Ceramic Materials O. Odawara SHS Refractory Materials “Furnon” and their Practical Implementations in Kazakhstan and Russia Z.A. Mansurov

263 273 282 287 297 302 312

SECTION VI – LAYERED AND FUNCTIONALLY GRADED MATERIALS The Potential of Spark Plasma Sintering (SPS) Method for the Fabrication on an Industrial Scale of Functionally Graded Materials M. Tokita Comparison of Microwave and Conventional Sintering of LHA Ceramics and Functionally Graded Alumina-LHA Ceramics Z. Negahdari and M. Willert-Porada Fabrication of Functionally Graded ZTA Ceramics Using a Novel Combination of Freeze Casting and Electrophoretic Deposition A. Preiss, B. Su, S. Collins and P. Ellison Effects of Strain-Graded Plastic Deformation on Mechanical Properties of Metals K. Matsuura and M. Ohno Defect Crystal Structure of Low Temperature Modifications of Li2MO3 (M=Ti, Sn) and Related Hydroxides N.V. Tarakina, T.A. Denisova, Y.V. Baklanova, L.G. Maksimova, V.G. Zubkov and R.B. Neder Atomic and Electronic Structure of Zinc and Copper Pyrovanadates with Negative Thermal Expansion T. Krasnenko, N. Medvedeva and V. Bamburov Layered Alumina Ceramics with Porosity Steps E. Gregorová, W. Pabst and M. Chmelíčková Fabrication of Porous Intermetallic Thick Films by Metallic Powder-Liquid Reaction T. Ohmi and M. Iguchi High-Strength Reaction-Sintered Silicon Carbide for Large-Scale Mirrors - Effect of Surface Oxide Layer on Bending Strength S. Suyama and Y. Itoh Development of Functionally Graded Coating Based Plasma Facing Materials for Fusion Reactor C.C. Ge, S.Q. Guo, Y.B. Feng, Z.J. Zhou, J. Du, H.B. Zhou and C. Wang Relationship Between Microstructure and Hardness of ZrN/TiN Multi-Layers with Various Bilayer Thicknesses Y. Aoi, S. Furuhata and H. Nakano

322 332 340 348 352 358 364 370 374 383 392

d

12th INTERNATIONAL CERAMICS CONGRESS PART B

SECTION VII – OTHER PROCESSING ROUTES Mechanical Properties of Si3N4-SiC Composites Sintered by HPHT Method P. Klimczyk Effect of the Hydrothermal Heat Treatment Conditions of Titanium Substrates on the BioMimetically Grown “Bone-Like” Apatite Coatings D. Teker, C.P. Sağ, M. Dinçer, S. Alkoy and K. Öztürk Highly Porous Hydroxyapatite Ceramics for Engineering Applicatios H. Ivankovic, S. Orlic, D. Kranzelic and E. Tkalcec Lithium Disilicate Glass-Ceramic Obtained from Rice Husk-Based Silica F.A. Santos, C. dos Santos, D. Rodrigues Júnior, D.R.R. Lazar, D. Faviero de Castro, D.G. Pinatti and R.A. Conte Reactive Milling and Mechanical Alloying in Electroceramics R. Rivas-Marquez, C. Gomez-Yanez, I. Velasco-Davalos and J. Cruz-Rivera Crystal Growth of Calcite Nano-Plate by Alternate Soaking Method, Using CDS Crystals K. Hayashi, M. Tomohara, K. Fujino, G. Sakane and Y. Katayama

396 402 408 414 420 425

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.1

Spray Freeze Granulation of Nano Powders for Die Pressing Jon Binner1,a, Ketharam Annapoorani1,b and Bala Vaidhyanathan1,c 1

Department of Materials, Loughborough University, Loughborough, LE11 3TU, UK

a

[email protected], [email protected], [email protected]

Keywords: Nanopowders, spray freeze drying, die pressing.

Abstract The processing of nanocrystalline yttria doped zirconia powder via dry forming routes has been investigated via the granulation of the powder using spray freeze drying (SFD). Free-flowing and crushable powders suitable for either die or isosatic pressing have been achieved via the combination of SFD with additions of up to 2 vol% of Freon 11; the latter reducing the strength of the granules whilst not affecting the powder flowability into the die. The approach has allowed relic-free green bodies of up to 55% of theoretical density to be produced using pressures as low as 250 MPa. Introduction It has been demonstrated that a number of properties, including electrical, electronic, magnetic, optical and mechanical are all influenced by the presence of a nano grain structure in materials [1]. In addition, nanopowders can be sintered at relatively low temperatures [2], opening up possibilities of co-sintering with lower cost metals. As a result, nanostructured ceramics are likely to have a positive impact in the future on a range of industries, including aerospace, transport, communications and medical. The large scale industrial production of nanocrystalline ceramics still remains a challenge, however. It is well known that nanopowders intrinsically display poor flowability, very high compaction ratios and very low green densities when pressed due to the extremely large surface areas that they possess. They also display a very high tendency to agglomerate [3]. Die and isostatic pressing are two of industries’ favoured processing routes for the production of engineering ceramic components, however, since the technique is versatile and high production rates are achievable [4]. If the advanced ceramics industry is to be able to use dry forming routes with nanopowders then it is going to be important to granulate them to enhance the flowability. Whilst spray drying is the most widely used technique [5], it has previously been shown that it yields flowable but hard granules that don’t crush even under pressures as high as 500 MPa [6,7], approximately twice the pressure typically used by the ceramics industry, unless produced hollow [8]. However, the presence of even a single remnant can result in a severe loss of strength by acting as a Griffith flaw [9]. It is therefore important to control the granule strength so that crushable as well as flowable granules are obtained [10]. Spray freeze drying has been reported to yield soft granules from nanosized ceramics [11], eliminating the formation of segregated binder layer commonly found in spray dried granules [12]. This paper discusses the use of spray freeze drying to achieve granules that have very similar flowabilities to existing, commercial submicron powders whilst crushing completely at pressures as low as 250 MPa; values that will enable industry to use existing equipment on its factory floor. The bulk of the work has been undertaken using yttria stabilised zirconia (YSZ) nanopowders with a primary particle size of ∼16 nm, however the principles have also been demonstrated to work with a coarser, commercial nanoalumina powder.

2

12th INTERNATIONAL CERAMICS CONGRESS PART B

Experimental Aqueous nanosuspensions containing 3 mol% yttria stabilized zirconia (3YSZ) with an average particle size of ~16 nm and a solid content of 5.5 vol% (26 wt%) with no other additives and an acidic pH of ~2.4 were obtained from MEL Chemicals Ltd, UK. They were concentrated to ~20 vol% (60 wt%) using a route described in detail elsewhere [13] but which involved changing the suspension pH to ~10 using tetra-methyl ammonium hydroxide, TMAH, (Aldrich Chemicals Ltd., Dorset, UK), then adding ~3 wt% tri ammonium citrate, TAC, (Fisher Scientific UK Ltd, Loughborough, UK) as a dispersant before heating gently at 60oC using a water bath under constant stirring whilst exposing the suspension to ultrasound (Soniprep 150 Ultrasonicator, MSE Scientific Instruments, Manchester, UK) at regular intervals to assist in breaking up any agglomerates present. Further batches were produced that also contained 1 and 2 vol% of flurotrichloromethane (Freon 11, Fisher Scientific UK Ltd, Loughborough, UK) introduced directly after the addition of the TMAH and TAC but before thermal concentration. The viscosity of all the final suspensions had to be 1000 cSt Gelation and drying 100ºC-11 days, 150ºC-24h, 170ºC-4 days dried gel, shard Grinding gel powders Preliminary heating (1oC/min) until selfignition Final temperature treatment and grinding

Li2MnO4 or LiNixCo1-xO2 Fig.1. Flow chart for preparation of Li2MnO4 or LiNixCo1-xO2 by routine CSGP (unspecified metallic species are denoted Me and acetates as Ac). Ni

Ni

O O

Ni ASC Ni

O

C

O

O

Fig. 2. Strongly bonded carbon formed during thermal treatment of ASC-containing species.

15

16

12th INTERNATIONAL CERAMICS CONGRESS PART B

Other negative features of CSGP processing include foaming during thermal treatment, long gelation times, and formation of highly voluminous macro-porous gels. In the case of LiMn2O4, addition of HNO3 before gelation of the Me acetate precursors decreased foaming [6]. Recent combustion synthesis papers for LiNixCo1-xO2 described use of Me nitrates as oxidizers and glycine [13, 14], glycol [15], or starch [16] as (effectively) the fuel. Only with use of starch were foaming and ignition observed [16]. In our previous work, foaming and ignition have always occurred, which indicates that the decomposition products of ASC act as an efficient fuel. However, in our CSGP process, ASC cannot be decreased without precipitation of Me hydroxides in the alkalization step. Consequently, the goal of this work was to examine the possibility of decreasing the second component, the fuel which consists of decomposition products of ASC. We hypothesized that this result could be achieved by radically decreasing the times of the drying and gelation steps. Experimental Details Starting sols were prepared from aqueous solutions of 1.5 M Mn2+ or 1 M of xNi2+ and (1-x)Co2+ acetates that contained ASC by alkalizing them with LiOH and NH4OH [6]. The resulting sols were gelled within a Buchi RE 121 Rotavapor unit (Flawil, Switzerland) by drying at 70ºC under a pressure of ~15 mm Hg for 3 h and then at 100ºC for 1 h. They were subsequently heated at ≥ 300ºC/h to 750ºC (Fig. 3). Starting complex sols Gelation and drying at 70oC, 3 h under reduced pressure (~15 mm Hg), drying 3 h, then 1 h at 100oC Gel Heating rate (~300oC/min) with self-ignition until 750oC

Grinding

Li2MnO4 or LiNixCo1-xO2 Fig. 3. Preparation of Li2MnO4 or LiNixCo1-xO2 by modified CSGP. All thermal treatments were observed visually and were filmed. Self-ignition temperatures were measured with a Type EMT-300 pyrometer (Czaki, Poland). All products were analyzed by X-ray diffraction (XRD) with a Rigaku Miniflex diffractometer (Tokyo, Japan); Cu-Kα radiation was used, the tube output voltage was 30 kV and the tube output current was 15 mA, 2θ = 3-90°, with steps of 0.02° and a scanning rate of 2°/min. Raw data were smoothed by the Savitzky method, background was eliminated by the Sonnevelt method, and Kα2 was eliminated. Infrared measurements were conducted with a Perkin Elmer Model 983 Spectrometer (Waltham, MA); the potassium bromide pellet technique was adopted. Carbonate concentrations were determined by

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

17

internal standardization [17], with use of sodium nitride and 2 carbonate bands at 875 cm-1. This method, to our knowledge not previously used by others in examination of similar materials, is quite specific and highly accurate for carbonate determination. Results and Discussion The gels obtained by the modified CSGP could be heated at rates exceeding 300ºC/h (HHR) without occurrence of foaming. Self-ignition was observed always. Soft aggregates were produced, which could be easily ground to powders. Infrared spectra were obtained for the various layered oxides (Fig. 4). The ν2 carbonate band at 875 cm-1 was very weak, except for a specimen that contained LiCoO2 plus Na2CO3 at a concentration of 5% carbonate, which was examined as a standard. Even for pure LiNiO2, only traces of carbonates (290nm, Kenko L41 Pro (W) for λ>400nm, and Fuji triacetyl cellulose filter for λ>510nm. It has been reported that during the photocatalytic destruction, about 20% of NO is directly reduced to N2, and the other 80% is oxidized to NO3- species [22]. For comparison, the photocatalytic reaction was also carried out using a standard commercial undoped titania (Degussa P25). Results and Discussion Table 1. Fe contents and specific surface areas (S.S.A.) of the samples Sample

TiO2 (P25)

TiO2-xNy

TiO2-xNy+1 mol % Fe

TiO2-xNy+5 mol % Fe

Fe content [mol %]

0

0

1.1

4.2

48

166

163

162

2

S.S.A. [m /g]

To confirm the existence of Fe in the obtained samples, EDX analysis was carried out. It was observed that the detected Fe contents were very close to that added in the solution as shown in Table 1. Both N doped TiO2 and Fe and N co-doped TiO2 showed the similar specific surface areas

38

12th INTERNATIONAL CERAMICS CONGRESS PART B

of above 160 m2/g, which were more than three times higher than that of the commercial TiO2 (P25). Figure 1 shows the XRD patterns of the obtained samples. For the N doped titanium dioxide, strong diffraction peak around 25o was observed, which could be corresponding to the main peak of anatase phase TiO2. At the meanwhile, diffraction peaks around 27o and 36o, assigned to rutile TiO2, were also observed. By the co-doping with Fe, almost no changes in XRD patterns were observed, even the co-doping amount of Fe increased from 1.0 to 5.0 mol.%.

(a) Intensity / a.u.

Intensity / a.u.

(c) (b) (a) Rutile 89-4920 Anatase 78-2486 10

20

30 40 50 60 o 2θ (CuKα) / Fig. 1 XRD patterns of (a) N doped TiO2, and Fe and N co-doped TiO2 with (b) 1.0 mol.%, and (c) 5.0 mol.% of Fe

A A

R A

A: Anatase R: Rutile (b) A

800 700 600 500 400 300 200 100 -1 Raman shift / cm Fig. 2 Raman spectra of (a) N doped TiO2, and (b) Fe and N co-doped TiO2 with 5.0 mol.% of Fe

The crystalline phases of the samples were further confirmed by Raman spectra. As shown in Fig. 2, strong Raman shifts corresponding to anatase TiO2 could be obviously observed in N doped TiO2, and a typical shift of rutle TiO2 around 450 cm-1 was also observed. For the Fe and N co-doped sample, shifts corresponding to anatase and rutile TiO2 could still be observed, but the intensity of Raman shifts became weak, probably due to the partial disordering in the lattices caused by the co-doping of Fe. The photo-absorption property of the sample was characterized by the diffuse reflectance spectrum. The commercial P25 TiO2 showed no absorption ability under visible light region, due to the large band gap energy of ca. 3 eV. The N doped TiO2 showed a two step absorption; the first absorption around 390 nm should be related to the band gap of TiO2, and the second absorption edge around 500 nm could be attributed to the newly formed N 1s orbit by the nitrogen doping in the molecular structure [19]. By further co-doping Fe into the samples, absorption under visible light could be further enhanced. At meanwhile, the color of the prepared samples also became dark with an increase in Fe content. Figure 4 shows the photocatalytic activities of the samples for the oxidative destruction of NOx gas under irradiation of a high pressure Hg lamp with various wavelengths. The commercial undoped TiO2 (P25) showed low activity under the irradiation of visible light (λ>510nm), due to its large band gap energy. The activity under visible light irradiation could be significantly increased by doping nitrogen due to the narrowing of band gap. The N doped TiO2 showed exceeded activity

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE 100

(a)

(b) 60

(d)

40

40

(a) (b) (c) (d)

30 20 10

20

200

DeNOx ability / %

(c)

80 Reflectance / %

50

39

300

400 500 600 700 Wavelength / nm Fig. 3 Diffuse reflectance spectra of (a) commercial P25 TiO2, (b) N doped TiO2, and Fe and N co-doped TiO2 with (c) 1.0 mol.% and (d) 5.0 mol.% of Fe

0

>510

>400 >290 Wavelength / nm Fig. 4 Photocatalytic deNOx activities of (a) commercial P25 TiO2, (b) N doped TiO2, and Fe and N co-doped TiO2 with (c) 1.0 mol.% and (d) 5.0 mol.% of Fe

DeNOx ability / %

under both visible light and UV light irradiation to the commercial P25 TiO2, due to its mixed crystalline phases, and larger specific surface area [20]. Co-doping of Fe and N was proved to be negative for the photocatalytic activity. As shown in Fig. 4, 1 mol.% of Fe showed almost no influence on the activity, but by increasing the content of Fe to 5 mol.%, the deNOx ability decreased drastically, although Fe co-doping actually showed no influence on either the crystalline phase or the specific surface area of the sample. The sample contained 5 mol.% co-doped Fe even showed the highest visible light absorption ability The decrease in the photocatalytic activity of Fe and N co-doped TiO2 should be attributed to the increase of anion vacancy in the sample. In the N doped TiO2, the charge balance was destroyed because of the substitution of N3- into O2- sites. To maintain the electro neutrality, anion vacancies would be generated in the lattices as TiO2-2-1.5yN3-y□0.5y. It was believed that the anion vacancy would act as the recombination centres of photo-induced electrons and holes to depress the photocatalytic activity. By further co-doping with 35 (a) Fe3+, the anion vacancy would increase as (b) 30 Ti4+1-xFe3+xO2-2-0.5x-1.5yN3-y□0.5X+0.5y. This may be (c) (d) 25 the reason why Fe co-doping resulted to decrease the photocatalytic activity. 20 In order to overcome this problem, a higher 15 valence metal ion, Nb5+, was co-doped into the N 10 doped TiO2 instead of Fe3+ to reduce the anion 5 vacancy as Ti4+1-xNb5+xO2-2-1.5y+0.5xN3-y □ 0.5y-0.5x. The preparation method was similar to that of Fe 0 627(R) 530(G) 445(B) 390(UV) co-doping. Nb co-doping also showed almost no Wavelength / nm noticeable influence on the crystalline phase and Fig. 5 Photocatalytic activities in deNOx by (a) commercial P25 TiO2, (b) N doped TiO2, and specific surface area of the samples. The Nb and N co-doped TiO2 with (c) 2.0 mol.% photocatalytic deNOx activities under LED lamps and (d) 6.0 mol.% of Nb

40

12th INTERNATIONAL CERAMICS CONGRESS PART B

with various wavelengths are shown in Fig. 5. It is obvious that co-doping with Nb could effectively promote the photocatalytic activity under photoirradiations of all wavelengths, probably due to the effective reduction of the vacancy in the lattice by doping with higher valence metal ion. Conclusions Different valences of metal ions such as Fe3+ and Nb5+ were co-doped with nitrogen ion into titanium dioxide by hydrothermal method. The co-doping of low-content metal ions caused no noticeable influence on the crystalline phase and specific surface area of the samples, however, coping with Nb ion could increase the deNOx ability, but co-doping with Fe ion depressed it. Co-doping with higher valence metal ion was considered to be effective in reducing the vacancy in the lattice which acts as the recombination centers of photo-induced electrons and holes, and achieving higher photocatalytic activity. Reference [1] A. Mills, and S. L. Hunte, J. Photochem. Photobiol. A: Chem., 108 (1997) 1 [2] J. Wang, S. Yin, M. Komatsu, Q. Zhang, F. Saito, and T. Sato, Appl. Catal. B.Environ., 52 (2004) 11 [3] J. Wang, S. Yin, Q. Zhang, F. Saito, and T. Sato, J.Mater.Sci., 39 (2004) 715 [4] P. Zhang, S. Yin, R. Li, and T. Sato, J. Ceram. Soc. Jpn, 115 (2007) 898 [5] K. Hashimoto, H. Irie and A. Fujishima, Jpn. J. Appl. Phys. 44 (2005) 8269 [6] R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 293 (2001), 269 [7] T. Morikawa, R. Asahi, T. Ohwaki, K. Aoki, and Y. Taga, Jpn. J. Apply. Phys., 40 (2001), 561 [8] S. U. M. Khan, M. Al-Shahry and W. B. Ingler Jr, Science, 297(2002), 2243 [9] T. Umebayashi, T. Yamaki, S. Tanaka, and K. Asai, Chem. Lett., 32(2003),330 [10]Q. Zhang, J. Wang, S. Yin, T. Sato, and F. Saito, J.Am.Ceram.Soc., 87 (2004) 1161 [11] D. Huang, S. Liao, J. Liu, Z. Dang, J. Photochem. Photobiol. A: Chem., 184 (2006) 282 [12] J. Xu, J. Lia, W. Dai, Y. Cao, H. Li, K. Fan, Appl. Catal. B: Environ., 79 (2008) 72 [13] J. He, J. Zhao, T. Shen, H. Hidaka, N. Serpone, J. Phys. Chem. B, 101 (1997) 9027. [14] J. Yu, J. Xiong, B. Cheng, Y. Yu, J. Wang, J. Solid State Chem., 178 (2005) 1968. [15] X.H. Tang, D.Y. Li, J. Phys. Chem. C., 112 (2008) 5405. [16] J.C. Yu, J.G. Yu, W.K. Ho, Z.T. Jiang, L.Z. Zhang, Chem. Mater., 14 (2002) 3808. [17] W. Choi, A. Termin, M.R. Hoffmann, J. Phys. Chem., 98 (1994) 13669. [18] Y. Aita, M. Komatsu S. Yin, and T. Sato, J. Solid State Chem., 177 (2004) 3235 [19] S. Yin, Y. Aita, M. Komatsu, J. Wang, Q. Tang, and T. Sato, J. Mater. Chem., 15 (2005) 674 [20] P. Zhang, B. Liu, S. Yin, Y. Wang, V. Petrykin, M. Kakihana, and T. Sato, Mater. Chem. Phys., 116 (2009) 269 [21] S. Yin, H. Hasegawa, D. Maeda, M. Ishitsuka, T. Sato, J. Photochem. Photobiol. A: Chem., 163 (2004) 1 [22] M. Anpo, Recent Development on Visible Light Response Type Photocatalyst, NTS, Tokyo, 2002, p.9, ISBN 4-86043-009-03

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.41

Tailored Silica Based Xerogels and Aerogels for Insulation in Space Environments Luisa Durães1, a, Marta Ochoa1,b, António Portugal1,c, Nelson Duarte2,d, João Paulo Dias2,e, Nuno Rocha3,f and Jessica Hernandez3,g 1

Department of Chemical Engineering, Univ. Coimbra, 3030-790 Coimbra, Portugal 2

Led&Mat, IPN-Instituto Pedro Nunes, 3030-199 Coimbra, Portugal

3

AST-Active Space Technologies, IPN, 3030-199 Coimbra, Portugal

a

[email protected], [email protected], [email protected], [email protected], [email protected], f [email protected], [email protected]

Keywords: Sol-gel, nanostructures, silica based xerogels/aerogels, insulation, Space applications.

Abstract. In this work, the sol-gel technology is used to produce silica based xerogels and aerogels suitable for insulation applications in Space. The properties of the obtained materials are tailored varying the precursor – Methyltrimethoxysilane (MTMS) or Methyltriethoxysilane (MTES), and the solvent – methanol or ethanol. A two-step acid-base catalyzed synthesis is used, being the obtained gels dried at atmospheric pressure, in the case of xerogels, and in supercritical conditions, for aerogels. Density and thermal conductivity must be made as low as possible for the sought application and only highly porous materials can fulfill this requirement. The obtained xerogels and aerogels, either with MTMS or MTES, show very promising properties for thermal insulation in Space, when methanol is used as solvent. The more suitable materials are obtained with MTMS and exhibit very low density (80-100 kg/m3), very high surface area (~ 400 m2/g) and small pore size (~ 30-40 Å). They also show moderate flexibility and a remarkable hydrophobic character (~ 150º). Introduction The development of new and more powerful techniques for the analysis of materials at molecular scale has enhanced the study of chemical systems at this level. Following this trend, the materials synthesis technologies based on bottom-up approaches, namely the sol-gel technology, has experienced a significant dissemination. This technology offers many advantages over traditional methods: high homogeneity and purity of the product, low temperature required, control of the product microstructure in the synthesis and drying stages, preparation of advanced materials with tailored properties. It is a wet technology, which starts from a homogeneous solution that, through hydrolysis and condensation reactions of the precursor, gives rise to a colloidal solution – sol. By polycondensation reactions, the sol evolves to an integrated solid network with the liquid phase in the pores – gel. A xerogel or an aerogel is obtained, if the gel is dried at ambient pressure or in supercritical conditions, respectively. Silica aerogels synthesized with tetra-alkoxysilane precursors (Si(OR)4) exhibit a nanostructured three-dimensional solid network with low density (~ 100-200 Kg/m3), high porosity (> 90 %) and surface area, low thermal conductivity (~ 0.01-0.02 W/mK) and high transparency (~ 90 %). These unique properties allow their use as thermal or acoustic insulators, dielectric or optical materials, filters and catalysts [1,2]. However, they are brittle, absorb moisture and deteriorate with time. Less dense, flexible and hydrophobic xerogels/aerogels would improve the performance of these materials, for example, in Space applications. The key property to tailorable xerogels/aerogels is the porosity – a high porous structure results in very low thermal conductivity and density. To make these materials competitive with current materials for insulation in Space, the following requirements must be met: working pressure – 10-103 Pa, working temperature – -150-500 ºC, density – 10-100 kg/m3, thermal conductivity – 3-16 mW/(m K) and moderate flexibility. If the

42

12th INTERNATIONAL CERAMICS CONGRESS PART B

application involves insulation of electrical/electronic devices, it is also required a material with high hydrophobicity. Aerogels may have several applications in Space, such as: thermal and structural insulators for re-entry and Mars vehicles, spacecraft devices and cryogenic tanks; coatings for solar panels; semi-flexible coatings for cables, Printed Circuit Boards and Multi Layer Insulation Blankets in spacesuits and spacecrafts; collection of Space debris; acoustic insulators for spacecrafts and the International Space Station; windows and optical instruments [3]. Using MTMS or MTES precursors in a two-step acid-base catalyzed sol-gel chemistry, as proposed by Rao, Bhagat, Nadargi and co-workers [2,4-8], the obtained silica based xerogels and aerogels can fulfill the targets referred above. In MTMS or MTES, one alkoxy group of Si(OR)4 is replaced by the group CH3. This group does not suffer hydrolysis and remains in the gel structure after the condensation reactions (Fig. 1), providing hydrophobicity and flexibility to the final materials. The presence of methyl groups within the structure increases its degree of disorder, leading to an increase in porosity and hence a reduction in density. The porosity is also controlled by the drying technique, being the xerogels less porous and usually denser than aerogels. However, both have very low density, very high surface area and porosity, and consequently very low thermal conductivity. Hydrolysis reaction: OR H3 C

OR + 3 H2O

Si OR

Acid catalyst Solvent

OH H3 C

Si

OH

+ 3 ROH

OH

Condensation reaction: OH 2 H3 C

Si OH

OH OH

OH

+

H3 C

Si OH

OH

Basic catalyst

H3 C

Si

OH O

O H3 C

Si OH

Si

CH3

+

O O

Si

4 H2 O

CH3

OH

Figure 1. Hydrolysis and condensation reactions of trialkoxysilanes: R=CH3 for MTMS and R=CH2CH3 for MTES.

In this work, silica based xerogels and aerogels were obtained via sol-gel technology with MTMS and MTES precursors, following the experimental procedures described by Rao, Bhagat et al. [2,4] and used in an earlier work [9]. An optimal solvent/MTMS molar ratio found in the literature [2,46,9] for improved properties of monolithicity, density, porosity and flexibility, i.e. a value of 35, was applied for both precursors. In the case of MTES, obtaining gels with this dilution degree is an innovation, since the maximum solvent/MTES molar ratio indicated in the literature for gel formation is 20 [7,8]. In addition, the total replacement of the synthesis solvent from methanol to ethanol was investigated, as methanol is harmful to operators and environment. In literature, for the used precursors, gels were only obtained with a partial substitution of the solvent [10]. Gels were successfully obtained in all tested conditions, although the final products did not always exhibited the desired properties for Space applications. Materials and Methods The applied sol-gel technology to synthesize the silica based xerogels and aerogels includes a twostep acid-base catalyzed synthesis, pursued by ageing and drying stages. The experimental procedure follows the description of Rao, Baghat et al. [2,4] and uses the instrumentation already presented in an earlier work [9]. MTMS (98 %, Aldrich) and MTES (≥ 98 %, Fluka) precursors and methanol (99.8 %, Riedel-de Haёn) and ethanol (99.5 %, Panreac) solvents were used as received. Oxalic acid (99 %, Fluka) and ammonium hydroxide (25 % in water, Fluka) catalysts were used in the form of 0.001M and 10 M aqueous solutions, respectively. In a typical synthesis, the precursor

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

43

was diluted in methanol or ethanol, and water was added, by joining the oxalic acid solution, to promote the hydrolysis of the precursor in acidic conditions – acid step; after 24 h, the ammonium hydroxide solution was added, drop by drop (each 20 s), and the condensation reactions began, forming a sol – basic step. The acid and basic steps were carried out at 25 ºC and the pH evolution was registered in the basic step. The used molar ratios of precursor:solvent:acidic water:basic water were 1:35:4:4. The obtained sol was placed in an oven at 27 ºC and left for gelation. During gelation, the alkaline medium favors the polycondensation reactions, resulting in the building of a tri-dimensional solid network of Si-O-Si with the liquid solvents and catalysts inside – gel. The resulting gel was aged for two days at 27 ºC to promote the cohesion of the solid network. In this period, condensation reactions continues as long as silanols are close enough to react, and syneresis and coarsening processes also occur [11,12]. The concentrations of the acid and basic catalysts solutions, the amounts of added acidic and basic water and the ageing period were defined according to the optimal values found in the literature for improved density and flexibility [2,4-9], and considering a minimum amount of acid to avoid the damage of the equipments. The obtained gels were dried by Ambient Pressure Drying (APD), to produce xerogels, and by Supercritical Fluid Drying (SFD), to produce aerogels. In the first case, the gel was put into a ventilated oven, at atmospheric pressure, and submitted to a temperature cycle – 24 h at 60 ºC, followed by three stages, at 100 ºC, 150 ºC and 200 ºC, of one hour each. This evaporative method (Fig. 2) leads to the shrinkage of the pores due to surface tension effects. In the second case, the gel was placed in an autoclave, which was then filled up with the solvent. The temperature of the autoclave was increased, causing a simultaneous increase in the pressure inside it. When the critical point of the solvent (Tc(CH3OH) = 239.4 ºC and Pc(CH3OH) = 8.09 MPa; Tc(CH3CH2OH) = 243.0 ºC and Pc(CH3CH2OH) = 6.3 MPa [1]) was exceeded, the liquid became a supercritical fluid and was released, at constant temperature, leading to the pressure decrease. Near atmospheric pressure, the autoclave was flushed with N2 to remove the remaining gaseous solvent. In this process, the shrinkage is reduced because the liquid-gas equilibrium line is not crossed (Fig. 2). P Liquid Solid

Supercritical region c

t - triple point c - critical point APD path SFD path

0.1 MPa t

Gas 240 ºC

T

Figure 2. Schematic representation of APD and SFD processes in a P-T equilibrium diagram.

The elemental composition (C,H,N,O), chemical structure, microstructure, surface area, density and contact angle of the xerogels and aerogels were accessed by elemental analysis (EA 1108 CHNS-O, Fisons Instruments), FTIR (Magna-IRTM System 750, Nicolet), SEM (JMS–5310, JOEL), nitrogen gas adsorption (ASAP 2000, Micromeritics), weight-volume measurements and a contact angle technique (Contact Angle System OCA 20, Dataphysics), respectively. Results and Discussion During the basic step of the synthesis, a sharp pH increase was observed with the addition of the first two drops of the basic catalyst solution (~ 0.1 cm3); then, the pH increased slowly towards its final value (pH ≈ 11) (Fig. 3). The pH evolution curves do not depend significantly of the used solvent. Contrarily, the type of precursor has a remarkable effect in the region of sharp increase of pH: for MTMS, the pH starts at ~ 3.5 and increases up to ~ 9.5 and, with MTES, the pH starts at ~ 9.5 and increases to ~ 10.5. These pH differences must be more deeply studied in future works. They can be caused by chemical mechanism differences and/or by a retardation of the hydrolysis in

44

12th INTERNATIONAL CERAMICS CONGRESS PART B

systems with MTES when compared to MTMS. This retardation is due to the steric effect caused by a larger alkoxy group [11-13] and decreases the number of silanols in the system. It is known that the acidity of the silanol groups increases with increasing extent of hydrolysis [12]. The gel times were 5 and 6 hours for the systems MTMS/methanol and MTES/methanol, respectively, and ~ 5 days for the systems with ethanol. The difference registered between MTMS and MTES in methanol was already explained in the last paragraph. In this case, as the condensation pursues in a basic medium and there are not significant retardation effects caused by methanol, it is expected that condensation reactions occur faster than the hydrolysis reactions, producing more ramified polymers [11,12]. The large differences in the gel times obtained when the solvent is exchanged can be explained by lower overall reactions rate in ethanol, which is due to: lower charge activity of the catalysts, caused by the smaller dielectric constant of ethanol; higher steric effects and more complex hydrogen bonding structures with water and silicic acid, what decrease the mobility of the species in the medium and, thus, the probability that they have to react [10,13,14]. As the condensation reactions are significantly retarded in these conditions, it is expected that they will be slower than the hydrolysis reactions, being the formed polymers more linear and dense [11,12]. Monolithic aerogels and xerogels were obtained for the system MTMS/methanol (Fig. 4 - top). For the case of MTES/methanol, the final materials were fissured monoliths with powdery surfaces, exhibiting the xerogels a weaker structure. All the materials synthesized using ethanol as solvent were very fragmented and rigid, being the aerogels more opaque (Fig. 4 – bottom). 12 11 10 9 8 7 6 5 4 3

pH

MTMS_MeOH MTES_MeOH MTMS_EtOH MTES_EtOH

-3

0

3

6

9 12 15 18 21 24 27 30 Time (min) Figure 3. pH evolution during the basic step of the syntheses.

Figure 4. Typical appearance of monolithic and non monolithic final products.

The FTIR spectra obtained for all the synthesized xerogels and aerogels were almost superimposed. A typical FTIR result for these materials is presented in Fig. 5. The assignment of the peaks was based on data for similar systems [15]. The found vibration bands confirm the expected chemical structure for these materials (Fig. 1): a predominant inorganic network based on Si-O-Si bonds (silica), with a methyl group per Si (hybrid materials) and –OH terminal groups at the network ends. This structure appears to be independent of the used solvent and precursor, for the studied cases. The elemental analysis relevant results and the contact angle values are summarized in Table 1. The theoretical elemental ratios calculated for complete condensation, neglecting the OH groups at the structure ends, are 1Si:1C:3H:1.5O and 41.8Si:17.9C:4.5H:35.8O in molar and %wt base, respectively. For the case of non condensation of one OH group of the monomer, these ratios are 1Si:1C:4H:2O and 36.9Si:15.8C:5.3H:42.0O. A comparison between the theoretical and experimental %wt C will be used to evaluate the extent of condensation reactions, since the values for H and O are affected by other factors, namely: i) residues in the samples; ii) the OH terminal groups of the network were not considered in the theoretical estimates; iii) the used temperature in the elemental analysis furnace (1060 ºC) is not enough to break the Si-O bonds. The first item is confirmed by the non null experimental values found for the %wt N (~ 0.3), which are due to basic catalyst residues. The last item justifies the very low %wt O (1.5-4.3) found in the samples. The experimental values of %wt C are in better agreement with the theoretical hypothesis of complete condensation, except in the case of xerogels obtained with ethanol (Table 1). These xerogels were

0 3600

3200

2800 2400 2000 1600 -1 Wavenumber (cm )

1200

45

vs Si-O-Si v Si-O d O-Si-O

v Si-C

vs Si-O

20 4000

das C-H ds C-H

40

d O-H

60

vs C-H vas C-H

80 O-H and Si-OH

Transmittance (%)

100

vas Si-O-Si both

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

800

Figure 5. Typical FTIR spectrum for the obtained xerogels/aerogels (v - stretching 400 vibration; d - bending vibration; s – symmetric; as – asymmetric).

Table 1. Elemental analysis results for C and H %wt and contact angle for the synthesized materials. Aerogels: Xerogels: System %wt C %wt H Contact angle [º] %wt C %wt H Contact angle [º] MTMS_MeOH 19.841 5.502 148.4 ± 1.8 19.721 5.439 150.9± 4.3 MTES_MeOH 18.590 5.498 154.9 ± 4.7 16.721 5.490 144.6 ± 2.3 MTMS_EtOH 15.287 5.315 131.9± 9.9 17.691 5.355 140.2 ± 5.7 MTES_EtOH 12.635 4.661 137.7 ± 2.9 18.227 5.590 141.0 ± 7.1

very dense and presented a toasted color in the surfaces in contact with the glass tubes. Probably, the CH3 groups were oxidized in these regions, lowering the C content. The %wt H is affected by many factors. It is underestimated in the calculus and its experimental values do not show a specific trend. The obtained contact angles (Table 1) are typical of highly hydrophobic materials and do not exhibit a notorious dependence on the C content, maybe because the later doesn’t vary much. The systems with ethanol present slightly smaller values, but these were obtained with compressed powders. Table 2 summarizes the density and porosity related values for the synthesized materials. Considering that the required materials must have low density and high porosity, it can be concluded that methanol is the more adequate solvent. With this solvent, materials have been synthesized with a bulk density as low as ~ 90 kg/m3 and with porosities higher than 90 %. Accordingly, the surface area of these less dense materials is very high (~ 400 m2/g). In general, the xerogels are denser materials, as expected. The synthesized materials have a mean pore size of 27-44 Å, which is near the boundary between micropores and mesopores (20 Å). SEM images for the studied materials are presented in Figure 6. The xerogels/aerogels obtained using ethanol have nearly the same structure for both precursors. The structures shown in Figure 6 confirm the discussion made up to this point, since denser materials show more closed and less ramified patterns. The system of MTMS in methanol gives rise to cloudy and diffuse structures, with very small interlinked units, although this is not very clear in the images. The materials obtained with MTES in methanol show larger interlinked units, which may be due to the different pH conditions and to the steric effect of the alkyl group, that may enhance the growing of units and its regularity. Also, the evaporative drying appears to contribute to the growing of the units. Table 2. Density, surface area, porosity and contact angle of the obtained materials. System Bulk density Skeleton density Porosity BET surf. area [kg/m3] [kg/m3] [%] [m2/g] MTMS_MeOH - xerogel 88.4 ± 4.5 1183.1 ± 49.3 93 408.8 ± 10.2 MTMS_MeOH - aerogel 102.8 ± 14.1 1220.7 ± 38.5 92 411.8 ± 7.3 MTES_MeOH - xerogel 151.1 ± 12.8 1190.9 ± 28.3 87 40.9 ± 0.5(b) MTES_MeOH - aerogel 100.6 ± 2.7 1184.4 ± 47.7 92 361.8 ± 20.3 MTMS_EtOH - xerogel 394(a) 1313.3 ± 21.5 70 112.6 ± 1.1 MTMS_EtOH - aerogel 325(a) 1268.8 ± 31.1 74 190.6 ± 2.8 MTES_EtOH - xerogel 429(a) 1287.4 ± 11.7 67 191.1 ± 3.5 MTES_EtOH - aerogel 304(a) 1278.4 ± 8.3 76 119.5 ± 2.5 (a) Volume measured by immersion in ethanol (low accuracy); (b) Part of this sample became a analysis.

BJH desorption pore size [Å] 38 34 33 27 27 44 31 37 powder during the

46

12th INTERNATIONAL CERAMICS CONGRESS PART B

MTMS_MeOH - xerogel

MTES_MeOH - xerogel

MTES_EtOH - xerogel

MTMS_MeOH - aerogel MTES_MeOH - aerogel Figure 6. SEM images of the synthesized materials.

MTMS_EtOH - aerogel

Conclusions The sol-gel technology was applied to produce silica based xerogels and aerogels, using MTMS or MTES as precursors and methanol or ethanol as solvents. Considering the material properties required for insulation protection in Space, the most promising systems were those where methanol was used. Moreover, with this solvent, the MTMS gave rise to materials with the lower density, the higher surface area and porosity, and also with moderate flexibility and high hydrophobicity. Aknowledgements This work was developed under the project "AerTPS - Aerogel Thermal Protection Systems" by the Consortium "AST/FCTUC/IPN", funded by QREN, under the Operational Programme for Competitiveness Factors granted by the European Regional Development Fund. References [1] A.C. Pierre and G.M. Pajonk: Chem. Rev. Vol.102 (2002), p.4243. [2] A.V. Rao, S.D. Bhagat, H. Hirashim and G.M. Pajonk: J. Colloid and Interface Sci. Vol.300 (2006), p.279. [3] J. Hernández and R. Patricio, Active Space Technologies, ESA Type A contract no. 19528/NL/06/CO (2008). [4] S.D. Bhagat, C.-S. Oh, Y.-H Kim, Y.-S Ahn and J.-G. Yeo: Microporous Mesoporous Mater. Vol.100 (2007), p.350. [5] N.D. Hedge and A.V. Rao: J. Mater. Sci. Vol.42 (2007), p.6965. [6] D. Nadargi, S. Latthe and A. Rao: J. Sol-Gel Sci. Technol. Vol.49 (2009), p.53. [7] D. Nadargi and A. Rao: J. Alloys Compd. Vol.467 (2009), p.397. [8] D. Nadargi, S. Latthe, H. Hirashima and A. Rao: Microporous Mesoporous Mater. Vol.117 (2009), p.617. [9] L. Durães, S. Nogueira, A. Santos, C. Preciso, J. Hernandez and A. Portugal, in: Proc. 10th Int. Chem. Biol. Eng. Conf. – CHEMPOR 2008, edited by E. Ferreira and M. Mota, Department of Biological Engineering of University of Minho, Braga, (2008), p.563 and CD-ROM. [10] S. Laschober, M. Sulyok and E. Rosenberg: J. Chromatogr. A Vol.1144 (2007), p.55. [11] L.L. Hench and R. Orefice, in: Kirk-Othmer – Encyclopedia of Chemical Technology, 4thEd., Vol.22, edited by J.I. Kroschwitz and M. Howe-Grant, John Wiley & Sons, NY (1997), p.497. [12] C.J. Brinker and G.W. Scherer: Sol-Gel Science, Academic Press, Boston (1990). [13] A.V. Rao, G.M. Pajonk and N.N. Parvathy: J. Mater. Sci. Vol.29 (1994) p.1807. [14] H. Jiang, Z. Zheng and X. Wang: Vib. Spectrosc. Vol.46 (2008), p.1. [15] R. Al.Oweini and H. El-Rassy: J. Mol. Struct. Vol.919 (2009) p.140.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.47

1

In-situ TEM Observation of Crystallization Process for LiNbO3 and NaNbO3 Hiromi Nakano1a, Yoko Suyama2b

Toyohashi University of Technolog, Tempaku, Toyohashi, 441-8580, Japan 2 Shimane University, Nishikawatsu-cho, Matsue, 690-8504, Japan a

[email protected], [email protected]

Keywords: High-temperature TEM, nano-grain, microstructure, in-situ observation

Abstract. Fabrication of advanced electronic components requires high-quality powders. In this work, nano-powders of Li or Na niobates are synthesized from (Li or Na)-Nb ethoxide by a sol-crystal method. A single crystal of (Li or Na)-Nb ethoxide is decomposed to an amorphous matrix below 473 K. Next, small crystals are grown by heating at the appropriate temperature for each specimen. The sol-crystal method provides homogeneous quality and fine grains by heating at lower temperature. Structural analysis of the powders is performed by a transmission electron microscope (TEM) and X-ray diffraction. As a result, LiNbO3 turns to dense-powders, but NaNbO3 forms nano-porous powders. In order to understand this difference, we try to observe in-situ the crystallization and grain growth processes by high-temperature TEM. We successfully observe in-situ this processing and discuss the structural change and formation mechanism of LiNbO3, comparing these features with those of NaNbO3. Introduction Lithium niobate (LiNbO3) is known as an important ferroelectric material because of its piezoelectrical, electro-optical, and photo-refractive properties [1]. LiNbO3-NaNbO3 and NaNbO3-based piezoelectric ceramics have been investigated as Pb-free materials [2–4]. Fabrication of advanced electrical components requires high-quality powders, which must provide homogeneous quality, fine grains, and uniform grain size. In the case of LiNbO3, the conventional process requires high sintering temperature and causes Li2O loss. Suyama et al. reported a new preparation route for the compounds BaTiO3 and Ba5Nb4O15, in which those compounds could be successfully synthesized from double-metal alkoxides [5, 6]. This method could achieve synthesis at low temperature and obtain fine particles. In our previous study, NaNbO3 nano-porous grains were synthesized by the sol-crystal method [7]. Optical and electrical properties of ceramics are strongly influenced by the structural defects and morphology [8]. For improving the morphology and microstructure, the crystallization process should be controlled under the experimental condition. However, the crystallization behavior is not well understood. In this study, the crystallization behavior and microstructure were investigated by TEM and high-temperature TEM for LiNbO3 nano-dense grains prepared by the sol-crystal method. Consequently, the obtained features were compared with those of NaNbO3. Experimental Procedure Nb(OEt)5 and (Li or Na)(OEt) were synthesized in ethanol from Nb2O5 and Li or Na metal, respectively. A crystal of (Li or Na)-Nb ethoxide was obtained from Nb(OEt)5 and (Li or Na)(OEt). This procedure has been reported in detail elsewhere [7]. The crystal was annealed at 373–1173 K in air for 1.5 h. Those specimens were structurally analyzed by powder XRD (Rigaku RINT-2100SL) with CuK
radiation and a high-resolution TEM (JEM-2100F, JEOL) equipped with energy-dispersive spectroscopy (EDS). The decomposition behaviors of Li-Nb and Li-Na-Nb ethoxides were measured by TG-DTA (Rigaku TG8120).

48

12th INTERNATIONAL CERAMICS CONGRESS PART B

The crystallization process was observed in-situ by TEM (JEM-2000EX) using a thermal stage (EM-SHH4). The temperature on the stage was controlled manually from room temperature to 1038 K. TEM images were recorded in real time with a video camera. Results and discussion 

Intensity (a.u.)

LiNbO3

   





  

  

973K 873K 773K 673K

20

30

40 50 60 70 diffraction angle, 2/
(CuK)

80

Fig. 1 XRD patterns of Li-Nb ethoxide annealed at various temperatures for 1.5 h. Rectangular-shaped crystals of Li-Nb ethoxide with sizes of 1–5 mm were heated for 1.5 h. Figure 1 shows powder XRD patterns of specimens annealed at various temperatures. These crystals were decomposed to an amorphous state by heating. At >773 K, small peaks appeared, which indicated the existence of LiNbO3 crystals with a rhombohedral phase (a = 0.5148 nm, c = 1.3863 nm as a hexagonal cell)[Inorganic crystal structure database ICSD#28294]. An exothermic peak was detected at 758 K by TG-DTA measurement, possibly indicating the crystallization temperature of LiNbO3.

(a)

(c)

(b)

(d)

Fig. 2 HRTEM images of Li-Nb ethoxide annealed at 553 K in (a) and (b), and at 873 K in (c) and (d).

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

49

To determine the exact temperature at which the crystals begin to form in the amorphous matrix, microstructures were observed by TEM. Figure 2 shows TEM images of Li-Nb ethoxide annealed at 553 K and 873 K. At 99% density was achieved at 1760 o C, under 40 MPa pressure for 5 min. in vacuum atmosphere. It was seen that TiB2 addition helps densifying B4 C at lower sintering temperatures. Highest hardness value was measured as 34,5 GPa for 5% TiB2 containing sample and the highest fracture toughness was 6,9 MPa.m1/2 for 10% TiB2 containing sample, well higher than that of pure B4 C. References [1] Çınar Şahin F., Yeşilçubuk A., “B4 C-ZrB2 Kompozitlerinin Reaktif Sıcak Presleme İle B4 CTiB2 Kompozitlerinin Reaktif Sıcak Presleme ve Sıcak Presleme İle Eldesi, (2007) [2] Thevenot F., 1990. “Boron Carbide – A Comprehensive Review”, Journal of the European Ceramic Society, 6, 205-225. [3] Dudina D.V., Hulbert D.M., Jiang D, Unuvar C, Cytron S.J., Mukherjee A.K., 2008. “In Situ Boron Carbide - Titanium Diboride Composites Prepared by Mechanical Milling and Subsequent Spark Plasma Sintering”, J. Mater. Sci,. 43:3569-3576. [4] Cai K.F., Nan C.W., Schmuecker M., Mueller E., 2003. “Microstructure of Hot-pressed B4 C TiB2 Thermoelectric Composites”, J. All. and Comp., 350, 313-318. [5] Roy T.K., Subramanian C., Suri A.K., 2006. “Pressureless Sintering of Boron Carbide”, Ceramics International, 32, 227-233.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

73

[6] Tokita, M., “Mechanism of Spark Plasma Sintering”, Sumitomo Coal Mining Company Ltd., Kanagawa 213, Japan. [7] Skorokhod, V., Krstic, V.D., 2000. “High Strength-High Toughness B4 C-TiB2 Composites”, Journal of Materials Science Letters, 19, 237-239.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.74

The effects of codoping Y2O3 on MgO doped Spark Plasma Sintered Al2O3 Burcu Apak1,a, Gultekin Goller1,b, Onuralp Yucel1,2,c and Filiz Cinar Sahin1,2,d 1

2

Istanbul Technical University, Metallurgy and Materials Engineering Department 34469 Maslak, Istanbul, Turkey

Istanbul Technical University Prof. Adnan Tekin Applied Research Center of Materials Science & Production Technologies 34469 Maslak, Istanbul, Turkey. a

[email protected], [email protected], [email protected], [email protected]

Keywords: Spark Plasma Sintering, Alumina ceramics, Doping

Abstract. Nanocrystalline alumina (Al2O3) powders were sintered by Spark Plasma Sintering (SPS) method in a vacuum atmosphere to obtain highly dense and fine grained final ceramic products. In the first section of experiments, 0.4 % wt MgO doped and 0.4 wt % Y2O3 doped Al2O3 were sintered at high temperatures and under high pressure with a SPS system. Later sintering procedures were carried out with codoping Y2O3 with the cathodic ratio of 0.4 wt % in order to investigate dopant effects on spark plasma sintered alumina. The microstructures of all samples were observed using scanning electron microscope and the properties such as density, hardness and fracture toughness were examined. Introduction Recently, densification of ceramics with the grain size on the nanometer scale has been attracting considerable attention because finer grain size structures bring the improved mechanical, thermal, and optical properties with. Nowadays, a new manufacturing process which aimed to obtain fully dense ceramics while retaining small grain size, called Spark Plasma Sintering (SPS) is drawing attention. Spark plasma sintering is a consolidation technique that can combine high heating and cooling rates with an uniaxial applied pressure resulting in short processing time. Since SPS has the advantage of a rapid heating rate, it has widely been a candidate to sinter dense and fine-grained alumina at low temperatures within a short time [1,2] To modify the microstructure and mechanical properties of ceramics, an aliovalent dopant MgO and an isovalent dopant Y2O3 are added to alumina. During sintering of alumina, MgO doping enhances the densification rate [3,4]. It is known that beside the suppression of grain growth which can be defined as the primary role of magnesia, it is also beneficial for sintering with the influence on the surface diffusion [3]. Because of the ability of avoiding abnormal or anisotropic grain growth MgO is a leading dopant for alumina. Yttria is an important dopant for increasing creep resistance of alumina and hindering grain growth during sintering [6,9]. However, it has been reported that yttria doping of more than 2 mass % retards densification whereas that of 50 to 2000 ppm mass enhances it [5,6]. Also, yttria doping to alumina causes two different types of microstructures: for finer grain sizes, only grain boundary segregation of yttrium is observed whereas for coarser grain sizes, the grain boundaries are saturated with yttrium, resulting in the intergranular precipitation of a yttrium rich second phase [6,7] The aim of this study is to find out the effect of yttria doping, magnesia doping and yttriamagnesia codoping on the densification, microstructure, hardness and fracture toughness of spark plasma sintered alumina ceramics with keeping the cathodic ratio of dopants in 0.4 mass %.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

75

Experimental Procedure In this study, α-alumina powder (Inframat Advanced Materials, Farmington, CT) with an average particle size of 100 nm and a purity of 99.99%, was used. Batches were prepared by mixing Merck quality magnesium nitrate [Mg(NO3)2·6H2O] and/or Merck quaility yttrium nitrate [Y(NO3)3·6H2O] and alumina in ethanol medium by ball milling for 24 hours with alumina balls. The slurry was then dried and heated at 750 °C for 1 h. During heating, the magnesium nitrate transforms to MgO and yttrium nitrate to Y2O3. After screening, the dry powder was loaded in a graphite die for SPS. A graphite die with 50 mm in inner diameter was filled with the doped alumina. The sintering processes are carried out by using the SPS apparatus (SPS-7.40MK-VII, SPS Syntex Inc.). After applying a pressure of 80 MPa, specimens are heated with heating rate of 100 °C/min to 1300 and 1340 °C. The temperature of the die was measured by an optical pyrometer. All of the samples are sintered with soaking time of 5 minutes in a vacuum atmosphere. Furthermore, shrinkage, displacement, heating current, and voltage were recorded during the whole process for every 5 seconds. After sintering process, Archimedes method was used to determine the final densities of the compacts. Specimens, polished with a diamond paste having particle size of 1 µm, were thermally etched at 1100 °C for 1 hour. The micrographs of all sample surfaces were observed by scanning electron microscopy (SEM; Model JSM 7000F, JEOL, Tokyo, Japan). The hardness and fracture toughness of the samples at room temperature were evaluated by the Vickers indentation technique at a load of 98 N (Struers, Duramin A300). Results and Discussion Density values of spark plasma sintered ceramics at 1300 °C and 1340 °C applying 80 MPa pressure with 5 minutes soaking time under vacuum atmosphere are given in the Fig. 1. The densities are altered between 98.1 % and 99.3 % values. With increasing MgO wt % content, the density tended to increase. However in both sintering temperatures, codoping caused less densified products when compared with single doped alumina samples in case of keeping the cathodic ratio in 0.4 wt %. The density results do not show any remarkable change, especially in the specimens sintered at 1340 °C, till the MgO content reaches 0.4 wt %. Because the highest densities were obtained in the magnesia doped samples, it can be implied that yttria addition had a negative effect on densification of alumina in mentioned ratios.

Fig. 1. Relative density versus MgO % content of the samples sintered in 1300 °C and 1340 °C under 80 MPa pressure for 5 minutes in vacuum.

76

12th INTERNATIONAL CERAMICS CONGRESS PART B

Fig. 2. Microstructures of Al2O3 ceramics spark plasma sintered at different temperatures: (a) 0.4 % Y2O3 (b) 0.1 % MgO 0.3 % Y2O3 (c) 0.2 MgO % 0.2 Y2O3 (d) 0.4 % MgO at 1300 °C (e) 0.4 % Y2O3 (f) 0.1 % MgO 0.3 % Y2O3 (g) 0.2 MgO % 0.2 Y2O3 (h) 0.4 % MgO at 1340 °C

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

77

As it is shown in Fig. 2., the increasing sintering temperature led to larger grains. The finest grains are seemed in the samples that contain 0.1 % MgO and 0.3 % Y2O3 in both sintering temperatures. Also, more homogenous microstructures, as regards grain size, were obtained in the codoped specimens. In the study of Stuer et al. [8], Y2O3 doped alumina indicated finer grain size values for single doping strategy at 1350 °C whereas results were just the opposite at 1400 °C. The attained results of this study at the 1300 and 1340 °C are parallel to that 1350 °C of the mentioned study [8]. Furthermore, because the starting powder and final products had fine particles/grains, there is not any yttrium rich phase in the grain boundaries.

Fig. 3. The variation of hardness as a function of MgO content in ceramics SPSed at 1300 °C and 1340 °C applying 80 MPa pressure under vacuum with 5 minutes soaking time

Fig. 4. The variation of fracture toughness as a function of MgO content in ceramics SPSed at 1300 °C and 1340 °C applying 80 MPa pressure under vacuum with 5 minutes soaking time Fig. 3. illustrates the variation of hardness versus the amount of MgO in the ceramics for 1300 °C and 1340 °C sintering temperatures. The hardness values were close to each other ranging between 19.5 GPa and 20.5 GPa. Moreover, increasing sintering temperature resulted in higher hardness results which were because of higher densification rates in higher sintering temperatures. As can be seen, yttria addition to MgO doped alumina enhanced the hardness values in both sintering temperatures, codoping alumina with 0.2 wt % yttria and 0.2 wt % magnesia served the purpose in terms of hardness.

78

12th INTERNATIONAL CERAMICS CONGRESS PART B

Fracture toughness values versus MgO % content are given in Fig. 4. The fracture toughnesses are compared well with the hardness values. Fracture toughness results decrease with increasing hardness according to Fig 3. and Fig. 4. The highest fracture toughness is attained in the lowest densified sample contained 0.1 % MgO and 0.3 % Y2O3 sintered at 1300 °C. It is most probably due to the role of nanopores in hindering crack propagation by reducing the stress intensity at the crack tip when it interacts with an intergranular crack and thus increase the toughness [10]. Also, homogenous microstructure of same sample which can be clearly seen in Fig. 2. b., caused higher fracture toughness. Conclusion With the cathodic ratio 0.4 %, it is possible to produce MgO and/or Y2O3 doped alumina ceramics higher than 98.1 % densities by using spark plasma sintering at 1300 °C and 1340 °C under 80 MPa pressure in vacuum atmosphere. While codoping affected densification negatively, hardness and fracture toughness of the ceramics are improved. Acknowledgements This work has been supported by The Scientific and The Technological Research Council of Turkey-TUBITAK under Projection 109M584. References [1] Zhijian Shen, Mats Johnsson, Zhe Zhao, and Mats Nygren, “Spark Plasma Sintering of Alumina”, J. Am. Ceram. Soc., 85 [8] 1921–27 (2002). [2] Byung-Nam Kim, Keijiro Hiraga, Koji Morita and Hidehiro Yoshida, “Spark plasma sintering of transparent alumina”, Scripta Materialia 57, 607–610 (2007). [3] Stephen J. Bennison and Martin P. Harmer, “Effect of Magnesia Solute on Surface Diffusion in Sapphire and the Role of Magnesia in the Sintering of Alumina”, J. Am. Ceram. Soc., 73 [4] 833-37 (1990). [4] A. H. Heuer, “The Role of MgO in the Sintering of Alumina,” J. Am. Ceram. Soc., 62 [S-61 317-18 (1979). [5] P. Nanni, C.T.H. Stoddart, and E. D. Hondros, “Grain Boundary Segregation and Sintering in Alumina”, Mater. Chem., 1, 297-320 (1976) [6] E. Sato and C. Carry, Yttria Doping and Sintering of Submicrometer- Grained α-Alumina”, J. Am. Ceram. Soc., 79 [8] 2156-60 (1996). [7] P.Gruffel and C. Carry, “Effect of Grain Size on Yttrium Grain Boundary Segregation in FineGrained Alumina”, J. Eur. Ceram. Soc., 11, 189-99 (1993). [8] Michael Stuer, Zhe Zhaob, Ulrich Aschauer and Paul Bowena, “Transparent polycrystalline alumina using spark plasma sintering: Effect of Mg, Y and La doping”, J. Eur. Ceram. Soc. 30, 1335–1343 (2010). [9] Voytovych R., MacLaren I., Gulgun M.A., Cannon R.M., Ruhle M. “The effect of yttrium on densification and grain growth in alpha-alumina”, Acta Mater 2002;50:3453–63 [10] Dibyendu Chakravarty, Sandip Bysakh, Kuttanellore Muraleedharan, Tata Narasinga Rao, and Ranganathan Sundaresan, “Spark Plasma Sintering of Magnesia-Doped Alumina with High Hardness and Fracture Toughness”, J. Am. Ceram. Soc., 91 [1] 203–208 (2008)

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.79

Spark Plasma Sintering of Boron Carbide and Effects of Various Additives on Sintering and Material Properties Yusuf Çelik1,a, Gültekin Göller1,b, Onuralp Yücel1,c, Filiz Şahin1,d 1

Istanbul Technical Universty, Faculty of Chemical and Metallurgical Engineering, Department of Metallurgical and Materials Engineering; Maslak, Istanbul, 34469, Turkey a

[email protected], [email protected], [email protected], [email protected]

Keywords: Spark Plasma Sintering, SPS, B4C, Al2O3,Y2O3,SiO2

Abstract. The aim of this study was to produce dense and mechanically strong boron carbide ceramics with the help of different oxide additives. Physical properties of two different grades of pure boron carbide powders were analysed and sintered by Spark Plasma Sintering method. Starting powders were prepared by ball milling with the addition of 5 wt. % Y2O3, Al2O3, SiO2 and Y2O3 + Al2O3. In the sintering step, powder mixtures were sintered by SPS method in round-shaped graphite dies under 50 MPa for 5 minutes in the range of 1700-1800oC. In the characterization step, sintered sample compositions and microstructures were characterized by XRD and SEM analysis respectively. The hardness values were measured under 1000 g load and the density values were measured with Archimedes' principle. The fracture toughness analysis were also carried out. Introduction Boron carbide is a well known non-metallic hard material, such as alumina, silicon carbide, and diamond [1]. The main properties of boron carbide is high melting point, high mechanical strength combined with low density, high neutron absorbation capability and semiconducting characteristics, which could be beneficial for ballistic applications [2]. These potentials are prevented by the high sintering temperature of boron carbide due to the covalently-bonded structure [3]. After hot press and hot isostatic press, spark plasma sintering is a new method for sintering technical ceramics, especially covalently-bonded ones. Spark Plasma Sintering and related techniques are popular thanks to its fast, energy-efficient and highly effective sintering capability for producing fine-grained, fully dense materials. Moreover the combination of these properties combined with high pressure and controlled atmosphere, gives an ability to develop new materials with optimized microstructures [ 4]. Sintering of boron carbide can be easier by using additives, especially oxide compounds, which also supply better mechanical properties. In order to increase the density, carbon is the unique additive for presureless sintering [5]. Besides, elemental additives such as Mg, Al, B, Fe, Co, Ni and Cu provide lower sintering temperatures between 1750–1900oC as well as grain growth [6]. Addition of TiO2, which forms some TiB2 in the structure at 1500oC due to its reaction with B4C, ensure a two phase composite and increase the fracture toughness [7]. Alumina as an additive, increases the relative density through material transport and product yield due to the reaction with B4C. Al2O3 precipitates on grain boundries and increases the mechanical properties such as hardness, elastic modulus, flexural strength, and fracture toughness [8]. The effects of ZrO2, V2O5, Cr2O3, Y2O3 and La2O3, which can form borides at high melting points, are investigated as well to prevent mass loss by vaporization during sintering. However it is comprehended that lantanide oxides (Y2O3 and La2O3) are more influential than transition metal

80

12th INTERNATIONAL CERAMICS CONGRESS PART B

oxides on sintering [9]. The addition of ZrO2 on the other hand, when pressureless sintering boron carbide, is effective on sintering boron carbide with increased consolidation and mechanical properties [10, 11]. Although there aren’t many studies on spark plasma sintered boron carbide, addition of Fe and microstructural characterization of it has been investigated as well in recent years [12, 13]. In this study, the effects of some additives such as Y2O3 , Al2O3, SiO2 and Y2O3 + Al2O3 on sintering properties of spark plasma sintered B4C were invastigated. Experimental Procedure B4C, Y2O3 and Al2O3 powders used in this study were supplied by H.C. Starck, Germany and SiO2 by Alfa Aesar. The B4C powder and the additives were mixed by ball milling for 24 hours. The powder mixture was then dried in an oven for 24 hours at 150oC. 25 grams of HP and HS grade B4C powders and B4C powders with additives from each mixture were put into a graphite mold with an initial diameter of 50 mm and inserted into the SPS device. Sintering was carried out under 50 MPa pressure for 5 minutes under vacuum at elevated temperatures depending on the shrinkage in the sample. The sintering temperatures of each sample are given in Table 1. Table 1: Sintering temperatures of the powder mixtures Powder Mixture B4C (HP) B4C (HS) B4C+%5 Y2O3 B4C+%5 SiO2 B4C+%5 Al2O3 B4C+%5 Al2O3+%5 Y2O3

Sintering Temperature [oC] 1800 1770 1740 1700 1750 1700

The densities of the samples were measured according to the Archimedes’ principle. Sintered samples with 5 mm thickness were cut into mechanical testing pieces with a diamond disc and polished with diamond paste. Hardness and fracture toughness measurements were carried out by Streurs Duramin A300 Vickers micro-hardness tester under a load of 1000 g. Fracture toughness values were calculated by Antsis equation as seen below. KIc = 0,016 (E/H)1/2 x (P/C3/2).

(1)

Where KIc is the fracture toughness, E-elastic modulus, H-hardness, P-load and 2C-full crack length produced by Vickers Hv10 indentation. Microstructural inspections were carried out by JEOL FEG-7000F scanning electron microscope and the X-ray diffraction patterns were obtained from the polished surfaces of the samples by Panalytical X’Pert Pro diffractometer. Result and Discussion Table 2 shows the theoretical and measured densities of HP grade and HS grade, spark plasma sintered boron carbide samples. Density value of HS grade boron carbide is higher than that of HP

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

81

grade since its 1,77 µm particle size is smaller than 2,95 µm of HP grade. It is estimated that the smaller particle size results in higher densities and the SPS mechanism helps hindering the grain growth. Since HS grade powder yielded better density results, experiments with the additives was carried out using the HS grade B4C powders. Table 2: Density comparison of HP and HS grade B4C powders. Sample No 1 2

Sample

Temperature [oC]

B4C (HP) B4C (HS)

1800 1770

Density [g/cm3] 2,47 2,52

Theorical Density [%] 97,90 99,86

The effects of additives on densities can be seen in the Table 3. Theoretical densities of the samples are lower than that of additive-free (HS grade) samples. This can be explained with the formation of new phases during the sintering process as can be seen in Figure 1. While Al2O3 and Y2O3 forms borides and carbides, SiO2 and Y2O3 + Al2O3 additive yields amorphous glassy phases. Table 3: Density comparision of the samples sintered with additives Sample No 3 4 5 6

Sample

Temperature [oC]

B4C+%5 Y2O3 B4C+%5 SiO2 B4C+%5 Al2O3 B4C+%5 Al2O3+%5 Y2O3

1740 1700 1750

Density [g/cm3] 2,57 2,47 2,53

Theorical Density [%] 99,57 97,71 98,38

1700

2,58

97,90

Figure 1: XRD patterns of the samples

82

12th INTERNATIONAL CERAMICS CONGRESS PART B

Figure 2 compares the hardness and fracture toughness of the samples. The hardness and fracture toughness values of HS grade powder is higher than those of HP grade due to the smaller grain size. All of the additives increased the hardness values of the sintered samples. The reason of this increase is the new phases that formed and precipitated on grain boundries during the sintering such as carbides, borides and glassy phases, as seen in Fig. 3-a. Highest hardness values obtained at sample 3 which include Y2O3 as an additive, cause the samples have the lowest porosity and new phases like Y4C7 and YB12 on grain boundries. However, it was also seen that the additives decreased the fracture toughness. The only sample with a higher fracture toughness is sample 4, which included 5 wt. % SiO2 and its value was measured almost equal to that of pure B4C.

Figure 2: Hardness and fracture toughness graphs of samples Figure 3 shows the SEM pictures of the samples, taken from the fractured surfaces.

a.

b.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

c.

83

d.

e. Figure 3: SEM images of pure HS grade and additive-including B4C samples: a) HS grade B4C b) B4C + Y2O3, c) B4C + SiO2, d) B4C + Al2O3, e) B4C + Y2O3 + Al2O3 As seen in Figure 3-a, sintered HS grade B4C samples have 1-2 µm particle size. In Figure 3-b, the Y2O3 including sample can be seen and according to the XRD and EDS analysis, the white phases are Y4C7 and YB12. In Figure 3-c, it can be seen that SiO2 and B4C reacted and formed SiC. XRD analysis of the samples shows only SiC, but according to the EDS analysis, there is also a glassy phase including B, Si and O. Figure 3-d shows the sintered body of B4C+Al2O3 mixture, in which aluminum borates were formed during sintering. They are seen as white phases on SEM images. Finally, Figure 3-e shows the B4C sample which was sintered with Y2O3 + Al2O3 additives. According to the XRD and EDS analysis white phases in that image represent YAlO3. Conclusion In this study, the density, hardness and fracture toughness of pure and additive-including boron carbide samples, which were sintered by SPS at the highest possible sintering temperatures, were presented. The sinterability and the hardness properties of B4C increased with small particle precursors and sintering additives such as Y2O3, Al2O3, SiO2 and Y2O3+Al2O3. Small amount of additives decreased the sintering temperature of boron carbide. Although there were expectations of an increase in fracture toughness, the samples yielded lower values. It is thought to be due to the formation of brittle phases on grain boundries. Addition of Y2O3 can uniformly distribute the pores in the structure and yield higher hardness values than that of pure boron carbide. Further investigations will be carried out by TEM.

84

12th INTERNATIONAL CERAMICS CONGRESS PART B

References [1] G. Ferrari, G., The `Hows' and `Whys' of Armour Penetration," MILTECH, 81-96 (1988). [2] F. Thevenot, Boron carbide-a comprehensive review, Journal of the European Ceramic Society 6 (1990) 205–225. [3] H. Lee, R.F. Speyer, Pressureless sintering of boron carbide, Journal of the American Ceramic Society 86 (2003) 1468–1473. [4] R.S.Dobedoe, G.D.West,M.H. Lewis, Spark Plasma Sintering of Ceramics, (2003). [5] J.E. Zorzia, C.A. Perottoni, J.A.H. da Jornada, Hardness and wear resistance of B4C ceramics prepared with several additives, Materials Letters 59 (2005) 2932 – 2935. [6] M.V. Swain, Structure and properties of ceramics, in: R.W. Cahn, P. Haasen, E.J. Kramer (Eds.), Materials Science and Technology, vol. 11, VCH Publishers Inc., NY, USA, 1993, pp. 175–258. [7] L. Levin, N. Frage, M.P. Dariel, A novel approach for the preparation of B4C-based cermets , International Journal of Refractory Metals & Hard Materials 18 (2000) 131-135. [8] Hae-Won Kim, Young-Hag Koh, Hyoun-Ee Kim, Densification and Mechanical Properties of B4C with Al2O3 as a Sintering Aid, J. Am. Ceram. Soc., 83 [11] (2000) 2863–65

[9] A. Goldstein, Y. Yeshurun, A. Goldenberg, B4C/metal boride composites derived from B4C/metal oxide mixtures, Journal of the European Ceramic Society 27 (2007) 695–700 [10] C. Subramanian, T.K. Roy, T.S.R.Ch. Murthy, P. Sengupta, G.B. Kale, M.V. Krishnaiah, A.K. Suri, Effect of Zirconia Addition on Pressureless Sintering of Boron Carbide,; Ceramics International 34 (2008) 1543–1549 [11] Adrian Goldstein,* Ygal Geffen, and Ayala Goldenberg, Boron Carbide–Zirconium Boride In Situ Composites by the Reactive Pressureless Sintering of Boron Carbide– Zirconia Mixtures J. Am. Ceram. Soc., 84 [3] (2001) 642–44 [12] S. Hayun, S. Kalabukhov, V. Ezersky, M.P. Dariel, N. Frage, Microstructural Characterization of Spark Plasma Sintered Boron Carbide Ceramics, Ceramic International, 36 (2010) 451-457 [13] N. Frage, S. Hayun, S. Kalabukhov, and M. P. Dariel, The Effect of Fe Addition on the Densification of B4C Powder by Spark Plasma Sintering, Powder Metallurgy and Metal Ceramics, 46 (2007) 711-12

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.85

Microwave Absorbency Change of Zirconia Powder and Fiber during Vacuum Heating Saburo SANO1,a, Shoji KAWAKAMI1,b, Yasumasa TAKAO1,c, Sadatsugu TAKAYAMA2,d and Motoyasu SATO2,e 1

National Institute of Advanced Industrial Science and Technology (AIST), 2266-98 Anagahora, Shimoshidami, Moriyama-ku, Nagoya-city, 463-8560 JAPAN 2

National Institute for Fusion Science (NIFS), 322-6 Oroshi, Toki-city, 509-5292 JAPAN a

[email protected], b [email protected], c [email protected], d [email protected], e [email protected]

Keywords: microwave, absorbency, measurement, circular wave-guide fixture, vector network analyzer, high-temperature, zirconia, stabilized zirconia, powder, fiber

Abstract Stabilized zirconia shows rather high m icrowave absorbency at room tem perature, and the absorbency become higher with increasing tem perature. In this study, stabilized zirconia powder, partially stabilized zirconia powder and zirconia fiber were subjected for m icrowave absorption measurements at elevated temperature. Microwave absorption measurements were done by using a system consists of a microwave vector network analyzer, a circular wave-guide fixture and a vacuum furnace. Microwave absorbency was evaluated by thereflection power change from the sample in the circular wave-guide fixture under vacuum heati ng. Microwave absorbency of stabilized zirconia powder, partially stabilized zirconia powder and zirconia fiber gradually increased with the increase of tem perature. W e supposed that the increase of m icrowave absorbency is related to the ionic (oxygen) conduction behavior of stabilized zi rconia. Stoichiometric composition ZrO2 powder was also subjected for a m easurement to consider th e relation between m icrowave absorbency and ion conduction of zirconia. As the resu lt, stoichiometric composition ZrO 2 powder was not absorbed microwave power even when th e powder was heated up to 900 oC because it isn’t an oxygen ion conductor. Introduction In studies on ceramic sintering by microwave heating, many kinds of ceramics have been handled to investigate. One of the interesting ceramics on microwave heating researches is zirconia, because it leads to the runaway phenom ena, which cause s “onion”-type cracking during sintering, under microwave heating. The phenom ena was reported at the very first stage of research on m icrowave sintering of ceramics by Jammey et al. [1] They explained that “runaway” heating conditions were routinely encountered during which the therm ocouple temperature would ri se rapidly, even though the microwave power into the furnace m ight actually falling when zirconia was m icrowave heated. Quit a num ber of researchers thought the reason is an abrupt change of m icrowave absorbency of zirconia at a tem perature. One of objectives in th is study is to clarify this supposition. To prevent runaway phenom ena, “picket fence” arrangem ent of SiC rods [1] or use of higher frequency millimeter-wave [1,2] were successfully applied. As a basis of developing the microwave sintering technology, it is important to know the absorbency of m icrowave power of ceram ics, especially at high tem perature preferably up to sintering temperature. Som e attempts to m easure microwave absorption behaviors at high tem perature has been performed. In these attem pts resonant cavity, coaxial fixture, wave-guide fixture [3] and free space fixture [4] were used. Cross and Dim itakis reported a theory for high tem perature dielectric measurement by cavity perturbation [5].

86

12th INTERNATIONAL CERAMICS CONGRESS PART B

Our m icrowave absorption m easurement system us ed in this study is based on the wave-guide method [6]. Sam ple powder filled at bottom end in a circular wave-guide is heated in a vacuum furnace and m icrowave reflection spectrum around 13.5 GHz is m easured at the other side of wave-guide. In this paper, results of m icrowave absorption measurement for zirconia powders and fiber at elevated temperature up to 900oC will be shown. Experimental As m easurement objectives stabilized zirconia powder (Tosoh corporation: TZ-8Y), partially stabilized zirconia powder (Tosoh corporation: TZ- 3Y) and zirconia fiber (Shinagawa Refractories Co., Ltd.: Y7Z) were used. Stoichiom etric com position ZrO 2 powder (W ako Pure Chem ical Industries, Ltd.: chem ical reagent grade) was used as reference to consider the relation between microwave absorbency and oxygen ion conduction of zirconia. The microwave absorption m easurement system is schematically shown in figure 1. Sizes of the circular wave-guide fixture are 660m m long a nd 16m m internal diam eter, and the m atching frequency is adjusted around 13.0GHz. A microwave vector network analyzer (Wiltron: 37269A) is used to m easure the reflection spectrum from the ci rcular wave-guide fixture. Sam ple powder is packed in the bottom end in the wave-guide fixture and reflection signal is measured at another side with the microwave network analyzer. One end of the fixture, in which sample powder is packed, is heated in a vacuum furnace for preventing oxidation of wave-guide metal. Change of signal power from the sam ple during heating is m easured using tim e dom ain m ode of the m icrowave network analyzer at a frequency range from 13.0 to 13.8GHz (or from 13.0 to 14.2GHz). The m icrowave network analyzer was calibrated at the end of coaxial cable prior to the measurement. Figure 2 shows a typical reflection spectrum of frequency dom ain m ode m easured with the microwave network analyzer. In this case, the wave-guide fixture was em pty. Periodic resonant peaks are observed in Fig. 2. These resonant fr equencies are related to standing waves in the wave-guide. Figure 3 is obtained by a conversion from frequency domain data to time domain. Tow large broad observed in Fig. 2. These resonant fr equencies are related to standing waves in the wave-guide. Figure 3 is obtained by a conversion from frequency domain data to time domain. Tow large broad peaks are seen in Fig. 3. The peak around 0ns is reflection signal from upper end in the

Fig.1 Schematic illustration of microwave absorption measurement system.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

87

Reflection (S22) / dB

0

-5

-10

-15

-20 13

13.2

13.4

13.6

13.8

14

14.2

Frequency / GHz

Fig.2 Microwave reflection spectrum from empty circular wave-guide fixture. Frequency domain mode. 10

form upper end

Reflection (S22) / dB

0

from bottom

-10 -20 -30 -40 -50 -60 -70 -80 -1

0

1

2

3

4

5

6

7

Time / ns

Fig.3 Microwave reflection spectrum from em pty circular wave-guide fixture. Time domain mode.

circler wave-guide and the peak around 4.7ns is refl ection signal from the bottom. The value of the peak around 4.7ns is approximately 13dB. When the bottom of the wave-guide is filled with high loss material for microwave the peak value from the bottom will be reduced because the material adsorbs microwave energy. Therefore, we can evaluate th e microwave absorbency of a sam ple material by the change of this peak power change. Results and Discussion Figure 4 shows a m icrowave absorption m easurement result f or partially stabilized zirconia at elevated temperature up to 900 oC. A measurement result for empty wave-guide fixture (indexed as “blank”) is also plotted as a com parison. Abso rption of m icrowave power by partially stabilized zirconia powder is very small at low temperature; reflection power is almost overlapped with the data

12th INTERNATIONAL CERAMICS CONGRESS PART B

Reflection form sample / dB

88

-11

measured at 13.0-13.8GHz in vacuum blank

-12 -13

TZ-3Y

-14

1005mg, t=2.86mm ρ=29.1%T.D.

-15 -16 -17 0

200

400

600

800

1000

Temperature / oC

Reflection form sample / dB

Fig.4 Microwave absorption mesurment result for partiall stabilized zirconia (TZ-3Y) at elevated temperature.

-11

measured at 13.0-13.8GHz in vacuum blank

-12 -13

TZ-8Y

-14

1005mg, t=2.73mm ρ=30.5%T.D.

-15 -16 -17 0

200

400

600

800

1000

o

Temperature / C Fig.5 Microwave absorption mesurment result for stabilized zirconia (TZ-8Y) at elevated temperature. from an empty fixture. However, reflection power is decreased, i.e. microwave absorbency increases, with the increase of tem perature. Th e value of decrease reaches 3dB at 900 oC. This change is reversible for temperature, microwave reflection power from partially stabilized zirconia decreased with the decrease of temperature. Figure 5 shows a m icrowave absorption m easurement result f or stabilized zirconia at elevated temperature up to 900oC. Stabilized zirconia shows almost same microwave absorption change with partially stabilized zirconia. Reflection power is decreased with the increase of temperature, and this change is reversible for temperature change. The value of decrease reaches 4dB at 900oC. As seen in Figs. 4 and 5, temperature change of microwave absorbency is gradual, not abruptly occurs at a particular tem perature as m entioned in in troduction part. The occurring process of runaway

Reflection form sample / dB

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

-11

measured at 13.0-13.8GHz in vacuum

-12

89

blank

-13

Y7Z

-14

428mg, t=6mm ρ=5.9%T.D.

-15 -16 -17 0

200

400

600

800

1000

o

Temperature / C Fig.6 Microwave absorption mesurment result for zirconia fiber (Y7Z) at elevated temperature. during m icrowave sintering of zirconia ceram ics (p artially stabilized and stabilized zirconia) is considered as follows. At m iddle tem perature (around 500 oC), absorbency of stabilized zirconia becoming larger and the value is depend on tem perature. If a tem perature distribution exists in the heated body, high temperature part absorbs more microwave power. The excess heat will be diffused rapidly if the therm al conductiv ity of the heated body is high enough. However, the therm al conductivity of zirconia is low. Therefore, the sintering body could not diffuse som e excess heat when the microwave absorbency of zirconia become high enough and runaway occurs. Namely, it is thought that runaway occurs when the balance of heat generation by microwave absorption and heat diffusion from heated part is broken at a temperature. Figure 6 shows m icrowave absorption measurement result for zirconia fiber, which is stabilized by adding yttria, at elevated tem perature up to 900 oC. Zirconia fiber shows alm ost same microwave absorption change with partially stabilized zirconi a and stabilized zirconia. Reflection power is decreased with the increase of temperature, and this change is reversible for temperature change. The value of decrease reaches 3.5dB at 900oC. The value is almost same with though the sample weigh is almost half of partially stabilized and stabilized zirconia powders. Figure 7 shows a m icrowave absorption m easurement result f or stoichiom eric com position ZrO 2 powder for considering the relation between m icrowave absorbency and oxygen ion conduction of zirconia. As shown in Fig.7, stoichiometric com position ZrO 2 doesn’t absorb m icrowave power, reflection power plots overlapped with the data of blank measurement. Stoichiometric composition ZrO2 isn’t an ion conductor, it becomes ion conductor by adding divalent or teivalent atoms such as Ca, Mg, Y and so on. TZ-2Y, TZ-8Y and Y7Z are yttria added zirconia materials, which oxygen ion conductivity increase gradually with the increase of temperature. This tem perature change of ion conductivity possibly rerates to the tem perature change of m icrowave absorbency of oxygen ion conductive zirconia. Summary Microwave absorbency of zirconia powder and fiberwas measured at elevated temperature by using a system consists of a microwave vector network analyzer, a circular wave-guide fixture and a vacuum furnace. As object samples stabilized zirconia powder, partially stabilized zircona powder, zirconia fiber and stoichiometric composition ZrO2 were used. Obtained results are as follows.

12th INTERNATIONAL CERAMICS CONGRESS PART B

Reflection form sample / dB

90

-11

measured at 13.0-14.2GHz in vacuum blank

-12 -13

ZrO2

950mg, t=3.6mm ρ=21.9%T.D.

-14 -15 -16 -17 0

200

400

600

800

1000

o

Temperature / C Fig.7 Microwave absorption mesurment result zirconia (ZrO2, stoichiometric compositon) powder at elevated temperature. Microwave absorbency of oxygen conducting zirconia increases with the increase of tem perature though stoichiometric composition ZrO2 doesn’t absorb microwave power up to 900oC. Therefore, it is considered that m icrowave absorption beha vior of zirconia has strong relation with ion conductivity. Microwave absorbency change of oxygen ion conductorzirconia gradually increases with increase of temperature. Therefore, it is thought that runawa y occurs when the balance of heat generation by microwave absorption and heat diffusion from heated part is broken at a particular temperature. References [1] M. A. Janney, C. L. Calhoun and H. D. Kimmly: J. Am. Ceram. Soc. Vol. 75 (1992) p. 341 [2] S. Sano, T. Bannno, K. Oda, S. Kinoshita, Y. Setsuhara and S. Miyake: Proc. Int. Sym p. On Microwave, Plasma and Thermochemical Processing of Advanced Materials, (1997) p. 101 [3] N. H. Harris, J. R. Chu, R. L. Eisenhat and B. M. Pierce: Ceramic Transactions Vol. 21 (1991) p 235 [4] R. D. Hollinger, V. V. Vardan, V. K. Vardan and D. K. Ghodgaonkar: Ceramic Transactions Vol. 21 (1991) p 243 [5] T. E. Cross and G. A. Dimitrakis, Proc. of 9th AMPERE, (2003) [6] S. Sano, Y. Hotta, T. Banno, K. Oda, Y. Sets AMPERE, (1999) p 111

uhara, Y. Makino and S. Miyake: Proc. of 7th

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.91

Microwave assisted reaction sintering of ZrSiO4/ α -Al2O3 mixtures O. Ertugrul1, a, S. Akpinar2, b, I. M. Kusoglu1, c, K. Onel1, d 1

Dokuz Eylul University, Department of Metallurgical and Materials Engineering/Izmir, TURKEY 2 Afyon Kocatepe University, TURKEY a [email protected], [email protected], [email protected], d [email protected]

Keywords: microwave sintering, zirconia, mullite, zirconium silicate, α -Al2O3.

Abstract. Mullite- zirconia composites have better mechanical properties than monolithic mullite ceramics and can be produced by reaction sintering of ZrSiO4 and α -Al2O3. The samples were prepared from high-purity (99.9%) α-alumina and fine zircon (ZrO2>65 wt.%) powders using PVA as binder. The powder mixtures were compacted under 80 MPa as coin shaped samples by uniaxial dry pressing and then sintered in a multimode microwave field of 2.45 GHz. The microwave effect on ZrSiO4 dissociation and mullite formation was evaluated by comparing the microwave sintered samples with those sintered conventionally. The as-reacted compacts were characterized by X-ray diffraction and scanning electron microscopy (SEM). The effects of sintering parameters on mullitization and mullite grain growth were investigated. Introduction Mullite ceramics posses low thermal expansion, good high-temperature strength, excellent creep resistance and chemical stability, and are suitable for high temperature applications [1-4]. Mullite, however, has low fracture toughness and relatively low strength at room temperature, when compared with other engineering ceramics. Mullite-based composites with dispersed zirconia particles have been widely studied to overcome these disadvantages [5]. These particles toughen the material mainly as a consequence of the tetragonal-monoclinic phase transformation that can occur during the fracture process or during post sintering cooling [6]. Reaction sintering of alumina and zircon (ZrSiO4) is an easy and inexpensive route to obtain homogenous composites containing dispersed zirconia with enhanced mechanical properties, and has been extensively studied by many researchers [7, 8, 9]. However, the fully dense compacts are difficult to achieve, due to the poor sinterability of the mixed powders [5]. Similarly, ZrSiO4/ α -Al2O3 ratio should also have a great influence on the sinterability of ZrSiO4/ α -Al2O3 mixed powders because the sinterability of the two powders is different [5]. To favor both chemical reaction and densification during sintering, various additives are commonly used. Using magnesia leads to a microstructure constituted of interlinked needlelike mullite grains with a dispersion of coarse intergranular (> 1 micron) and fine intragranular (~ 0.1 micron) zirconia. However, large amounts of glassy phase, undesirable to keep favorable mechanical properties and low creep deformation at elevated temperatures, are present in the material [9]. The formation mechanism of mullite from zircon and alumina comprises the following steps: (1) initial formation of glassy phase from the present impurities with some thermally dissociated zircon; (2) dissolution of zircon into this glassy phase; (3) dissociation of zircon and the formation of ZrO2 and higher SiO2 containing glassy phases; (4) dissolution of Al2O3 in this glassy phase and its concentration increase in the melt until the stoichiometric composition of mullite is obtained [10]. The sinterability is greatly dependent on both firing temperature and zircon content, which undergoes complete dissociation in all compositions. Zirconia is the major phase evolved in the

92

12th INTERNATIONAL CERAMICS CONGRESS PART B

studied composites beside alternative amounts of spinel and corundum. Mullite formation is independent of THA Al2O3/SiO2 ratio, but greatly dependent on the type and amount of glassy phase present. The incorporation of lower MgO and Al2O3 and higher ZrO2 contents led to the formation of excellent high temperature multicomposites [10]. In reaction sintering, a mullite matrix with zirconia inclusions or an alumina matrix with mullite and zirconia inclusions can be obtained as the resulting microstructure; in the latter case an additional toughening mechanism is present in the matrix (bridging) if the mullite inclusions appear in a needle shape. The disadvantage of using reaction sintering is that the dissociation reaction of zircon causes porosity that may lead to degradation of the mechanical properties [6]. Up to 1000 oC fracture strength and toughness values are quite high, which make these materials potential candidates for high temperature applications [11]. Besides increased heating rate, uniform heating and energy saving, microwave sintering enables higher density and mechanical properties of ceramics. Stabilized tetragonal zirconia is a transformation toughening agent and also a highly efficient microwave absorber. It can be preferentially heated in a microwave field [12]. The present work aims to produce in-situ zirconia-mullite ceramics by rapid and uniform heating using microwave energy and evaluate the microwave effect on ZrSiO4 dissociation and mullite formation. Experimental work The samples were prepared using high-purity (99.9%) α-alumina and fine zircon powder (66 wt. % ZrO2, 33 wt. % SiO2). The average particle sizes (d50) were 3.43 µm and 1.83 µm, respectively. As sintering aid 1 wt. % MgO with a mean particle size of 10.76 µm was added. The main impurities of the zircon powder were Al2O3 (0.50 wt.%), CaO (0.40 wt.%) and Fe2O3 ( ≤ 0.10 wt.%). The composition of mullite phase consists of SiO2 and Al2O3 in the weight ratio of 28.17 / 71.83 and 2 wt. % polyvinyl alcohol (PVA) was added to this mixture as the binder. The mixture was then ball-milled in distilled water for 8 hours to obtain a homogeneous mixture. After drying at 120 oC the mixture was uniaxially dry-pressed at 80 MPa to form green compacts, which were subsequently sintered using microwave heating and conventional heating. The expected reaction during sintering is given by the following reaction: 2ZrSiO4 + 3Al2O3 → 2ZrO2 + 3Al2O3· 2SiO2 The samples were heated in a conventional electrical resistant furnace to 1550 oC at a rate of 8 oC/ min. and sintered for 2 hours. In microwave sintering, the green compacts were placed in the furnace in a microwave transparent container with SiC susceptors. The temperature of the samples was monitored using an infrared pyrometer with the circular crosswire focused on the sample surface. The emissivity of the pyrometer (Kleiber 273 - LWL) was set to 0.85 in order to measure the exact temperature. Sintering durations were 1 or 2 hours for each sintering temperature. The samples were cooled to 300 oC in 2-3 h after sintering was completed in microwave furnace. The bulk density (ρb) and open porosity of sintered samples were measured by the Archimedes’ method. Phase analysis of the sintered samples was conducted by a standard powder X-ray diffractometer (RIGAKU, Japan). The microstructures and morphologies of sintered samples were observed using a scanning electron microscopy (SEM; JEOL, Japan). Thermal etching was performed on the sintered compacts at a temperature 50 oC below their sintering temperatures for 1 hour before SEM observations.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

93

Results and discussion The microwave sintering (MWS) cycle was considerably shorter than the conventional sintering (CS) cycle. Heating profiles for both conventional and microwave sintering are given in Fig. 1. Heating rate was 20 oC/min for MWS while it was 8 oC/min for CS. The complete cycle for MWS was around 300 min while it was 450 min for CS. Faster rates of both heating and cooling during MWS yields finer microstructures in the end product. The XRD patterns of sintered samples are given in Fig. 2. The phase composition of the conventionally sintered sample (1550oC/2 h) consists of zircon, corundum, mullite, t-ZrO2 and mZrO2 phases. It is clearly seen that not all the zircon phase in the structure was dissociated and transformed at this sintering temperature and duration. However, the phase transformation ratio in the MWS sample sintered at 1450oC for 2 hours is higher than the conventionally sintered one. It is also obviously seen that the dissociation of zircon phase was still not complete at 1450oC. It is another point that a good mullitization is reached while the microwave sintering temperature and duration is low. When the sintering temperature is raised up to 1500oC, there will be no zircon peaks observed at 2Ө-20o and 2Ө-27o. The existing phases in the samples sintered at 1500oC were corundum, mullite, t-ZrO2 and m-ZrO2. In consequence of increasing sintering temperatures in MWS up to 1570oC, some increase in the intensity of mullite phase (2Ө-28o) is observed. However, there is no remarkable change in the intensity of corundum phase (2Ө-35o) above 1500oC. It can be tought that at this sintering temperature the increase of mullite peaks despite corundum phase is due to the growth of mullite grains. Although the sintering temperature is increased, the presence of free corundum in the structure still remains. That is probably the result of inadequate SiO2 ratio in the stochiometry. Considering the XRD patterns given in Fig. 2, it is seen that in the samples microwave sintered at 1500 oC for 2 hours, zircon is dissociated completetly to form ZrO2 and mullite phases.

Figure 1. Heating profiles for MWS, CS.

94

12th INTERNATIONAL CERAMICS CONGRESS PART B

Figure 2. XRD patterns of the samples sintered by conventional heating (CS) and microwave heating (MWS). The physical properties of samples sintered at different temperatures are given in Table 1. Open porosity values in Table 1 indicates that densification is better in microwave sintering than conventional one and increases with increasing sintering temperature. Table 1. Physical properties of sintered mullite ceramic foams. Sintering

Sintering

Temperature

Duration

(oC)

(hour)

1450

1

1500

Specific

Open

Gravity

Porosity

(g/cm3)

(%)

3,138

3,753

16,54

2

3,167

3,697

14,51

1525

2

3,197

3,586

11,04

1550

2

3,138

3,675

14,76

1570

2

3,167

3,525

10,34

1550(CS)

2

3,149

3,817

17,67

ρb (g/cm3)

As seen in Fig 3, BEC images give information on the transformed and untransformed regions. In Fig. 3e and f the micrographs show three different regions representing different constituents. EDS analysis of different regions show that white regions are zirconia, light gray grains are corundum and dark gray matrix is mullite. White grains around the coarse grain in Fig. 3e indicates that transformation to zirconia starts at the boundaries of coarse zircon grains but the transformation of zircon to zirconia is not completed. Fig 3c indicates clearly that coarse zircon grains are completely transformed to fine zirconia grains. However, in Fig 3d there are some small light gray corundum grains. It can also be seen from the microstructures that ZrO2 (white grain) and mullite grain size in MWS is slightly smaller than in those sintered conventionally. Also, in accordance with the results

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

95

of Zhao et al. [13], mullite formation might follow the nucleation and growth mechanism within the amorphous matrix.

(a)

(c)

(e)

(b)

(d)

(f)

Figure 3. SEM micrographs of the samples; a-b) 1500oC-2h/MWS, c-d) 1550oC-2h/MWS, e-f) 1550oC-2h/CS. Conclusions Stochiometric mixtures of ZrSiO4/α-Al2O3 compacts to form in-situ ZrO2-mullite composites were sintered successfully by using microwave energy. Using 960 W of microwave energy the compacts were heated to 1570 oC. The sintering cycle for MWS was considerably shorter than that for CS. Microwave has a significant effect on zircon-mullite transformation even when the sintering temperature is 50oC lower than CS. In-situ Zircon-mullite transformation is completed within 2 hours by microwave sintering at 1500 oC. Some untransformed corundum phase found in the structure, the reason might be related to the starting stochiometric ratios of raw materials. Furthermore, increasing the sintering temperature has positive effect on densification of the

96

12th INTERNATIONAL CERAMICS CONGRESS PART B

compacts. According to SEM images finer zircon particles are supposed to result in more homogeneously distributed zirconia grains in mullite matrix so finer zircon grains should be used as starting raw material. The results indicate that microwave interaction with zircon accelerates zircon dissociation to provide ZrO2 formation. Further studies can be focused on preparing new mixtures with different stochiometric ratios of raw materials for the full transformation to in-situ ZrO2mullite composites. References [1] H. Ohnishi, K. Kriven, T. Nakamura and T. Kawanami: Mullite and Mullite Composites, Ceramic Transaction, vol. 6 (1990), p. 605–612 [2] N.M. Rendtorff, L.B. Garrido and E.F. Aglietti: Ceram. Int. 34 (8) (2008), p. 2017–2024 [3] G.M. Anikumar, U.S. Hareesh, A.D. Damodaran and K.G.K. Warrier: Ceram. Int. 23 (6) (1997), p. 537– 543 [4] P.C. Dokko, J.A. Pask and K.S. Mazdiyasni: J. Am. Ceram. Soc. 60 (3– 4) (1977), p. 150–155 [5] S. Zhao, Y. Huang, C. Wang, X. Huang and J. Guo: Ceramics International 29 (2003), p. 49-53 [6] A.C. Mazzei and J.A. Rodriguez: J. Mat. Sci. 35 (2000), p. 2807-2814 [7] P.Descamps, S. Sakaguchi, M. Poorteman and F. Cambier: J. Am. Ceram. Soc. 74 (10) (1991), p. 2476–2481 [8] T. Koyama, S. Hayashi, A. Yasumori, K. Okada, M. Schmucker and H. Schneider: J. Eur. Ceram. Soc. 16 (2 – 3) (1996), p. 231– 237 [9] J.M. Wu and C.M. Lin: J. Mater. Sci. 26 (17) (1991), p. 4631– 4636 [10] M. Awaad, M. F. Zawrah and N. M. Khalil: Ceramics International Vol. 34 (2) (2008), p. 429434 [11] G. Orange, G. Fantozzi, F. Cambier, C. Leblud, M.R. Anseau and A. Leriche: J. Mat. Sci. 20 (1985), p. 2533-2540 [12] Y. Fang, J. Cheng, R. Roy, D.M. Roy and D.K. Agrawal: J. Mat. Sci. 32 (1997), p. 4925-4930 [13] S. Zhao, Y. Huang, C. Wang, X. Huang and J.Guo: Materials Letters 57 (2003), p. 1716-1722

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.97

Hybrid Foams, Colloids and Beyond: Advanced Ceramics through Integrative Chemistry

Nicolas Brun1,a, Simona Ungureanu1,b , Florent Carn2,c Béatriz Julián-López3,d and Rénal Backov1,e 1

Centre de Recherche Paul Pascal, CRPP-UPR CNRS 8641, 115 Avenue Albert Schweitzer 33600 Pessac, France.

2

Laboratoire Matière et Systèmes Complexes, UMR 7057, Université Denis Diderot-Paris 7, Bâtiment Condorcet, Case Courrier 7056, 10, rue Alice Domon et Léonie Duquet, 75205 Paris Cedex 13, France.

3

Departamento de Quımica Inorganica y Organica ESTCE, UniVersitat Jaume, IAVda., Sos Baynat s/n12071 Castellòn, Spain a

[email protected], [email protected],

c

[email protected], [email protected], [email protected]

Keywords: Integrative chemistry - biliquid foams – porous ceramics -hybrid materials- sol-gel process – photonic s- catalysis.

Abstract. Today chemistry of materials and as such the ceramic field of research are addressed through more and more complex synthetic methodologies in order to optimize final material performances. The notion of complexity in chemical science is illustrated inhere through the concept of integrative chemistry. Particularly the integration between bi-liquid foams, sol-gel process, organo-silane functionnalization, lanthanides complexation and Pd heterogeneous nucleation is proposed as a non-exhaustive synthetic tool box to reach specific advanced ceramics. The first section is dealing with the synthesis of the first series of Si(HIPE) macrocellular foams where the oil volume fraction of the starting emulsion allows a nice tuning of the foams macroporosity. The second section is dealing with Europium complexation of β−diketone and malonamide hybrid Organo-Si(HIPE) leading to the Eu3+@Organo-Si(HIPE) luminescent foams, while the third part is dedicated to Pd heterogeneous nucleation within host hybrid foams. This last series of macrocellular ceramics are labeled Pd@Organo-Si(HIPE) which demonstrates good turn over number (TON) and turn over frequencies (TOF) when acting as supported catalysts for the Mizoroki-Heck coupling reactions. In the above mentioned foams the HIPE acronym is for High Internal Phase Emulsion. Introduction Nowadays chemists are request to envisage materials more and more complex in nature and structure, bearing polyfunctionnality and, at the extreme, being able to develop some degree of autonomy taking inspiration from living organisms. To construct such high standard performing materials, chemical science cannot be restricted to a single specific domain of chemistry but has to be addressed through a strong interdisciplinary approach, crossing thereby chemical boundaries and beyond. In such a context, innovative synthetic pathways are certainly reaching the notion of “complexity” in chemical science [1]. When it turns to chemistry of materials, there is a crucial need for a “rational design” of functional architectures where the first parameters to bear in mind is not the “shaping for the shaping” or even the competence and methodology in use, but rather the enhanced function or real competitive application to be reached from what will then ensue the overall synthetic pathway to be apply. From this way of thinking has recently emerged the concept

98

12th INTERNATIONAL CERAMICS CONGRESS PART B

of Integrative Chemistry [2,3]. This new transversal domain of chemical science be defined as an “interdisciplinary tool box” where advanced functional materials having hierarchical structures can be tailor made via the smart integration of soft chemistry based pathways and the versatile processing conditions offered by soft-matter physical-chemistry. Beyond it offers the possibility of positioning and confining chemical reactors at divers length scales, those reactors acting either in cooperative or partitive action modes [4]. Here we can sense that the chemical reactions should now operate at low temperature in order to preserve either the architectures’ organic counterparts or the soft textural modes addressed during the synthetic routes. When it turns to generate ceramics, solgel process [5], also known as “soft chemistry” is appearing as a candidate of choice to sequester chemical reactions within complex fluid versatile templating modes as for instance; bi-liquid foams concentrated[6] and diluted [7], air-liquid foams [8], extrusion process [9], lyotropoic mesophases [10], multilamellar vesicles [11], Evaporation Induced Self Assembly (EISA) techniques [12] and so forth. Herein we would like to describe non-exhaustive rational synthetic pathways based upon the integration between lyotropic mesophases, direct concentrated emulsions and sol-gel process. The first section will be dedicated to the rational design of purely inorganic silica macrocellular foams where the macropore diameters are tune through varying the oil volume fraction of the starting direct concentrated emulsions. The second and third sections will extend this synthetic route toward the wide hybrid organic-inorganic materials platform [13] where photoluminescent foams and palladium-based heterogeneous macrocellular catalyst will be respectively described.

Sol-gel Process, Lytropic Mesophase and Bi-liquid Foams toward Synthesizing Inorganic Silica Foams (Si-HIPE) Bi-liquid foam is a powerful tool to generate hierarchical porosity when combined with lyotropic mesophases, and divers synthetic routes have been reviewed by Cooper et al [14]. Beyond the syntheses of macroporous solids it seems important to be able to tune the macroscopic void space diameters. To accomplish this study we played with the starting oil volume fraction (ρo) of a direct concentrated emulsion (dispersion of oil droplets within aqueous continuous phase) [15] Results are depicted in Fig. 1.

a)

c)

70 µm

b)

e)

70 µm

d)

70 µm

f) 4 µm

19 µm

13 µm

13 µm

Figure 1. SEM visualization of the inorganic monolith-type material macrostructure. a) and b) 2SiHIPE0.035, c) and d) 3Si-HIPE0.035, e) and f) 4Si-HIPE0.035. RSC copyright 2004.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

99

We can observe that whatever the oil volume fraction conditions, the general texture resembles aggregated hollow spheres. For the 1Si-HIPE0.035 (ρo = 0.67, and pH near 0.04) and 2Si-HIPE0.035 (ρo = 0.70, pH near 0.04) porous monoliths we observe almost the same macrocellular average cell size without any dramatic increase in the cell sizes (not shown here). When we increase the starting emulsion’s oil volume fraction, the macrocellular cell sizes diminish drastically (Fig. 1 c, e and g). In fact, considering the rheology of the bi-liquid foams, it is well known that the viscosity of direct emulsions increases dramatically when the oil volume fraction reaches values above 0.74 [16]. For the starting emulsions of the materials 3Si-HIPE0.035 (ρo = 0.73 and pH near 0.04) and 4Si-HIPE0.035 (ρo = 0.78, and pH near 0.04), this enhanced viscosity increases the shear applied to the oily droplets, which induces smaller macrocellular cells within the solid state replica. Also, when the oil volume fraction is increased-up from 2Si-HIPE0.035 to 3SiHIPE0.035 and 4Si-HIPE0.035 the size of the largest cell junctions decreases from 1.4 to 0.5 and 0.25 µm respectively whereas the average smallest cell junction sizes remains constant at 50 nm for the compounds 2Si-HIPE0.035, 3Si-HIPE0.035 and 4Si-HIPE0.035. Beyond control over the cell diameters playing with the starting oil volume fraction we shown that the Plateau borders morphology could be tune playing with the synthetic pH conditions (Figure 5). At pH 0.5 close to the silica isoelectric point, the inorganic polymer possesses a strong fractal charater and the polymer extends completely within the continuous aqueous phase. When the pH is lowered, the inorganic skeleton is more Euclidian, the dense polymerisation will be this time focussed rather close to the oil-water interface, providing therefore this “hollow spheres” aggregation state. All those porous materials possess a secondary micro-mesoporosity around 800m2.g-1, porosity promoted by using TTAB as concomitant mesoscopic texturing agent. At higher scale it is possible to align the macropores by substitution of the basic oil (dodecane) with hydrophobic ferrofluid and applying an external magnetic field during the condensation process [17]. Eu3+@Organo-Si(HIPE) Macro-Mesocellular Hybrid Foams Generation and Photonic Properties Herein, we intended to take benefit of β-diketone and malolamide organosilane derivatives to complex lanthanide ions within new Organo-Si(HIPE) matrices (Fig. 2).

a)

30 mm

b)

c) 5 mm

7 mm

Figure 2. As synthesized Organo-Si(HIPE) monoliths. a) Eu3+@gβ-diketone-Si(HIPE) (left) and Eu3+@gmalonamide-Si(HIPE) (right), b) Eu3+@gmalonamide-Si(HIPE) when irradiated under UV light (350 nm), c) Eu3+@gβ-diketone-Si(HIPE) when irradiated under UV light (350 nm). ACS Copyright 2008.

100

12th INTERNATIONAL CERAMICS CONGRESS PART B

Due to their photophysical properties, the design of efficient lanthanide complexes as molecular devices became an important issue in the 1990s as their photophysical properties has grown considerably since Lehn [18] proposed the chelation of lanthanide ions with different ligand families. We have prepared [19] a first set of luminescent foams via a two-step process in which βdiketone or malonamide organosilane derivates were grafted to a previously prepared macrocellular Si(HIPE). A third material was obtained via a “one-pot” co-condensation of the silica precursor (tetraethyl-orthosilane, TEOS) and a trialcoxysilylated β-diketone precursor. Luminescence studies of the Eu3+-impregnated materials were carried out in order to get information about the lanthanide environment, about the efficiency in the complexation depending on the synthetic strategy, and their suitability as luminescent material. From each synthetic pathway we obtained self standing monolith-type materials showing luminescent properties upon UV light exposure (Fig. 2). Roomtemperature excitation spectra of Eu-doped Si(HIPE) materials monitored around the more intense emission line of the Eu3+ at 615 nm are shown in Fig. 3. With the aim of comparing the optical features, the spectra were normalized by one of the bands located out of the matrix or complex absorption, i. e. at 465 nm.

Figure 3. Excitation spectra of the Eu-doped organically derived Si(HIPE) hybrid materials monitored the Eu3+ emission at 615nm. ACS Copyright 2008. All spectra exhibit the signals characteristic of the f-f transition levels (assigned in the figure) of the lanthanide ion. However depending on the nature of the complexing group, important differences in intensity and in the profile are appreciated. Indeed, the spectra of both β-diketone derived compounds are dominated by the intense absorption in the UV and blue region, that can be assigned to the π→π* excitation of the ligand, thus reducing the direct excitation of the Eu3+ energy levels. This behavior is especially important in the sample prepared by a co-condensation of the TEOS and organically-derived precursors, indicating that the charge of chelating agent in the Si(HIPE) foam is higher, as already revealed by elemental analysis. In this situation, the energy is mainly absorbed by the lowest triplet level of the β-diketone ligand and efficiently transferred to the resonance energy level of the central Eu3+ ion. In contrast, for the Eu@gmalonamide-Si(HIPE) sample, a weaker band is detected around 275 nm. It indicates that the excitation bands of Eu(III) in the UV region are less affected by the complexation with malonamide groups. It could be explained by the low energy overlapping between the absorption levels of Eu and this ligand, as observed in

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

101

the absorption spectrum. The photoluminescence emission spectra (PLE) of the Eu3+-organically derived Si(HIPE) materials were obtained using two different excitation wavelengths corresponding to the absorption maximum positions of the ligands (270 and ~305 nm for malonamide and βdiketonate, respectively) and the most intense excitation of the Eu3+ (396 nm). As an example, the spectra registered for the Eu@β-diketone-Si(HIPE) sample is shown on the left side of Figure 4. They consist on a set of emission lines corresponding to the intra 4f electronic transitions from the lowest excited state, 5D0, to the ground state manifold, 7FJ (J=0-4). Upon ligand excitation, from the analysis of the absorption bands (not shown here), a great enhancement of the PLE emission is detected for the sample prepared in a one-pot process in comparison with the analogue synthesized by the grafting process. This result can be explained in terms of the more effective energy transfer between the complexing ligand and the Eu3+ ion. This effect is really interesting for the design of light emitting devices since the aim is to find a high efficiency in the PLE spectrum. For the Eu3+@gmalonamide-Si(HIPE), a high emission intensity is also observed but upon excitation at the Eu 5L6 level, in accordance with the excitation spectrum (emission spectra not shown). Thus, depending on the excitation wavelength, the stronger PL intensity can be achieved through complexation by the β-diketone (exc. at 306 nm) or the malonamide chelate (exc. at 396 nm). In addition, some of the Eu(III) emission bands being sensitive to changes in the first coordination sphere are particularly useful analytical tools . In particular, the intensity of the electric dipole 5D0→7F2 transition (~615 nm) is known to be hypersensitive and, consequently, each spectrum has been normalized against the magnetic dipole 5D0→7F1 transition, located at ~590 nm, which is not affected by the environment of the fluorescent ion. The normalized spectra of the three samples after ligand excitation are depicted on the right in Figure 4.

Figure 4. Emission spectra of (left) the Eu@β-diketone-Si(HIPE) upon Eu3+ excitation, and (right) those of the three samples upon ligand excitation after normalization. ACS Copyright 2008. The intensity of the electrical dipole transition increases when the lattice environment is distorted and contains certain components of non-inversion symmetry. Thus, the ratio between the areas of the 5D0→7F2 and 5D0→7F1 transition bands can be considered as a parameter to probe the “asymmetry” of the Eu sites and it is a good measure of the strength of the interaction between Eu(III) and the ligands. The spectra were fitted by Gaussian curves and the ratio values obtained were 6.8, 4.7 and 3.5 for the Eu3+@β-diketone, Eu3+@gβ-diketone and Eu3+@gmalonamide-

102

12th INTERNATIONAL CERAMICS CONGRESS PART B

Si(HIPE)s, respectively. Note that these values were very similar to those obtained upon excitation at 396 nm, what indicates that Eu ions are mainly occupying a unique site in each system. Comparing the sensitive ratio for the two samples with β-diketone ligands indicates that Eu ions are located in a more asymmetric environment when prepared via a one-pot process than via the grafting one. Also, decay curves were recorded upon chelate excitation in the UV region or upon direct excitation of Eu ions at 394 nm, with the fixed detection at the most intense emission wavelength of Eu3+ (615 nm). The luminescent relaxation profiles (figures not shown) were almost single exponential (ln I(t) α -1/τ, were I(t) is the luminescence intensity in function of time and τ is the lifetime value), indicating that the Eu ions occupy the same average coordination environment. The lifetime values obtained by fitting are given in Table 1. Sample

Eu@βdiketoneSi(HIPE)

Eu@gβdiketoneSi(HIPE)

τ (µs) 110 130 Exc. 394Eu τ (µs) 150 150 Exc. ligand Table 1. Lifetime values of the Eu3+@Organo-Si(HIPE) samples.

Eu@gmalonamideSi(HIPE) 450 460

Overall, we can consider that europium ions are mostly complexed by the chelates, where the dual presence of OH groups and divers ligands nature could be responsible for the differences observed in the decay times. Further investigations would be necessary to reveal more precise information on this subject. Moreover, this new series of luminescent Eu3+ complexed OrganoSi(HIPE) materials are bearing a wide range of promising applications in catalysis, optics, sensors, absorbers etc.

Pd@Organo-Si(HIPE) Open-Cell Hybrid Monoliths Generation Offering Cycling Heck Catalysis Reactions Palladium heterogeneous nucleation within the macrocellular Organo-Si(HIPE) materials [20] have been performed (Fig. 5a). The resulting supported catalysts were called Pd@gAminoSi(HIPE), Pd@gMercapto-Si(HIPE) and Pd@Mercapto-Si(HIPE) depending on the starting Organo-Si(HIPE) material employed [21]. Upon the synthetic route in use, it was observed that the monoliths are macroporous (Fig. 5a) and become homogeneously black from the outer to the inner part of the monoliths. This feature is an indication that Pd nanoparticles have been macroscopically homogeneously nucleated within the Organo-Si(HIPE) compounds (Fig. 5b). The chosen salt Pd(OAc)2 reduction occurs through using NaBH4 in a THF/H2O (50 /50 v/v) mixture in the presence of triphenylphosphine (PPh3) as zero-valent state nanoparticles stabilizing agent [22] (4/1 PPh3/Pd ratio). Fig. 5c shows that the nanoparticles have been generated, with average size diameters around 5-10 nm. In fact, if non stabilized Pd nanoparticles are good enough for operating under reducing atmosphere, especially for heterogeneous hydrogenation catalysis reactions [23], their capability to promote good catalytic yields under oxidative conditions is minimized and needs combination of matrices stabilization, using entities bearing amino or mercapto groups, and the addition of, for instance, triphenylphosphine as a co-stabilizing agent [23]. Herein, the XPS spectrum of the grafted Pd shows two main peaks centred, at 335 eV and 340.8 eV respectively corresponding to the 3d5/2 and 3d3/2 of metallic zero-valent Pd nanoparticles (Fig. 5d). The Pd (Weight %) loading was 3.9%, 4.1% and 3.9% for Pd@gAmino-Si(HIPE), Pd@gMercaptoSi(HIPE) and Pd@Mercapto-Si(HIPE) respectively as estimated by Elemental Analysis.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

103

b)

a)

2µm

d) Counts ( ar bitrary units)

c)

100 nm 325

330

335

3 40

345

3 50

355

Binding ener gy (eV)

Figure 5. a) Textural macroporosity observed through Scanning Electron Microscopy (SEM), b)Typical monolith after the Pd nanoparticles heterogeneous nucleation (the white arrow indicates the monolith inner part), c) Pd nanoparticles observed through Transmission electron Microscopy (TEM), d) Typical X-Ray Photoelectron Spectroscopy (XPS) performed on the Pd@OrganoSi(HIPE) compounds, focussed on the bands centred at 335 eV and 340.8 eV that correspond respectively to the Pd 3d5/2 and 3d3/2. ACS Copyright 2008. Palladium supported monoliths were then used as catalysts for the Mizoroki-Heck coupling(Fig. 6a) reaction between iodobenzene 1 and styrene 2 using triethylamine as a base and DMF as solvent. The reactions were performed at 155°C in a closed reactor with a lateral frit (Fig. 6b). Conversions of styrene (2) and iodobenzene (1) in (E + Z) stilbene (3+4) were followed by Gas Phase Chromatography (GPC). Each of the batch experiments was conducted for 3 h, then the liquid medium was filtered and a new reactive mixture added to the remaining supported catalyst in order to perform a new catalytic run. The Pd/iodobenzene molar ratio was settled at 0.004 and 0.002. The first comment refers to the high selectivity observed in all cases for the E product isomer (E/Z : 96/4). Considering kinetics reported on Fig. 6c, all catalysts tested behave similarly on the first use; conversion being close to completion in all cases after 3 hours. However, under recycling, supported catalysts bearing a mercapto group appear to be less sensitive to deactivation/leaching than those functionalized with an amino group. For instance mercapto-derivatives are still active on the seventh cycle bearing a conversion yield of 97%. These results seem to confirm the previously reported observation that a mesoporous silica modified with a mercapto-propyl group provides a fairly good scavenging power toward Pd nanoparticles, thus reducing as far as possible Pd leaching in heterogeneous catalysis of the Mizoroki-Heck reaction [24]. Beyond, in order to investigate the limits of these monolith catalysts toward their cycling performance, runs were pursued with Pd@Mercapto-Si(HIPE) support, showing a slow decreases in conversion yield from 92 to 75 % for the eighth and ninth run respectively.

104

12th INTERNATIONAL CERAMICS CONGRESS PART B

a)

b) I

Pd / SS

+

Et3N, Et3 N, DMF DM F 1 55°C 155°C

11

3 :: E 3 E 4 :: Z Z

2

c) cycle 1

cycle 2 cycle 4 cycle 6 cycle 3 cycle 5 cycle 7

cycle 8 cy cle 9

C onversion yie ld (%)

100

80

60

40

20

0

0

200

40 0

600

800

10 00

1 200

1400

16 00

Time ( minutes)

Figure 6. a) Expression of the Mizoroki-Heck coupling reaction between the starting iodobenzene and styrene, b) typical reactor in use for the catalysis reactions. c) Cycling Heck coupling reactions and conversion yields.
Pd@gAmino-Si(HIPE),
Pd@gMercapto-Si(HIPE),
Pd@MercaptoSi(HIPE)  Pd@gAmino-Si(HIPE), in this case, 0.055 g of support were used instead of the 0.11 g used for all the other tests. Conversion yields are the average of two GPC analyses. ACS Copyright 2008. In order to further explore the catalytic properties of these hybrid materials limiting the Pd content, a new set of catalytic cycling was performed using Pd@gAmino-Si(HIPE) support where the Pd/iodobenzene molar ratio was decreased to 0.002 (instead of 0.004). Corresponding TON and TOF numbers have been calculated and are reported in Table 2. Samples

TON

TOF (h-1)

gAmino-Si(HIPE) [a] gAmino-Si(HIPE) [b] gMercapto-Si(HIPE) [a] Mercapto-Si(HIPE) [a]

1300 2961 1409 2783

62 141 78 103

Table 2 The M-H reaction of iodobenzene with styrene over different catalysts synthesised. [a] using a 0.004 Pd/iodobenzene molar ratio, [b] using a 0.002 Pd/iodobenzene molar ratio

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

105

The values obtained, in the good average for heterogeneous catalysis of Mizoroki-Heck reaction performed with powdered compounds [24], are reaching the best results for silica-based supports, considering the work from Crudden et al [25]. Despite this honourable performance, it can be noticed that, in any case, the conversion yield, being constant to a high level in the first runs, slowly decreases after 7 cycles. These observations suggest that the grafted hybrid derivatives allow some Pd leaching while performing cycling catalysis. In fact, palladium content analysis of the reaction media after each cycle using Pd@gAmino-Si(HIPE) (0.002 Pd/iodobenzene molar ratio), confirm this assumption: 3.5ppm, 3.5ppm, 4ppm, 5ppm and 15ppm of Pd were detected in solution after cycle 1 to cycle 5 respectively. The literature dealing with Mizoroki-Heck catalysed reaction has clearly established that, in most cases, heterogeneous catalysts are merely reservoirs for highly active soluble forms of Pd [27] from where the metal can be leached in solution to perform an homogeneous catalysis. Pd leached in solution is considered to re-precipitate on the support surface after completion of the reaction [28]. The final slow activity decrease of the support after several reuses arises from successive Pd loss by solution leaching in each cycle of the run coming from the insufficient ability of the soluble Pd species to reintegrate the support macroporosity at the end of the reaction. Summary The notion of complexity in chemical science is illustrated inhere through the concept of integrative chemistry when applied tot the generation of advanced funtional ceramics. Particularly the integration between bi-liquid foams, sol-gel process, organo-silane functionnalization, lanthanides complexation and Pd heterogeneous nucleation is proposed as a non-exhaustive synthetic tool box to reach specific advanced ceramics. The first section is dealing with the synthesis of the first series of Si(HIPE) macrocellular foams where the oil volume fraction of the starting emulsion allows a nice tuning of the foams macroporosity. The second section is dealing with Europium complexation of β−diketone and malonamide hybrid Organo-Si(HIPE) leading to the Eu3+@Organo-Si(HIPE) luminescent foams. The third part is dedicated to Pd heterogeneous nucleation within host hybrid macrocellular ceramics. This last series of porous materials are labeled Pd@Organo-Si(HIPE) which demonstrates good turn over number (TON) and turn over frequencies (TOF) when acting as supported catalysts for the Mizoroki-Heck coupling reactions. References [1] G.M. Whitesides and R.F. Ismagilov: Sience vol. 284 (1999), p. 89. [2] R. Backov: Soft Matter vol. 2 (2006), p. 452. [3] E. Prouzet, S. Ravaine, C. Sanchez and R. Backov: New J. Chem. vol. 32 (2008), p. 1284. [4] R. Backov: L’Acualité chimique vol. 329 (2009), p.III. [5] C.J. Brinker and G.W. Scherrer, in :The physics and chemistry of sol-gel processing, Academic, san Diego (1990) [6] F. Carn, M.-F. Achard, O. Babot, H. Deleuze, S. Réculusa and R. Backov: J. Mater. Chem. vol 15 (2005), p. 3887. [7] M. Destribats, V. Schmitt and R. Backov: Langmuir vol.26 (2010), p.1734. [8] F. Carn, P. Masse, H. Sadaoui, B. Julian, H. Deleuze, S. Ravaine, C. Sanchez, D.R. Talham and R. Backov: Langmuir vol. 26 (2006), p. 5469. [9] H. Serier, M.-F Achard, N. Steunou, J. Maquet, J. Livage, C. Leroy, O. Babot and R. Backov: Adv. Funct. Mat. Vol. 16 (2006), p. 1745.

106

12th INTERNATIONAL CERAMICS CONGRESS PART B

[10] L. Lecren, T. Toupance and R. Backov: Materials letters vol. 59 (2005), p. 817. [11] O. Regev, R. Backov and C. Faure: Chem. Mat. Vol. 16 (2004), p. 5280. [12] C. Boissière, D. Grosso, H. Amenitsch, A. Gibaud, A. Coupe, N. Bacile and C. Sanchez: Chem. Commun. (2003), p. 2798. [13] C. Sanchez and F. Ribot: New J. Chem. vol 18 (1994), p. 1007. [14] H. Zhang and A.I. Cooper: Soft Matter vol. 1 (2005), p. 107. [15] F. Carn, A. Colin, M-.F. Achard, M. Pirot, H. Deleuze and R. Backov: J. Mat. Chem. vol. 14 (2004), p. 1370. [16] T.G. Mason, J. Bibette and D.A. Weitz: J. Colloid Interface Sci. vol.179 (1996), p. 439. [17] F. Carn, A. Colin , V. Schmidt, F.-L. Calderon and R. Backov: Colloids and Surfaces A: Physicochemical and Engineering Aspects vol. 263 (2005), p. 341. [18] J.-M. Lehn: Angew. Chem. vol. 102 (1990), p.1347. [19] N. Brun, B. Julian-Lopez, P. Hesemann, L. Guillaume, M.-F. Achard, H. Deleuze, C. Sanchez and R. Backov : Chem. Mater. Vol. 20 (2008), p. 7117. [20] S. Ungureanu, M. Birot, L. Guillaumme, H. Deleuze, O. Babot, B. Julian, M.-F. Achard, M. I. Popa, C. Sanchez and R. Backov: Chem. Mater. Vol. 19 (2007), p. 5786. [21] S. Ungureanu, H. Deleuze, M. I. Popa, C. Sanchez and R. Backov: Chem. Mater. vol. 20 (2008), p. 6494. [22] A. Desforges, H. Deleuze, O. Mondain-Monval and R. Backov: Ind. Eng. Chem. Res. vol. 44 (2005), p. 8521. [23] A. Desforges, R. Backov , H. Deleuze and O. Mondain-Monval: Adv. Funct. Mat. vol. 15 (2005),p. 1689. [24] O.Aksin, H.Turkmen, L. Artok, B. Cetinkaya, N. Chaoyinj, O. Buyukgungor, E.J. Ozkal: Organomet. Chem. vol. 691 (2006), p. 30227. [25] C.M. Cruddden, M. Satteesh and R. Lewis: J. Am. Chem. Soc. vol. 127 (2005), p. 10045. [26] H. Katayama, M. Nagao, F. Ozawa, M. Ikegami, T. Arai: J. Org. Chem. vol. 71 (2006), p. 2699 [27]C.C. Cassol, A.P. Umpierre, G. Machado, S.I. Wolke, J. Dupont: J. Am. Chem. Soc. vol 127 (2005), p. 3298.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.107

Panoscopic Assembling of Ceramic Materials for High Performance UV-ray Shielding Application Tsugio SATOa, Xiangwen LIUb and Shu YINc Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, JAPAN a

[email protected], [email protected], [email protected]

Keywords: Ceria nanoparticles, Plate-like titanate, Panoscopic assembling, UV-shielding, Comfort

Abstract. The hybrid materials consisting of plate-like potassium lithium titanate (K0.81Li0.27Ti1.73O4) micro particles coated with calcia-doped ceria (Ce0.8Ca0.2O1.8) nano particles were prepared by the co-precipitation method and sol-gel method. Broad-spectrum UV-shielding composite materials with good comfort and low oxidation catalytic activity were successfully synthesized. The comfort when applied on skin and UV-shielding ability of the composites prepared by the sol-gel method were superior to those by the co-precipitation method. Introduction Panoscopic assembling is the hierarchical structure control of ceramic materials from nano structure to macro structure [1]. It is expected that such assembling may result in the creation of new functional materials possessing the synergistically integrated functions of nano material, micro material and macro material. In response to the damages caused by UV-rays various kinds of UV-shielding materials have been designed. In order to cut off UV-ray less than 400 nm in wavelength by inorganic materials, the electric transition of semiconductors possessing bandgap energy about 3 eV is used. In addition, UV shielding materials are also expected to be transparent in the visible light range. The light scattering can be depressed by decreasing particle size, and the particles become quite transparent when the particle size becomes less than 20 nm, since the Rayleigh scattering (s) is in proportion to the particle size to the sixth power as shown by eq. (1). s = Kd6/λ4 (1) K: constant, d: grain size, λ: light wavelength Therefore, nanoparticles of zinc oxide (ZnO) and titania (TiO2) possessing the band gap energies of ca. 3 eV are widely used as inorganic UV-shielding materials in personal products, but their high refractive indices can make the face look unnaturally white, and their photocatalytic activities which can facilitate the generation of reactive oxygen species cause the safety worry [2]. Recently, the calcia-doped ceria (Ce1-xCaxO2-x) nanoparticles which have lower refractive index, low photocatalytic activity, low oxidation catalytic activities and broad-spectrum UV-shielding ability have also been developed as the UV-shielding material [3-7]. However, the comfort and covering ability of inorganic nanoparticles are generally modest due to the agglomeration. In contrast, the plate-like micro particles such as mica and talc show excellent comfort when applied on skin [8] and have been used to improve the comfort of nanoparticles, but the combination of UV-shielding inorganic nanoparticles and these plate-like micro particles generally results in decreasing the UV-shielding ability, sicne they do not possess UV-shielding ability. We reported that the comfort of calcia-doped ceria, Ce0.8Ca0.2O1.8, nanoparticles could be improved by coupling with lepidocrocite type plate-like titanate, K0.81Li0.27Ti1.73O4, micro particles which possesses UV-shielding ability in stead of general plate-like particles, however, it was difficult to coat Ce0.8Ca0.2O1.8 nanoparticles effectively on plate-like K0.81Li0.27Ti1.73O4 micro particles. In the present study, Ce0.8Ca0.2O1.8 nanoparticles were coated on the surface of palte-like K0.81Li0.27Ti1.73O4 micro particles by two different methods, coprecipitation methods and sol-gel methods, and the UV-shielding performance of the products were evaluated.

108

12th INTERNATIONAL CERAMICS CONGRESS PART B

Experimental Sample preparation: The plate-like K0.81Li0.27Ti1.73O4, was prepare by flux method as follows. The powders of K2CO3, Li2CO3 and TiO2 (anatase form) in molar ratio of 3:1:13 and 50 mass% KCl flux were mixed intimately and the mixture was placed in a Pt crucible and then heated up to 1000 °C at a heating rate of 600 °C h-1. The temperature was fixed at 1000 °C for 5 h and then decreased to room temperature naturally. The product was washed with boiling water and filtrated to remove the KCl flux and dried at 150oC for 2 h. Synthesis of the K0.81Li0.27Ti1.73O4 /Ce0.8Ca0.2O1.8 composites by coprecipitation method was as follows. After dispersing the desired amount of plate-like K0.81Li0.27Ti1.73O4 micro particles into deionized water at 40°C by ultrasonic irradiation, 1 M NaOH aqueous solution and 0.8 M CeCl3-0.2 M CaCl2 mixed aqueous solution were simultaneously dropped, where the solution was vigorously stirred and the pH value was kept at a certain value throughout the reaction. Then, the desired amount of 2 M H2O2 solution was dropwise added. The slurry was filtrated and washed with water and methanol in turn for 3 times. Finally, the as-prepared precipitate was calcined at 700 °C for 1 h to coat white Ce0.8Ca0.2O1.8 nanoparticles on the surface of K0.8Li0.27Ti1.73O4 micro particles. Synthesis of K0.8Li0.27Ti1.73O4/Ce0.8Ca0.2O1.8 composites by sol-gel method was as follows. After dissolving Ce(NO3)3·6H2O (1.34 mmol) and Ca(NO3)2·4H2O (0.34 mmol) into 40 mL absolute ethanol at 40°C, 10 mL glacial acetic acid (HAc) was added under vigorous stirring with an HAc/(Ce+Ca) molar ratio of 10. After stirring the solution for 30 min, 1 g K0.81Li0.27Ti1.73O4, which was dispersed into 10 mL absolute ethanol, was dropwise added to the solution. Then, the solution was heated at 60oC for ca. 2 h to get gel-like substance. This gel was dried in a vacuum oven at 333 K for 5 h. Then, the product was collected and ground in an agate mortar followed by calcination at 800oC for 2 h to remove organic materials. Analysis: The morphology of K0.81Li0.27Ti1.73O4/Ce0.8Ca0.2O1.8 composites were evaluated by a field-emission scanning electron microscope (FE-SEM; Hitachi: S-4800). The catalytic ability for the oxidation of organic material was determined by a conductometric determination method (Rancimat method) [9] using cosmetic grade castor oil as an oxidized material. The sample powder (0.5 g) was mixed with the castor oil (10 g) and set at 120°C with bubbling 10 L h-1 of air, where the air from the reactor was introduced into 100 mL deionized water. The oxidation catalytic ability of sample was determined by measuring the increase in the electric conductivity of deionized water by dissolving the volatile molecules coming from the oxidation of castor oil on heating. The UV-shielding abilities of the prepared samples were evaluated by measuring the transmittance spectra of thin films uniformly dispersed the sample powders with an UV-Vis spectrophotometer (Shimadzu; UV-2450), where 0.5 g of the sample, 1 g of nitrocellulose of industrial grade, 2.5 g of ethyl acetate and 2.25 g of butyl acetate were mixed uniformly by ball milling with a plastic container and 10 g of zirconia bead with 1.0 mm in diameter for 40 h. Then, the dispersion mixture was applied onto a quartz glass plate to form the film of 12.5 µm in thickness with an applicator. The feeling of wearing the composite powder on the skin was evaluated by measuring the dynamic friction coefficient using a friction tester (Katotech: KES-SE) which employs a piano wire as a friction sensor and the artificial leather as the substrate where the sample was uniformly applied on using a make-up brush. Results and discussion It is well known that the hetero coagulation of the particles possessing different surface charges is useful to prepare coupled ceramics. The zeta-potentials of pate-like K0.81Li0.27Ti1.73O4 and Ce0.8Ca0.2O1.8 are shown in Fig. 1 as function of solution pH. It can be seen that the isoelectric points of pate-like K0.81Li0.27Ti1.73O4 and Ce0.8Ca0.2O1.8 are around 2.5 and 7, respectively. Therefore, it is

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

109

expected that negatively charged K0.8Li0.27Ti1.73O4 and positively charged Ce0.8Ca0.2O1.8 particles are strongly combined by electrostatic interaction at pH between 2.5 and 7.

Fig. 1 Isoelectric points of (a) K0.81Li0.27Ti1.73O4 and (b) Ce0.8Ca0.2O1.8 nanopaericles. The scanning electron micrographs of K0.81Li0.27Ti1.73O4/Ce0.8Ca0.2O1.8 composites prepared at pH 6.5 and 12 are shown in Fig. 2. As expected, at pH 6.5 the surface of plate-like K0.81Li0.27Ti1.73O4 was homogeneously covered with Ce0.8Ca0.2O1.8 nanoparticles. In contrast Ce0.8Ca0.2O1.8 nanoparticles were agglomerated at pH 12 and could not homogeneously cover the surface of the plate-like K0.81Li0.27Ti1.73O4. It can be seen that the surface of the plate-like K0.81Li0.27Ti1.73O4 micro particle is covered with Ce0.8Ca0.2O1.8 nanoparticles of about 15-20 nm in diameter. The nanocomposite before calcinations at 700oC was pale yellow and that after calcinations changed to whitish.

Fig. 2 Plate-like K0.81Li0.27Ti1.73O4/40 mass% Ce0.8Ca0.2O1.8 composites prepared by coprecipitation at pH (a), (c), (d) 6.5 and (b) 12.0. The composites were also prepared by sol-gel method. Typical SEM images of the composites coated with various amounts of Ce0.8Ca0.2O1.8 nanoparticles by one step sol-gel method are shown in Fig. 3. The surface of the composite with a 10 mass % Ce0.8Ca0.2O1.8 was smooth, but the part of surface of plate-like K0.81Li0.27Ti1.73O4 was uncovered. Some large agglomerates of calcia-doped ceria nanoparticles were formed on the surface with 30 wt% of Ce0.8Ca0.2O1.8. In contrast, the surface of the composite with 20 mass % of Ce0.8Ca0.2O1.8 was smooth and uniform indicating less agglomeration of Ce0.8Ca0.2O1.8 nanoparticles. The grain size of Ce0.8Ca0.2O1.8 nanoparticles coated on plate-like K0.81Li0.27Ti1.73O4 was ca. 10~20 nm. Therefore, the optimal loading content of Ce0.8Ca0.2O1.8 nanoparticles by one step coating was determined to be 20 mass %. The plate-like K0.81Li0.27Ti1.73O4/

110

12th INTERNATIONAL CERAMICS CONGRESS PART B

Ce0.8Ca0.2O1.8 composites with higher amounts of Ce0.8Ca0.2O1.8 loading were obtained by muliti step-coating with repeating the optimal coating course , i.e., 20 mass % Ce0.8Ca0.2O1.8 coating.

Fig. 3 Plate-like K0.81Ti1.73Li0.27O2/ Ce0.8Ca0.2O1.8 composites prepared by one step sol-gel method. The amounts of Ce0.8Ca0.2O1.8 coated on K0.81Ti1.73Li0.27O2 (mass %):(a) 0, (b) 10, (c) 20, (d) 30. In order to evaluate the comfort when sample powders are applied to skin, the kinetic friction coefficients of the artificial leather before and after applying sample powders on were determined using a friction tester, where the relationship among the friction force (f), dynamic friction coefficient (µ) and applied force (N) to the material after moving can be written by eq. (2).

f = µN

(2)

Figs. 4 and 5 show the change in the dynamic friction coefficients (µ/µo) of the samples prepared by coprecipitation method and sol-gel method, respectively, where µo and µ are the dynamic friction coefficient of artificial leather before and after applying the sample power on, respectively. The plate-like K0.81Li0.27Ti1.73O4 showed lower kinetic friction coefficient than those of nanoparticles of Ce0.8Ca0.2O1.8 and titania. As expected, the kintetic friction coefficient of Ce0.8Ca0.2O1.8 nanoparticle could be decreased by coupling with plate-like K0.81Li0.27Ti1.73O4, indicating that plate-like K0.8Li0.27Ti1.73O4 is useful as a means to improve the comfort of applying K0.81Li0.27Ti1.73O4 nanoparticles on the skin. It can be seen in Fig. 4 that the kinetic friction coefficients of the composites containing 10-40 wt% of Ce0.8Ca0.2O1.8 nanoparticles prepared by coprecipitation method are lower than that of plate-like K0.81Li0.27Ti1.73O4. It may be due to the rolling effect by spherical particles. In addition, the dynamic friction coefficient of the composite prepared at pH 6.5 is lower than that at 12.0. These results agreed with the more uniform distribution of Ce0.8Ca0.2O1.8 nanoparticles at pH 6.5 than that at pH 12.0 (see Fig. 2). The increase in the kinetic friction coefficient by increasing the amount of Ce0.8Ca0.2O1.8 nanoparticles up to 70 wt% is due to the change in the friction mechanism from the rolling friction to slip friction. The composites contained 15-60 mass % of Ce0.8Ca0.2O1.8 nanoparticles by sol-gel method also showed lower friction coefficient than that of plate-like titanate. These behavior was almost identical to that of the samples prepared by the coprecipitation method. The UV-shielding material for cosmetics is required to show low activity for the oxidation of organic materials. The catalytic ability for oxidation of organic material was determined by a conductometric determination method, so-called Rancimat method, using cosmetic grade castor oil as an oxidized material. The sample powder was mixed with castor oil and set at 120oC with bubbling air, where the air was introduced into deionized water. The catalytic ability was determined by

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

Fig. 4 Dynamic friction coefficients of various samples prepared by coprecipitation method at pH 6.5 and 12.0. A: TiO2 (Degussa P25), B: Ce0.8Ca0.2O1.8 (pH=12.0), C: K0.81Ti1.73Li0.27O4 before coating; D: K0.81Ti1.73Li0.27O4/10 mass % Ce0.8Ca0.2O1.8 (pH 6.5); E: K0.81Ti1.73Li0.27O4/ 40 mass% Ce0.8Ca0.2O1.8 (pH 6.5); F: K0.81Ti1.73Li0.27O4/40 mass% Ce0.8Ca0.2O1.8 (pH12.0); G: K0.81Ti1.73Li0.27O4/ 70 mass % Ce0.8Ca0.2O1.8 (pH 6.5)

111

Fig. 5 Dynamic friction coefficients of various samples prepared by sol-gel method. A: TiO2 (Degussa P25), B: Ce0.8Ca0.2O1.8 (pH=12.0), C: K0.81Ti1.73Li0.27O4 before coating; D: K0.81Ti1.73Li0.27O4 /10 mass % Ce0.8Ca0.2O1.8, E: K0.81Ti1.73Li0.27O4/15 mass % Ce0.8Ca0.2O1.8, F: K0.81Ti1.73Li0.27O4/20 mass % Ce0.8Ca0.2O1.8, G: K0.81Ti1.73Li0.27O4/40 mass % Ce0.8Ca0.2O1.8, H: K0.81Ti1.73Li0.27O4/40 mass % Ce0.8Ca0.2O1.8.

measuring the increase in the electric conductivity of water by trapping the volatile molecules coming from the oxidation of castor oil on heating. Fig. 6 shows the results of the Rancimat test. It can be seen that the increase in the electric conductivity, that means the oxidation catalytic activity of as-prepared sample increased with increasing the amount of calcia-doped ceria nanoparticles. The activity of the sample prepared at pH 6.5 is higher than that at pH 12. It may be due to the dissolution of calcium ion from ceria at low pH. The oxidation catalytic activity of both samples could be greatly decreased by calcination at 700oC. The oxidation catalytic activities of the composites by sol-gel method were quite low, since the samples were calcined at 800oC (data are not shown here).

Fig. 6 Results of the Ransimat test for the evaluation of the oxidation catalytic activities of the K0.81Li0.27Ti1.73O4/ Ce0.8Ca0.2O1.8 composites prepared by coprecipitatiojn method. A: Before calcination, B: After calcination at 700oC for 1 h. The amounts of Ce0.8Ca0.2O1.8 coated on K0.81Ti1.73Li0.27O2 (mass %):(a) 0, (b) 7, (c) 20, (d) 30, (e) 40, (f) 50.

112

12th INTERNATIONAL CERAMICS CONGRESS PART B

The UV-Vis transmittance spectra of the thin films of plate-like K0.81Li0.27Ti1.73O4/ Ce0.8Ca0.2O1.8 composites containing various amounts of Ce0.8Ca0.2O1.8 nanoparticles by sol-gel method are shown in Fig. 7. The onsets of absorption of the plate-like K0.81Li0.27Ti1.73O4 and Ce0.8Ca0.2O1.8 were ca. 330 and 400 nm, respectively. The Ce0.8Ca0.2O1.8 nanoparticles showed excellent UV-shielding ability and transparency in the visible light region. The UV-shielding abilities of the plate-like K0.81Li0.27Ti1.73O4/ Ce0.8Ca0.2O1.8 composites increased with the increase of Ce0.8Ca0.2O1.8 content. The UV-shielding ability of the composite with 60 mass % Ce0.8Ca0.2O1.8 which was prepared by 3 times coating was almost identical to that of only Ce0.8Ca0.2O1.8 nanoparticles.

Fig. 7 UV-Vis transmittance spectra of K0.81Ti1.27Li0.73O4/Ce0.8Ca0.2O1.8 composites with various Ce0.8Ca0.2O1.8 contents prepared by sol-gel method In order to evaluate the UV shielding ability and transparencey in the visible light region, the transmittances of thin films at 300 nm and 700 nm were determined for the samples prepared by both coprecipitation method and sol-gel method. As shown in Fig. 8, the transparency in visible light region decreased a little with an increase in Ce0.8Ca0.2O1.8 content, probably due to the agglomeration of Ce0.8Ca0.2O1.8 nanoparticles. The UV-shielding abilities of the composites with 60 mass % and 70 mass % Ce0.8Ca0.2O1.8 by sol-gel method and coprecipitation method, respectively, were almost identical to that of only calcia-doped ceria nanoparticles. It can be seen that the UV-shielding abilities of the composites prepared by sol-gel method were more excellent than those prepared by coprecipitation method, although the transparency in the visible light region was almost the same for

Fig. 8 Transmittances of at (a), (b) 300 nm and (c), (d) 700 nm of the K0.81Ti1.27Li0.73O4/Ce0.8Ca0.2O1.8 composites thin films with various Ce0.8Ca0.2O1.8 contents prepared by (a), (c) sol-gel method and (b), (d) coprecipitation method.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

113

both methods, indicating the better quality of the samples by the sol-gel coating method. It may be attributed to the better distribution of Ce0.8Ca0.2O1.8 nanoparticles after coupling with plate-like K0.81Ti1.27Li0.73O4 by sol-gel method. These results excemplify the importance of coupling the UV-shielding ceramic nanoparticles with plate-like semiconductor microparticles to improve the UV-rays blocking performance.

Conclusions From the present results, following conclusions may be drawn. (1) Nanocomposites of plate-like titanate microparticles/calcia-doped ceria nanoparticles could be prepared by coprecipitation method and sol-gel method. (2)Panoscopic assembling of calcia-doped ceria nanoparticles with plate-like titanate was useful to improve the comfort when applied to the skin as well as decreasing the oxidation catalytic activity. (3) The UV-shielding performance of the nanocomposite prepared by sol-gel method was superior to that by coprecipitation method. These results suggested that the panoscopic assembling of nanoparticles with plate-like semiconductor micro particles is useful to develop new inorganic UV-shielding materials with safe, comfort and excellent UV-shielding ability References [1] G.A. Ozin, Chem. Comm. (2000) 419-432. [2] R.Cai, K.Hashimoto, K. Itoh, Y.Kubota and A. Fujita, Bull. Chem. Soc. Jpn. Vol. 64 (1991), p. 1268. [3] S. Yabe, M. Yamashita, S. Momose, K. Tahira, S, Yoshida, R. Li, S. Yin and T. Sato, Int. J. Inorg. Mat. Vo. 3 (2001), p. 1003. [4] S. Yabe and S. Momose, J. Soc. Cosmet. Chem. Jpn. Vo. 32 (1998), p. 372. [5] S. Yabe and T. Sato, J. Solid State Chem. 171 (2003) 7-11. [6] R. Li, S. Yabe, M. Yamashit, S. Momose, S. Yoshida, S. Yin and T. Sato, Mater. Chem. Phys. Vol. 75 (2002), p. 39. [7] R. Li, , S. Yin, S. Yabe, M. Yamashit, S. Momose, S. Yoshida and T. Sato, Bri. Ceram. Trans. Vol. 101 (2002), p. 9. [8] C. Sato, M.A. El-toni, S. Yin, T. Sato, Jpn. Soc. Powder Powder Metallurgy Vo. 55 (2008), p. 253. [9] T. Miyazawa, K. Fujimoto, M. Kinoshita and R. Usuki. J. Am. Oil Chem. Soc. Vo. 71 (1994), p. 343.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.114

Lightweight Hybrid Foam with Dimensional Stability Ming Y. Chen,1,a Chenggang Chen,1,2,b 1

Air Force Research Laboratory, Materials & Manufacturing Directorate, Wright-Patterson AFB OH 45433-7750 2 University of Dayton Research Institute, 300 College Park, Dayton, OH 45469-0060 a

[email protected], [email protected]

Keywords: hybrid foam, nanoclay, zirconium tungstate, microballon, epoxy

Abstract. Availability of advanced materials has opened up opportunities in meeting several functional requirements through hybridization. Hybrids consisting of ceramics, metals and high performance polymers could benefit many aircraft and space satellite applications. They could meet requirements of low weight, high environmental stability, and high thermal or dimensional stability. In this study, hybrid materials consisting of high performance polymer, porous ceramics (glass microballoons) and other constituents such as Zircornium Tungstate (with negative coefficient of thermal expansion (CTE)) and nanoclay were studied. Specimens were successfully produced with a range of density from 0.4 to 1.1 g/cm3 depending on the degree of fill in the syntactic foams. CTE tailoring was achieved to greatly reduce the residual stress arising from processing and CTE mismatch of dissimilar materials. The evaluations of dimensional stability were examined from thermomechanical analysis. The synergistic effects of resin, ceramic constituents and pores on the hybrid properties will be presented. Introduction Demands for future aircraft and space satellite applications continue to push the envelope for quicker, cheaper, and more reliable access to space, requiring advanced aerospace materials with multifunctional properties. Some of the applications include large, lightweight, high precision aerospace mirrors, reentry long-range strike, reusable access to space, and persistent strike and hypersonic vehicles. Hybrids consisting of ceramics, metals and high performance polymers offer the potential to lower weight, higher environmental stability, and higher thermal or dimensional stability to achieve higher skin and inner warm structure temperatures. Their usage in aerospace applications depends on the availability of novel materials to lower the density, enhance durability in aerospace environment and reduce thermal stress arising from mismatch in the coefficient of thermal expansion (CTE) between dissimilar materials. For instance, polymers are generally easily worn away in aerospace environment such as low earth orbit with very high atomic oxygen flux. The utilization of low density materials would allow higher payload to be launched to outerspace. The mismatch of the coefficients of thermal expansion between the polymer (40 - 70 ppm/K) and carbon fiber (~0 ppm/K) in the traditional composites and that between the polymer and metal (3 – 20 ppm/K) in metal/polymer hybrid is often the root cause for failure. Availability of advanced materials has opened up opportunities to address the aerospace requirements. Previous research on the addition of silicate nanolayers to epoxy resins has shown significantly improvement in the survivability of the materials under space environment [1, 2]. The enhanced durability was attributed to the formation of inorganic ceramic-rich layers,

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

115

which prevented further erosion of the polymeric materials. Filler materials for optimal CTE tailoring of polymers to match the CTE of ceramics or metals should have very low CTE values. This would allow minimal amounts of filler material to be added in order to minimize its impact to the hybrid structural properties. Simultaneously, the residual stress level at the filler/matrix interface also needs to be designed and controlled to avoid cracks resulting from the sharp difference in CTE. Negative thermal expansion fillers offer a wide range of CTE tailorability, including a near zero CTE, giving a dimensionally stable material. Glasses in the titania-silica family have been reported to have a negative CTE at room temperature [3]. Silicon and germanium have a negative CTE at very low temperatures (1500°C. As view port, a hole of 1 cm diameter was drilled into the top cover of the casket and covered with quartz window to avoid heat loss by convection. A sketch of the experimental set–up upon microwave hybrid heating is shown in Fig. 1. Density and porosity of samples were measured by the method based on Archimedes principle. In addition a Micromeritics, Autopore III mercury porosimeter was used to measure the open pore

334

12th INTERNATIONAL CERAMICS CONGRESS PART B

size distributions of sintered samples. LEO 1530 Gemini Field emission gun scanning electron microscope (FE–SEM) was used for microstructural characterization. Pyrometer

Casket made of insulating fiber boards + samples

Home-built 6 KW microwave furnace

Alumina/LHA composites LaCrO3 susceptor Insulating fiber boards Alumina samples samples

Figure 1, Schematic drawing of the experimental set–up upon microwave hybrid heating.

Results Homogenous LHA ceramics. Among the homogenous alumina/LHA composite ceramics, the composites with 20vol.% LHA and 60vol.% LHA were taken as representative composites for low and high LHA content ceramics, respectively. Relative density and porosity results of the conventionally (CS) and hybrid microwave (MWS) sintered 20vol.% LHA and 60vol.% composites are tabulated in Table 1. Table 1, Relative density and porosity of the conventionally (CS) and microwave sintered (MWS) 20 and 60vol.%LHA composite ceramics Relative density [%] Open porosity [%] Closed porosity [%]

Conventional Sintering 20vol.%LHA 60vol.%LHA 97 65 0 8 3 17

Microwave Sintering 20vol.%LHA 60vol.%LHA 97 88 0 2 3 10

The relative density decreased from 97% to 65% for the CS-20vol.%LHA to CS-60vol.%LHA composites and from 97% to 887% for MWS-20vol.%LHA to MWS-60vol.%LHA composites. Retardation of densification with increased LHA content was accompanied with an increase in the open porosity and the average pore size from less than ~20nm to ~200nm (Fig. 2). The relative density and pore size distribution of MWS-20vol%LHA did not reveal any significant differences as compared to the CS sample. At 20vol.%LHA ~3vol.% closed porosity was found in all the samples, regardless of the sintering method. For the MWS-60vol.%LHA relative density of 88% was obtained. Also the MWS-60vol.%LHA had a bi–modal pore size distribution like the CS60vol.%LHA but the contribution of open pores with average diameter of 200µm decreased to 2vol.% in comparison to 8vol.% for the conventionally sintered sample. A maximum of 10% and 17% closed porosity was reached for MWS and CS 60vol.%LHA, respectively, which in the

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

335

Log Differential Intrusion [ml/g]

mercury porosimetry results is the range of pores with average diameter less than 40nm. The reduction in open and closed porosity of the MWS-60vol.%LHA composite comparison with the CS sample is attributed to enhanced densification.

0.20

MWS-Alumina/20 vol % LHA MWS-Alumina/60 vol % LHA CS-Alumina/20 vol % LHA CS-Alumina/60 vol % LHA

0.15

0.10

0.05

0.00

200

20

2 0.2 Diameter [µm]

0.02

0.002

Figure 2, Comparison of pore–size distribution for the conventionally (CS) and the microwave sintered (MWS) 20vol.%LHA and 60vol.% LHA composite ceramics.

(a) CS-20vol.%LHA

(b) CS-60vol.%LHA

(c) MWS-20vol.%LHA (d) MWS-60vol.%LHA Figure 3, FE–SEM micrograph of the cross section of alumina/20vol.%LHA and alumina/60vol.%LHA composite ceramics after conventional and hybrid microwave sintering (SE: Secondary electron image, BSE; Backscattered electron image).

336

12th INTERNATIONAL CERAMICS CONGRESS PART B

As exhibited in Fig. 3 CS-20vol.%LHA and MWS-20vol.%LHA possessed similar microstructures (Fig. 3a,c) composed of very fine equiaxed–grained alumina matrix (in black contrast in BSE images) and well distributed plate–like LHA grains (in white contrast in BSE images). It can also be compared that upon microwave sintering the uni–modal grain size distribution of the alumina matrix in 20vol.%LHA shifted to the smaller grain sizes, meanwhile, the aspect ratio of LHA platelets was increased. Although by further increase in the LHA content to 60vol.% (Fig. 3b,d) porosity was increased and the average grain size of alumina matrix was decreased upon conventional sintering but substantial grain growth and enhanced densification were observed upon microwave sintering. In contrast with CS, MWS resulted also in enhanced anisotropic grain growth of the LHA grains inside the composite ceramics. Functionally graded composite. As it is observed here and have already been discussed in [16], microwave sintering enhances solid–reaction of LHA formation and reaction sintering in the alumina/LHA composite ceramics with LHA content higher than 40vol.%. This was expected to assist in-situ reaction sintering in the functionally graded alumina/LHA composite (FGLHA) in order to promote LHA formation and densification in the layers with higher LHA content and to prohibit alumina grain growth in the layers with lower LHA content. Fig. 4 shows low-magnification SEM micrographs of the multilayer configuration of FGLHA fired conventionally (CS-FGLHA) and by utilization of hybrid microwave heating at 100°C lower sintering temperature than the CS (MW-FGLHA). Crack–free interfaces shown in Fig. 4 (i-v) confirm a controlled green processing for FGLHA development. Due to the volume increase in layers containing higher amount of LHA after solid– state reaction between Al2O3 and LaAlO3 (refer to [17]), a specific residual compression is developed on the surface of layered composite, which increases damage resistance and surface–flaw tolerance of the layered composite. In microstructures, a rather smooth interface, indicating minor interdiffusion of La–ions, was seen only for the 20vol.%–40vol.% composites. In all other composites a tortuous interface have been developed, which could enhance bonding between the individual layers. The ‘smoothening’ of the compositional differences between the layers upon sintering was a major process parameter to transform a green ceramic laminate with abrupt changing functionality into a Functionally Graded Composite. Surprisingly, the interphase regions were developed without a detectable increase in porosity. The MW sintered FGLHA (Fig. 4b) had overall lower grain sizes in comparison with the conventionally sintered sample (Fig. 4a). The results of pore size distribution measurements for both CS and MWS samples shown in Fig. 5 indicate a weakly bi–modal pore size distribution. The majority of pores belonged to a size of 100–150 nm for CS-FGLHA and 300–400 nm for the MWSFLHA. However, it has to be taken into account, that the temperature of microwave sintering was 100°C lower than the conventional sintering, therefore, larger open pores could indicate less sintering. This would be in accordance with less grain growth. The other pore population, in the range of meso–pores with average size lower than ~20 nm was similar in both materials, and possibly resulting from LHA formation. The increase in volume fraction of macro–pores and their average size in the MWS sample could also be attributed to the probable higher thickness of composite layers with LHA content higher than 60vol.%. The significant microstructure difference, particularly the higher LHA content in microwave sintered FGLHA with even 100°C lower sintering temperature could indicate an enhancement in La3+ diffusion, caused by an additional driving force from a vacancy gradient induced by the microwave field. Discussion The enhanced LHA formation, densification, pore removal, and anisotropic grain growth of LHA platelets by microwave heating in comparison with conventional heating was ascribed to the improved mass transport in materials due to an additional driving force induced by microwaves.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

337

Alumina/80 LHA v Alumina/60 LHA

iii v

Alumina/40 LHA ii Alumina/20 LHA

i

Alumina

i

iii

ii

(a) Conventionally sintered FGLHA at 1675°C/10min-1500°C/2h

Alumina/80 LHA v Alumina/60 LHA iii

Alumina/40 LHA

v ii

Alumina/20 LHA Alumina

i

i

ii

iii

(b) Microwave hybrid sintered FGLHA at 1575°C/10min-1400°C/2h Figure 4, FE–SEM micrograph of microstructures obtained for FGLHA after (a) conventional and (b) microwave sintering. (i-v) Magnified interfaces between Al2O3/LHA composite layers (SE: Secondary electron image, BSE; Backscattered electron image).

338

12th INTERNATIONAL CERAMICS CONGRESS PART B

LHA is a defective and nonstoichiometric compound, therefore with increase in the LHA content the defect concentration increased significantly in the composite ceramics. The net result is that in microwave electric fields charged defects can be driven away from, or pulled towards a surface, producing a near–surface depletion or accumulation. This could constitute a nonlinear driving force for diffusion flows of defects and ions within the material near grain boundaries, pores and free surfaces [10–13]. An additional driving force acting selectively on charged defects would promote LHA formation and also sintering of the alumina/LHA composite ceramics.

Log Differential Intrusion [ml/g]

0.14

CS, 1675 °C/10 min-1500 °C/ 2h MWS, 1575 °C/10 min-1400 °C/ 2h 0.008

0.12

0.006

0.10 0.004

0.08 0.002

0.06

0.000

0.01

1E-3

0.04 0.02 0.00 100

10

1

0.1

0.01

1E-3

Diameter [µm] Figure 5, Comparison of pore size distribution for FGLHA composite ceramic conventionally sintered at 1675°C/10 min–1500°C/2 h with the hybrid microwave sintered one at 1575°C/10 min– 1400°C/2h. The reduced Al2O3 grain size, the high solute concentration of La3+ and the increased area of grain boundaries with increase in the LHA content up to 60vol.% could be seen as a microstructure increasingly prone to interaction with an microwave field and to selective heating. In the experimental set up used in this study temperature measurements are taken at the surface of the ceramic body. Therefore direct evidence for a different temperature response of the ceramic by microwave dissipation into heat with increasing LHA-content could not be drawn. Indirect evidence is given by the enhanced reaction and grain growth which occurs with increasing LHA content in the ceramic bodies subjected to a microwave field upon sintering as compared to ceramics sintered conventionally. Heat conductivity of alumina/LHA ceramics decreases with the increase in LHA content [17]. The low heat conductivity and thermal diffusivity of LHA ceramics could also partially contribute in a thermal effect of local overheating, which can not be detected by surface temperature measurement upon microwave sintering and enhance the densification. In functionally graded LHA ceramics, microwave hybrid heating assists development of materials with high mechanical properties and low thermal conductivity which is difficult to achieve in the conventional sintering. Higher temperatures which are needed to complete the in-situ formation of LHA cause the alumina grain growth in the alumina rich layers.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

339

Summary In functionally graded alumina/LHA ceramics microwave hybrid heating enhanced the in-situ formation of LHA platelets, densification, and anisotropic grain growth of LHA in the composite layers with high LHA content in comparison with conventional heating despite 100°C lower sintering temperature. Up to 97% densification was observed in the alumina reach layers with out any significant grain growth in alumina matrix by microwave hybrid heating. Acknowledgement The financial support of Bavarian Science Foundation within the doctoral grant DPA-52/05 is gratefully acknowledged. References [1] I. A. Bondar, N.V. Vinogradova, Russ. Chem. Bull. Vol. 13 (1964), p. 737 [2] B. Saruhan, I. R. Abothu, S. Komarneni, C. K. Loong, J. Am. Ceram. Soc. Vol. 83 (2000), p. 3172 [3] M. Yasuoka, K. Hirao, M. E. Brito, S. Kanzaki, J. Am. Ceram. Soc. Vol. 87 (1995), p. 1853 [4] R. Gadow, M. Lischka, Surf. Coat. Tech. Vol. 151 (2002), p. 392 [5] R. W. Sidwell, H. Zhu, B. A. Kibler, R. J. Kee, D. T. Wickham, Appl. Catal. A-Gen. 2003, 255, 279 [6] FactSageTM Database, http://www.factsage.cn/ [7] R.C. Ropp, B. Carroll, J. Am. Ceram. Soc. Vol. 63, (1980), p. 416 [8] J. G. Park, A. N. Cormack, J. Eur. Ceram. Soc. Vol. 19 (1999), p. 2249 [9] J. G. Park, A. N. Cormack, J. Solid State Chem. Vol. 130 (1997), p. 199 [10] J. Wang, J. Binner, B. Vaidhyanathan, N. Joomun, J. Kilner, G. Dimitrakis, T.E. Cross, J. Am. Ceram. Soc. Vol. 89 (2006), p. 1977. [11] K. I. Rybakov, V. E. Semenov, Phys. Rev. B Vol. 49 (1994), p. 64 [12] M. Willert-Porada, in: Ceramic Transaction, Microwaves: Theory and Application in Materials Processing III, edited by D.E. Clark, D.C. Folz, S.J. Oda and R. Silberglitt, volume 59, p. 193, Am. Ceram. Soc., Westerville, OH (1995). [13] J. H. Booske, R .F. Cooper, S. A. Freeman, Mater. Res. Innov. Vol. 1 (1997), p. 77 [14] Z. Negahdari, M. Willert-Porada, J. Eur. Ceram. Soc. Vol. 30 (2010). p. 1381 [15] T. Gerdes, Ph. D. thesis, University of Dortmund, ISBN 3-18-343205-6, (VDI-Verlag GmbH, Germany 1996). [16] Z. Negahdari, M. Willert-Porada, Adv. Eng. Mater. Vol. 12 (2010), p. 216 [17] Z. Negahdari, Ph. D. thesis, University of Bayreuth, ISBN: 978-3-8322-8801-3 (Shaker Verlag, Germany 2009).

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.340

Fabrication of functionally graded ZTA ceramics using a novel combination of freeze casting and electrophoretic deposition Annemarie Preiss1, a, Bo Su1, b, Simon Collins2, c and Peter Ellison2, d 1

Biomaterials Engineering Group (BioMEG), Department of Oral & Dental Science, University of Bristol, Lower Maudlin Street, Bristol BS1 2LY, UK 2 Corin Ltd, Cirencester, Gloucestershire, GL7 1YJ, UK a

[email protected], b [email protected], c [email protected], d [email protected]

Keywords: Electrophoretic deposition (EPD), freeze casting, functionally graded, zirconia toughened alumina (ZTA)

Abstract: Functionally graded zirconia toughened alumina (ZTA) ceramics have been fabricated from aqueous suspension with an open porous and aligned lamellae structure on one side and a dense layer on the other side. A novel combination of two processes has been merged to achieve such graded structures, i.e. unidirectional freeze casting and electrophoretic deposition (EPD). A custom-designed apparatus has been built in which a controlled double side cooling has been realized in conjunction with the possibility to introduce an electric field over the ceramic slurry prior to the freezing process. A square wave pulsed DC voltage has been used in the EPD process in order to avoid electrolysis of water. Suitable duty cycle of applied pulse voltage could gain bubble-free deposition. The thickness of the dense layer is controlled by tuning voltage, duty of cycle, pulse width and deposition time. It was shown that thicknesses up to 500µm could be achieved. The microstructure of the porous part is controlled by adjusting the temperature during the freezing process. Using temperatures between -1 and -25°C the channel width changed from 220 to 40µm, respectively. Introduction: Functionally graded materials (FGMs) are of high interest nowadays and have been focussed on the optimisation and tailoring process in the last decade due to their outstanding properties and functions compared to conventional homogeneous materials [1, 2]. In this work we will focus on the production and optimisation of a ceramic preform in which a second phase can be introduced to produce functionally graded composite materials. The unique feature of this preform is a dense-porous ceramic which is produced in a continuous process that results in a gradually changing microstructure from a dense to a porous ceramic without the presence of an interface. The ceramic preform is produced with a novel combination of electrophoretic deposition (EPD) and unidirectional freeze casting (Fig. 1). EPD has been performed in this study to produce a dense ceramic layer. EPD from aqueous suspensions generally have the problem of electrolysis of water (over 1.23V at 25°C [3]) which occurs as gas evolution at the electrodes. Oxygen and hydrogen gas evolve at anode and cathode, respectively which gets entrapped within the deposit and results in poor quality of the resultant ceramic deposit. Furthermore it disturbs the electrophoresis of the ceramic particles [4, 5]. Therefore a pulsed DC has been used in this study to avoid the formation of gas at the electrodes [3, 6]. By applying an electric field over the slurry the ceramic particles are forced to one electrode prior to unidirectional freezing. Consequently the understanding of the colloidal processing of the ceramic slurries is essential to ensure the formation of good quality green bodies. Characterisations of slurries such as viscosity and zeta potential have therefore been carried out in this work.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

341

After the ceramic particles build a dense deposit via EPD, the suspension is unidirectional frozen by applying desired temperatures from both sides. By implementing a temperature gradient during freeze casting the microstructure and the composition gradually changed over the height.

Fig. 1: Flow chart of fabrication process for dense-porous ceramics

Several studies discussed the process of unidirectional freezing due to its simplicity and flexibility in changing microstructure. They showed that by varying the solid concentration and freezing velocity the ice crystal size and their morphology were adjustable [7-11]. Furthermore, it has been identified that it was possible to render tailored ceramic microstructures in terms of lamellae spacing from a double side cooling [12]. The objective of this study is to show the possibility to create dense-porous functionally graded ceramics using a novel combination of two known processes, i.e. EPD and freeze casting. The parameters during freeze casting, controlled by the double side cooling apparatus will be investigated as well as the effects of the electric field intensity on the microstructure of ceramics. We will demonstrate that with this novel combination, graded structures with tailored lamellae spacing are achievable. Experimental Procedure: Alumina powder (CT3000SG, Almatis GmbH, Germany) with an average particle size of 0.5µm and zirconia powder (TY-3YS-E, yttria partially-stabilized, Tosoh Corp., Tokyo, Japan) with an average particle size of 0.3µm were used in this study. Aqueous suspensions containing 17.5vol% ZTA with 10vol% zirconia relative to alumina were dispersed in distilled water using Dolapix CE64 (anionic dispersant, Zschimmer & Schwarz, Germany) at a ratio of 0.6wt% on the dry powder weight basis. The mixture was ball milled in polyethylene bottles using zirconia balls for 12h to obtain well dispersed slurry. After removal of the balls, polyvinyl alcohol (PVA, MW: 30,000-70,000, Sigma-Aldrich, USA) was added, at a ratio of 0.825wt% on the dry powder, which acts as an organic binder. The slurry was sonicated for 10min to break down agglomerates. The slurry was poured into a PTFE mould (Ø 60mm) connected to two copper rods on each side as showed in Fig. 2. The top copper rod was cooled by a cold finger and the bottom one was placed in an ethanol bath which was surrounded by liquid nitrogen. Band heaters and thermocouples were connected to the rods, the latter ones were inserted into the rod

342

12th INTERNATIONAL CERAMICS CONGRESS PART B

close to the slurry. Both, band heaters and thermocouples, were connected to a PID console (TEC9100, Tempco, USA) to control the temperature during freeze casting. The distance between the two surfaces could be adjusted as desired. In this study a distance of 10mm was used throughout all experiments.

Fig. 2: Schematic of the custom-designed double-side cooled apparatus

The fabrication of functionally graded ZTA ceramics was performed in two steps. Before freezing the suspension, an electric field was applied over the slurry to force the ceramic particles towards the bottom electrode. Pulsed EPD was conducted at constant voltage of up to 10V by applying a series of direct current pulse of equal amplitude and duration by periods of zero voltage using a function/arbitrary waveform generator (33220A, Agilent Technologies, USA). A simple square wave pulse is used in this project as shown in Fig. 3. To achieve a dense ceramic layer without bubble incorporation with a reasonable thickness (up to 500µm), a compromise between duty cycle, pulse width, voltage and deposition time needed to be made. The optimal settings for this application of pulsed EPD were 50% duty cycle, 10V, 0.005sec pulse width and 20min deposition time.

Fig. 3: Schematic of constant voltage pulse of 50% duty cycle

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

343

In order to keep a fixed distance between the two electrodes of 10mm and an identical height of the ceramic specimen, an amount of 29ml suspension has been used throughout this study. During EPD the temperature was held at +1°C on top and bottom. After the ceramic had deposited onto the electrode surface the suspension was frozen at temperatures between -1 and -25°C on the bottom with a cooling rate of 1.5°C/min while the top was kept constant at -1°C (Fig. 4).

Fig. 4: Flow chart of process steps and settings: ∆Τ1: temperature of bottom rod (-1 to -25°C), ∆T2: temperature of top rod (-1°C), ∆t1: time of EPD (5 to 20mins), ∆t2: time to cool down (cooling rate

1.5°C/min) The frozen samples were then freeze dried for 12h to sublimate the ice and subsequently sintered at 1600°C for 2h. To make the samples processable the ceramic specimen were then embedded with resin (SpeciFix: Epoxy Resin & 20Curing Agent, Struers A/S, Denmark) under reduced vacuum (Cast n’Vac, Buehler GmbH, Germany). To obtain a better contrast in the optical microscope a small amount of blue ink was added to the resin. After curing the samples were cut (Acutom-5, Struers A/S, Denmark) parallel and perpendicular to the ice front for microscopical examinations. The macro- and microstructure of the sintered samples were examined using digital microscopy (Hirox KH-7700, Hirox, USA) and scanning electron microscope SEM (Jeol JSM 6330F, Tokyo, Japan), respectively. The stability of the slurry was analysed in zeta potential measurement with a Nano Z (Zetasizer, Malvern Instruments Ldt., UK) using diluted suspensions at different pH values. The rheological behaviour (Bohlin Gemini HR Nano Rheometer, Malvern) was also determined. The viscosity was measured on 40vol% ZTA suspensions using a concentric cylindrical rotational viscometer. The suspension was prepared by mixing the adequate amount of alumina and zirconia powder with the corresponding amount of dispersant and water. The suspension was ball milled for 12h. After removal of the ceramic balls the suspension was ultrasonicated for 10min before measuring. The volume of open porosity in sintered samples was measured using the Archimedes. Results and Discussion Prior to the fabrication of the graded dense-porous ceramic specimen, the suspension was characterised. In order to decide which dispersant content is the optimum, rheological measurements were carried out on high concentrated ZTA suspensions. The viscosity was measured on slurries with different amounts of dispersant, ranging from 0 to 2wt% on the dry powder weight basis, as a function of a range of shear rates, from 0.1 to 1500s-1. As seen in Fig. 5 the viscosity changes with varying amount of dispersant in slurry. The optimum content of dispersant can be determined from the dependence of the apparent viscosity on the dispersant content and corresponds to a minimum of apparent. The lowest viscosity was found to be 0.6 wt% which was chosen as the optimal amount of dispersant and used throughout this study.

344

12th INTERNATIONAL CERAMICS CONGRESS PART B

0.40 30s-1 102s-1 450s-1 1213s-1

0.35

Viscosity [Pas]

0.30 0.25 0.20 0.15 0.10 0.05 0.00 0.5

0.6

0.75

0.8

1

Amount of Dolapix CE64 [w t%]

Fig. 5: Viscosity behavior of 40vol% ZTA suspension at different shear rates as a function of the dispersant content

For a successful EPD it is essential to choose a suspension with a high surface charge of the suspended particles. That means the system needs to be stable which depends on the individual particle interaction in slurry. Variables such as suspension stability i.e. sedimentation and agglomeration behaviour, migration velocity of the particles during EPD and density of the green ceramics are affected by the particle charge. Therefore the stability of the ZTA slurry has been investigated by measuring the zeta potential under consideration of additives (binder and dispersant). Binder as one additive in this suspension system is an important component in a ceramic suspension as it gives strength to the green body especially when highly porous ceramics are fabricated. PVA as a binder was selected due to the increased stability of the suspension at its originally pH. The measurement of the zeta potential showed that the particle surfaces of the pure ZTA suspension, without the addition of dispersant and binder, are positively charged in a wide range of pH. The more dispersant is added the more negative the particle charge. The IEP decreases from pH 8.7 to 4.1 with 0 to 1.2wt% dispersant, respectively. The IEP of the ZTA suspension with the determined optimal dispersant content of 0.6wt% is at pH 8 and decreases when 0.825wt% PVA is added to pH 4.5. Fig. 6 shows the dependence of the zeta potential on the pH of the ZTA suspension used in this project. The particle surfaces are negatively charged at its originally pH of 9.2. The graph shows that the IEP of the ZTA slurry is at 4.5. The slurry is negatively charged above pH 5.8 and gets stable below approximately pH 3.4 and above pH 7.5 which is important for EPD to achieve homogeneous and smooth deposition. Therefore, anodic EPD was performed at the inherent pH of the suspension.

Fig. 6: Zeta potential of ZTA suspension as a function of pH

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

345

Due to the fact that the thermocouples could not be placed directly on the surfaces of the cold plates the temperature difference (apparent temperature/measured temperature) was measured and adjusted with the PID console. Anodic pulse EPD was performed at constant voltage prior to freeze casting using ZTA suspension at its inherent pH. Preliminary EPD tests were performed in a glass beaker with two parallel positioned copper electrodes fixed at a distance of 10mm to determine the settings necessary to achieve smooth deposits. After verifying the adjustments, EPD was carried out in the PTFE mould, the two copper surfaces on top and bottom providing the electrodes. The preliminary experiments showed that with increasing deposition time and voltage and decreasing pulse width the deposit yield of the dense layer increases. Using pulse duty cycle the voltage interruption (Toff ), when no voltage is flowing through the suspension, interrupts the electrolysis process which affects the quality of the deposit. Therefore the pulse width and the voltage influence the evolution of gas. In Fig. 7 a bottom view and a cross-section of sintered dense areas are shown of a sample for which EPD was performed with 5V applied potential, a pulse width of 0.005sec and a deposition time of 10 min. It can be seen that the layer is dense and has a height of approximately 500µm. It has to be noticed that the actual running time of EPD process is just half of the deposition time due to 50% duty cycle.

Fig. 7: SEM micrographs showing (A) the dense bottom of a ZTA sample (view from the bottom side) and (B) the cross-section of the dense layer (perpendicular to ice front)

The thickness of the dense layer on the bottom of the samples was analysed using the digital microscope. The thickness ranges between 100µm and 500µm for 0.0005sec/5V and 0.005sec/10V, respectively. After accomplishing EPD, the ceramic suspension was frozen in the same mould. Fig. 9 shows images taken with the digital microscope of the top surface of the sintered ceramic specimen. It shows the influence of the freezing temperature on the thickness of the ice crystals. With decreasing temperature the ceramic lamellae width decreases. Temperatures from -1°C to -25°C were used to freeze the ceramic suspension unidirectional from the bottom to the top at a cooling rate of 1.5°C/min. With increasing distance to the cool bottom plate the lamellae spacing of the ceramic specimen increases (Fig. 8).

Fig. 8: Typical microstructure of resin infiltrated sample perpendicular to the ice front

346

12th INTERNATIONAL CERAMICS CONGRESS PART B

Using the double-side cooled apparatus, the microstructure is easily repeatable using the same freezing condition due to the fact that the ceramic suspension is completely insulated from the surrounded temperatures. In Fig. 10 the relation between freezing temperature and channel width is graphically displayed. Depending on the application of the ceramic the settings can be applied to achieve the desired microstructure.

Fig. 9: Top view of ceramic specimen at different freezing temperatures on bottom, (A)-1°C, (B) -12°C, (C) 20°C, (D) -25°C

Fig. 10: Influence of freezing temperature over the width of the channels at a height of 10mm

Using Archimedes method the open porosity of the sintered ceramic specimen was determined. Depending from the freezing temperature the open porosity is between 55.5% and 76.5% at freezing temperatures of -25°C and -1°C, respectively. Conclusion The method presented here shows the ability to produce a graded dense-porous ceramic structure in which no interface occurs. It has been showed that using the custom-designed apparatus where a suspension is completely insulated and by merging two processes, electrophoretic deposition and freeze casting, the fabrication of ceramic specimen with tailored microstructure could be reproduced. It has

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

347

been demonstrated that by applying an electric field over the ceramic suspension the thickness of the dense layer can be tuned by adjusting deposition time and voltage. Dense deposit of aqueous slurries up to 500µm has been achieved by using pulse waves of 50% duty cycle. Additionally, the lamellae spacing of the ceramic specimen can be tailored by tuning the freezing temperature. Using temperatures from -1°C to -25°C channel widths from 40µm to 220µm could be achieved. Acknowledgements The work was performed in the BioMEG group and funded by GWR and Corin Ltd. AP would like to acknowledge the financial assistance of the ALUMNI Foundation. References 1. Corbin, S.F., et al., Functionally graded metal/ceramic composites by tape casting, lamination and infiltration. Materials Science and Engineering A, 1999. 262(1-2): p. 192-203. 2. Binner, J., H. Chang, and R. Higginson, Processing of ceramic-metal interpenetrating composites. Journal of the European Ceramic Society, 2009. 29(5): p. 837-842. 3. Besra, L., et al., Application of constant current pulse to suppress bubble incorporation and control deposit morphology during aqueous electrophoretic deposition (EPD). Journal of the European Ceramic Society, 2009. 29(10): p. 1837-1845. 4. Van der Biest, O.O. and L.J. Vandeperre, Electrophoretic deposition of materials. Annual Review of Materials Science, 1999. 29: p. 327-352. 5. Tabellion, J. and R. Clasen, Electrophoretic deposition from aqueous suspensions for nearshape manufacturing of advanced ceramics and glasses—applications. Journal of Materials Science, 2004. 39(3): p. 803-811. 6. Besra, L. and M. Liu, A review on fundamentals and applications of electrophoretic deposition (EPD). Progress in Materials Science, 2007. 52(1): p. 1-61. 7. Deville, S., E. Saiz, and A.P. Tomsia, Ice-templated porous alumina structures. Acta Materialia, 2007. 55(6): p. 1965-1974. 8. Araki, K., Porous ceramic bodies with interconnected pore channels by a novel freeze casting technique. Journal of the American Ceramic Society, 2005. 88(5): p. 1108-1114. 9. Moritz, T. and H.J. Richter. Ceramic bodies with complex geometries and ceramic shells by freeze casting using ice as mold material. in 6th Students Meeting 2005. 2005. Novi Sad, SERBIA MONTENEG: Blackwell Publishing. 10. Moritz, T. and H.J. Richter, Ice-mould freeze casting of porous ceramic components. Journal of the European Ceramic Society, 2007. 27(16): p. 4595-4601. 11. Deville, S., Freeze-casting of porous ceramics: A review of current achievements and issues. Advanced Engineering Materials, 2008. 10(3): p. 155-169. 12. Waschkies, T., R. Oberacker, and M.J. Hoffmann, Control of Lamellae Spacing During Freeze Casting of Ceramics Using Double-Side Cooling as a Novel Processing Route. Journal of the American Ceramic Society, 2009. 92(1): p. S79-S84.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.348

Effects of Strain-graded Plastic Deformation on Mechanical Properties of Metals Kiyotaka Matsuura a and Munekazu Ohno b Division of Materials Science and Engineering, Faculty of Engineering, Hokkaido University, Sapporo, Hokkaido, 060-8628, Japan

a

[email protected], b [email protected]

Keywords: rolling, flow forming, plastic deformation, mechanical property, Al-based alloy

Abstract. In a flow forming process of automobile wheels of Al-based alloys, plastic deformation of the rim part is performed by rollers from the periphery side, while the inner periphery is fixed on a steel mandrel and is slightly deformed. Therefore, the rim part has a strain-graded microstructure in the thickness direction. In this study, the effects of the strain-graded plastic deformation on mechanical properties of an Al-Si cast alloy have been investigated. The strain-graded plastic deformation in this study was done by hot rolling of an Al alloy plate together with a steel plate. The two plates were joined at one end in the longitudinal direction and were rolled from the joined edge at 330 oC using a roller with a roll diameter of 200 mm and a rotation speed of 66 per minute. The chemical composition of the alloy was Al-7mass%Si-0.3mass%Mg-0.3mass%Fe. The rolled Al alloy plate had a strain-graded microstructure in the thickness direction; the strain was the highest at the roller side surface and the lowest at the steel plate side surface. The rolling also brought about a Si particle size graded microstructure. The eutectic Si rods were broken by the rolling deformation and the Si particle size was the smallest at the roller side surface and the largest at the steel plate side surface. On the other hand, a normal rolling deformation of the Al alloy plate without the steel plate was also performed for comparison. The rolled sample having the strain-graded and Si particle size graded microstructure exhibited much more excellent bending strength and ductility compared with the normally rolled sample. Introduction The use of Al-based alloys in automobile industries is increasing due to request of reduction in weight of the vehicles [1]-[3]. An Al-based alloy of AC4C is widely used for wheels. An excellent combination of mechanical strength and lightweight properties is strongly required in wheel production. Wheels are usually made of steel or Al-based alloys. The former is produced by plastic deformation while the latter by casting. Although the latter is more expensive, the production of them is increasing due to their excellent formability which enables their elaborate design. Recently, for better combination of the mechanical strength and lightweight properties, flow forming technique is used for Al alloy wheels production. In the flow forming, a preform of wheel is

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

349

fabricated by casting and the rim part of the cast is plastically deformed to give the final shape and to improve the mechanical properties by refining the microstructure. Figure 1 schematically illustrates the flow forming process. The preform is fixed on a steel mandrel (Fig. 1(a)), and is given the shape by rolling on the mandrel surface (Fig. 1(b)), and finally the pattern of the mandrel is transcribed to the rim (Fig. 1(c)). The plastic deformation of the rim part by rolling on the mandrel leads to difference in plastic strain in the thickness direction of the rim because the deformation of the work piece near the mandrel side is limited due to friction between the mandrel surface and the work piece while that near the roll side is significantly deformed by rolling. Therefore, the plastic strain gradually increases from the mandrel side to the roll side resulting in a strain-graded microstructure, which may affect the mechanical properties of the rip part. Generally, plastic deformation accompanied with heat treatment improves the mechanical properties. However, it is not well known how the difference in plastic deformation in the thick direction affects the mechanical properties of a plate-like metal material. In this study, therefore, we investigate the effects of strain-graded plastic deformation on the mechanical properties.

Fig. 1 Schematic illustration of the flow forming of a wheel.

Fig. 2 Casting of the alloy and strain-graded rolling of the cast.

350

12th INTERNATIONAL CERAMICS CONGRESS PART B

Experimental Procedure An AC4C alloy (Al-7mass%Si-0.3mass%Fe-0.3 mass%Mg) was melted in air and cast into a rectangular block (Fig. 2(a)). A 4-mm thick steel plate having a hole of a 10-mm diameter was placed in the cavity of the mold to be embedded in the Al-alloy after casting (Fig. 2(b)). The upper part of the cast having shrinkage cavities was removed (Fig. 2(c)) and the cast was rolled at 350 oC (Fig. 2(d)). The steel plate had a role of the mandrel to give a strain-graded deformation in the thickness direction of the work piece. Alloy cast without the steel plate was also rolled to produce the standard samples having homogeneous strain in the thickness direction. Rolling reduction was varied from 20 to 80 % for the homogeneously rolled standard sample, while it was fixed at 70 % for the inhomogeneously rolled sample with the steel plate. Rectangular samples of 3mm-4mm20mm were cut from the rolled samples, and 3-point bending tests were performed at room temperature after a

Fig. 3 Microstructures of rolled samples of (a) with and (b) without steel platel. Rolling reduction is 70 %.

T6 heat treatment. Results and Discussion Figure 3(a) shows a microstructure of the rolled sample. The rolling reduction was 70 %. Figure 3(b) shows another sample normally rolled without the steel plate at the same rolling reduction. The bright part and the dark part in the figures correspond to primarily solidified dendrite of -Al solid solution and the eutectic-solidified interdendritic area, respectively. The deformation of the primary dendrite part is homogeneous in Fig. 3(b), while in Fig. 3(a) the deformation is localized in the area near the roll side surface. The local plastic strain in the rolled alloy can be estimated from the distribution of eutectic Si particles as shown in Fig. 4. Because the Si particle colony indicates the eutectic-solidified interdendritic area, the distance between the Si colonies corresponds to the thickness of the primary -Al dendrite. The distance between the colonies was approximately 30 m for a 70% homogeneously rolled standard sample and was

Fig. 4 Microstructures (a) near the roll side surface and (b) near the center in the thickness direction of the rolled sample with mandrel .

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

351

approximately 24 m for an 80% homogeneously rolled standard sample. On the other hand, in the sample inhomogeneously rolled with the steel plate, the distance between Si particle colonies varied from approximately 20 m in the roll side surface area to 50m in the steel plate side area. Figure 5 shows load-displacement curves of the bending test. The distance between the supporting points was 15 mm. The displacement of the lording point is given in the horizontal scale of the figure. Four curves for the homogeneously rolled standard samples and one curve for the flow-formed sample are shown in the figure. In the curves for standard samples, the rolling reduction seems to have a little effect on the fracture load, while it significantly increases the fracture displacement. The increase in fracture displacement is considered to have resulted from refinement of Si particles and their homogeneous distribution due to rolling, as seen in Fig. 4(b). In case of the flow-formed sample in Fig. 5, the fracture displacement is much larger than that of 80%-rolled standard sample, although the rolling reduction is 70 %. This attributes to the locally high plastic strain near the roll side surface of the sample. It was suggested that the flow forming improves the ductility and toughness of the Al-based cast alloys.

Fig. 5 load-displacement curves of the bending test. Summary In plastic deformation of Al-based casts, increase in plastic strain increases the ductility and toughness. Especially, the flow forming gives a large strain near the roll side surface and it leads to excellent ductility and toughness compared with standard rolling. References [1] Y. Ishiguro, K. Mori, H. Ohsako, T. Nonaka, D. Sugiyama and O. Ebihara: J. Jpn. Soc. Tech. Plasticity Vol. 49 (2008), p. 60 [2] F. Fukuji, N. Hayashi, T. Ogawa, S. Yokoyama and I. Hori: J. Light Met. 55 (2005), p. 147 [3] M. Suzuki, S. Hayashi: J. Light Met. 55 (2005), p. 275

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.352

Defect Crystal Structure of Low Temperature Modifications of Li2MO3 (M=Ti, Sn) and Related Hydroxides Nadezda V. Tarakina1a, Tatiana A. Denisova1b, Yana V. Baklanova1c, Lidia G. Maksimova1, Vladimir G. Zubkov1, Reinhard B. Neder2 1

Institute of Solid State Chemistry UB RAS, Ekaterinburg 620990, Russian Federation

2

Kristallographie und Strukturphysik, Universität Erlangen, Erlangen D-91058, Germany a

[email protected], [email protected], [email protected]

Keywords: Defect crystal structure, stacking faults, X-ray powder diffraction, Li2SnO3, Li2TiO3, TiO(OH)2, layered double hydroxides.

Abstract. Crystal structures of Li2MO3 (M=Sn, Ti) and TiO(OH)2 have been studied in detail and refined using X-ray powder diffraction data. All compounds posses a high concentration of defects in the structure. The crystal structures of the Li2MO3 salts obtained at 700°C reveal stacking faults of LiM2 metal layers, which leads to the appearance of short-range order in three possible space groups: C2/c, C2/m, P3112. The possibility to stabilize this imperfect state increases the mobility of the Li+ ions in the Li2TiO3 structure and allows the complete exchange of lithium by hydrogen in acid water solutions with formation of TiO(OH)2. The crystal structure of TiO(OH)2 belongs to the layered double hydroxide structure type with the 3R1 sequence of oxygen layers and can be described as a stacking of charge-neutral metal oxyhydroxide slabs [(OH)2OTi2O(OH)2]. Introduction Lithium salts Li2MO3 (M = Sn, Ti, Zr, Mo, Pd, etc.) are candidate materials for Li+ ion conductors and solid breeders in a fusion reactor [1-3]. Li2SnO3 can also be used as a precursor for the synthesis of XSnO3 (X = Zn, Co, Ni) [4]. It has also been shown that, depending on the sintering conditions, the reactivity of Li2MO3 is different [5]. Previously, we have established that during the interaction processes between Li2TiO3 with acid water solutions Li+ ions can be fully exchanged for hydrogen ions [5, 6]. The exchange is realized through the formation of Li2−xHxTiO3 (03.23

>3.24

3.18

3.10

3.10

3.20

3.14

3.18

290

310

310

310

395

410

430

363

377

1500

1520

1450 (300g)

1650 (300g)

2500

2800 (100g Konopp)

2600 (300g)

2812 (300g)

3132 (300g)

6.7

6.0

7.0

4.5

3.0

4.6

4.3

5.6

4.9

Density, g/ccm Young’s modulus, GPa Vickers hardness, HV Fracture toughness, 1/2 MPa·m

Conclusions The influence of different (nano-, submicro- and micro-structured) initial Si3N4 and SiC powders on mechanical properties of Si3N4–SiC composites sintered by the HPHT method was investigated. Nanocomposites are characterized by the lowest physical-mechanical properties of the three granulometric types of the investigated materials. The poor quality of these materials can be attributed mainly to the high gas content in the samples due to the absorption process on nanoparticles of initial powders. Composites obtained from submicron powders display the highest density, Young’s modulus, hardness and fracture toughness. Submicro-Si3N4–SiC composites have a better combination of mechanical properties than comparable commercial materials.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

401

The investigated properties predispose submicro-structured Si3N4–70 vol.% SiC composites to application in cutting tools. Wear tests and cutting tests (intended in the future) will show the range of applications of this material in machining. Acknowledgements The present study was perform within the framework of bilateral cooperation between The Institute of Advanced Manufacturing Technology in Cracow and SSPA "Scientific-Practical Materials Research Center of NAS of Belarus" in Minsk. Project fund by The Polish Ministry of Science and Higher Education (Project number: DPN/N111/BIALORUS/2009). References [1] [2] [3] [4]

[5]

[6] [7]

[8] [9] [10] [11] [12]

[13] [14]

[15]

H. Peng, Spark Plasma Sintering of Si3N4-Based Ceramics, Doctoral Dissertation, Department of Inorganic Chemistry, Stockholm University 2004. H.O. Pierson, Handbook of refractory carbides and nitrides, Noyes Publications, Westwood, New Jersey 1996. D. W. Richerson, Advanced ceramic materials, in J. K. Wessel (edt), Handbook of advanced materials, John Wiley & Sons, Inc., Hoboken, New Jersey 2004. F. Eblagon, B. Ehrle, T. Graule, J. Kuebler, Development of silicon nitride/silicon carbide composites for wood-cutting tools, Journal of the European Ceramic Society 27 (2007) 419428. S.M. Lee, T.W. Kim, H.J. Lim, C. Kim, Y.W. Kim, K.S. Lee, Mechanical Properties and Contact Damages of Nanostructured Silicon Carbide Ceramics, Journal of the Ceramic Society of Japan 15/5 (2007) 304-309. S. Suyama, T. Kameda, Y. Itoh, Development of high-strengths reaction-sintered silicon carbide, Diamond and Related Materials, 12 (2003) 1201-1204. K. Niihara, T. Kusunose, S. Kohsaka, T. Sekino, Y.-H. Choa, Multi-Functional Ceramic Composites trough Nanocomposite Technology, Key Engineering Materials 161-163 (1999) 527-534. A. W. Weimer, R. K. Bordia, Processing and properties of nanophase SiC/Si3N4 composites, Composites Part B: Engineering, 30/7 (1999) 647-655. A. Bellosi, Si3N4/SiC Ceramic Nanocomposites, Materials Science Forum, 195 (1995) 79-86. M. Sternitzke, Review: Structural Ceramic Nanocomposites, Journal of the European Ceramic Society 17 (1997) 1061-1082. B. Derby, Ceramic nanocomposites: mechanical properties, Current Opinion in Solid State and Materials Science, 3/5 (1998) 490-495. H. Mabuchi, H. Tsuda, T. Ohtsuka, T. Matsui, K. Morii, In-situ synthesis of Si3N4 - SiC composites by reactive hot-pressing, High Temperatures - High Pressures 31 (5) (1999) 499506. J. Wan et al., Silicon nitride/silicon carbide nano-nano composites, United States Application 20040179969. P. Klimczyk, D. Vallauri, P. Figiel, L. Jaworska, I. Amato, Development of TiC – TiB2 nanoceramic composites obtained from metastable powders by HPHT method, The 24th International Manufacturing Conference (IMC 24) August 2007, Waterford Ireland, Proceedings of the IMC24, 2 (2007) 813-820. Company data sheets from: H.C. Starck Ceramics GmbH& Co. KG, Ceradyne Inc. and SaintGobain Advanced Ceramics.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.402

Effect of the Hydrothermal Heat Treatment Conditions of Titanium Substrates on the Bio-Mimetically Grown “Bone-Like” Apatite Coatings Dilek TEKERa, Can Poyraz SAĞb, Mehmet DİNÇERc, Sedat ALKOYd and Koray ÖZTÜRKe Gebze Institute of Technology, Materials Science and Engineering Department, Cayirova Campus Gebze, Kocaeli, 41400, Turkey a

b

[email protected], [email protected], [email protected], [email protected], e

[email protected]

Keywords: Hydrothermal modification, Simulated Body Fluid, Biomimetic, Coatings

Abstract: Hydrothermal surface modifications of the titanium specimens were performed (i) at 200 ºC under 1.5 MPa pressure, and (ii) at 230 ºC under 2.5 MPa pressure. To see the effect of an ion implantation, two different aqueous hydrothermal environments were selected: (i) de-ionized water and (ii) calcium containing de-ionized water. Hydrothermally treated titanium surfaces were analyzed by X-ray photoelectron spectroscopy (XPS) and found to become rich with Ca when Cacontaining hydrothermal environment was used. The surface-modified titanium specimens were then kept immersed in 1.5X simulated body fluid (SBF) for 1, 2, 3 and 4 weeks. The biomimetically formed coatings were characterized using scanning electron microscopy (SEM) and Xray diffractometer (XRD). Crack formations and, consequently, severe peelings were observed after drying for all the coatings on the substrates that were treated hydrothermally using only de-ionized water. The Ca implanted titanium surfaces, on the contrary, were able to develop crack-free and quite cohesive coatings. Up to two weeks of immersion and after drying, no-cracks were observed in the coatings when the substrates were treated at higher temperature and under higher pressure (230 ºC and 2.5 MPa for the present investigation). Introduction

Hydrothermal treatment is one of the methods to modify the outer-most layer of the metal substrates for bio-mimetic coating [1,2]. Titanium metal surfaces have been treated in various aqueous environments which contain ions such as calcium, sodium, potassium and phosphate to contribute to the formation of apatite coating [3-7]. The starting substrate surfaces can, therefore, be charged with the desired ions from the solution. The surface of the passive oxide layer on the metal is changed into a very thin and gel-like or amorphous layer [8,9]. This type of layers are reported be effective in facilitating the nucleation of the “bone like” apatite mineral in SBF through the tentative ion-exchange mechanism [10]. The apatite formation is initiated by the exchange reaction between the ions on the substrate (e.g. Ca2+) and the H3O+ ions in the adjacent SBF. Through this reaction, the surface becomes rich in OH- groups. Then, the positively charged Ca2+ ions in the SBF migrate back to the surface to form an amorphous calcium and titanium containing layer (probably calcium titanate). This layer, then, reacts with the negatively charged phosphate ions in the SBF and turns into an amorphous calcium phosphate. Since amorphous calcium phosphate is metastable in the SBF, it transforms into an apatite and forms bio-mimetically grown “bone-like” layer on the substrate [11-13].

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

403

Experimental Procedures A commercially pure titanium plate (99.6 %; Goodfellow) with dimensions of 1mm x 300mm x 300mm was first cut into discs with the diameter of 15mm and the thickness of 1mm. Prior to hydrothermal heat treatment, the surfaces of the titanium discs were abraded with #800 and #1200 SiC paper and washed with acetone, ethanol and distilled water in an ultrasonic cleaner for 15 minutes each. The discs were then dried in a desiccator at 50 ºC. The aqueous solution used for the hydrothermal modification of the titanium surfaces contains 0.02 mol/lt CaO. Since CaO dissolves in de-ionized water easily, it transform into Ca(OH)2 on contact with water. The titanium discs were immersed in 127 ml of this aqueous solution that was placed in a teflon container. The teflon container was then fixed in the heat and/or pressure-controlled stainless-steel hydrothermal chamber (Parr 4566, Parr Instrument company, Moline, Illinois, USA). Pre-treatments were performed for 1 hour under the hydrothermal conditions at 200ºC and 230 ºC and under 1.5 MPa and 2.5 MPa pressures, respectively. After drying, the hydrothermally modified titanium surfaces were examined using X-ray photoelectron spectrometer (XPS, SPECS, Berlin, Germany) with Al Kα radiation under 10-10 bar pressure. The 1.5X SBF used in the experiments was TRIS (tris-hydroxymethyl-aminomethane)-HClbuffered and has the pH value adjusted to 7.4. All the reagents (NaCl, NaHCO3, KCl, Na2HPO4, MgCl2.6H2O, CaCl2.2H2O, Na2SO4, Tris and HCl) used to prepare the SBF were analysis grade and supplied from Merck Chemicals. The detailed preparation procedure for one liter of 1.5X SBF was given by Jolata et al. [14]. All the modified titanium discs described above were soaked at 36 ºC in 125 ml of 1.5X SBF in closed Nalgene® plastic bottles for 1, 2, 3 and 4 weeks. All the solutions were replenished in every 2 days to maintain the ion concentrations. At the end of each soaking, the coated discs were carefully removed from the SBF solution. The discs were then rinsed with distilled water and dried slowly in a closed and humid environment to prevent crack formations. The coatings were characterized by an X-ray diffractometer (XRD, Bruker Advance D8, Bruker AXS GmbH, Karlruhe, Germany) with Cu Kα radiation. The surface morphologies of the bio-mimetic apatite coatings were examined under a scanning electron microscope (SEM, PhilipsXL30, Amsterdam, The Netherlands).

Results and Discussion Surface Analysis of the Hydrothermally Treated Titanium Substrates. XPS studies were carried out in order to reveal the changes on the titanium surfaces during the hydrothermal treatment process. To make the comparison, metallographically prepared and untreated titanium substrate was selected as the reference material in the analysis. The other two specimens were hydrothermally treated in: (i) de-ionized water and (ii) 0.02 mol/lt CaO containing de-ionized water. Ca 2p signal was selected as the reference in the present XPS study of the surfaces and the results were compared with the literature [15,16]. Ca 2p signal splits into two Ca 2p1/2 and Ca 2p3/2 signals. The untreated titanium surface did not yield any Ca 2p peaks in the XPS spectrum (Fig. 1). The weak background signal from the specimen treated only in de-ionized water is due to residual Ca impurities in the hydrothermal chamber. On the other hand, the strong Ca 2p peaks were obtained from the specimen treated in Ca containing de-ionized water. The 2p1/2 peak and the 2p3/2 peak of Ca 2p shell were obtained at 349 eV and 346 eV, respectively (Fig. 1). The binding energy of Ca 2p shell changes depending on the compounds formed.

404

12th INTERNATIONAL CERAMICS CONGRESS PART B

Fig. 1. XPS peak pattern for Ca 2p.

Gerth et al. [16] reported that the 2p3/2 peak was located between 346.85 and 346.95 eV when Ca is bound within hydroxyapatite structure, located at 347.55 eV when Ca is bound within calcium fluoride structure and located at 345.80 eV when Ca is bound within Ca(OH)2 structure. On the other hand, when calcium is bound in CaTiO3 structure, the Ca 2p3/2 peak is located at around 346 eV, as reported by Asami et al. [15]. In the present investigation, the calcium implanted titanium surface yielded the Ca 2p3/2 peak at around 346 eV. Therefore, this peak can be attributed to Ca(OH)2 or CaTiO3. The results clearly indicate that Ti 2p peaks are lacking only for the titanium specimen which was hydrothermally modified in Ca-containing environment. This may be interpreted as: (a) the outer-most layer is made up of only Ca(OH)2, and (b) beneath the surface, CaTiO3 is formed as a thin film on the substrate. Crystal Structure of The Coatings. XRD analysis was employed to examine the crystal structure of the bio-mimetically grown layers as well as the white colloidal precipitates collected from the bottom of the plastic bottles for comparison. The XRD patterns of the different types of specimens, namely the precipitates and the coated substrates that were pre-treated under several hydrothermal conditions, is shown in Fig. 2 as a function of soaking time. The strong titanium peaks were observed for all the coated specimens due to the metal substrates. For both the precipitates and the coatings, two of the strongest peaks are located at 2θ degrees of ~26° and ~32°. These peaks belong to an apatite mineral and their intensities become more significant for the coated substrates generally after two weeks of immersion as the coating layers become thicker. The coatings on the substrates that were pre-treated in Ca-containing environment and de-ionized water were not thick enough at the beginning (1 and 2 weeks) and, therefore, did not yield any significant apatite peak (Figs. 2 (a) and (b)).

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

405

Fig. 2. XRD patterns of the precipitates and the coated substrates pre-treated in (a) 230°C Cacontaining de-ionized water and (b) 200ºC de-ionized water.

Microstructural Analysis of the Coatings. Apatite coated titanium substrates were analyzed under SEM to reveal the morphology of the coatings. The titanium substrate surfaces were completely covered with apatite at all immersion times. One week of immersion time, however, was not sufficient for the substrates to be completely covered with apatite if hydrothermal pre-treatments performed in Ca-containing de-ionized water. Although the nucleation and growth of apatite mineral started, one-week-soaked substrates were not covered completely. After two weeks of soaking, the coatings were complete on the entire surfaces. Apatite globules became combined into one and smoother coating layer. Crack formation in the coating layers may occur upon drying. Since bio-mimetically grown apatite has a porous structure, contraction of the coatings occurs due to evaporation. Crack formation was observed after two weeks of immersion and upon drying for the specimens that were pre-treated in Ca-containing de-ionized water at 200 °C (Fig. 3 (f)). In general, immersion times and hydrothermal treatment conditions influenced the quality of the coatings. Development of the apatite layers as function of soaking time are shown in Fig. 3 and Fig. 4 for the pre-treatments in Ca-containing de-ionized water and in de-ionized water, respectively.

406

12th INTERNATIONAL CERAMICS CONGRESS PART B

Fig. 3. SEM micrographs of the coatings on substrates pre-treated at 230 °C for (a) 1 week, (b) 2 weeks, (c) 3 weeks, (d) 4 weeks, and pre-treated at 200 °C for (e) 1 week, (f) 2 weeks, (g) 3 weeks, (h) 4 weeks. All pre-treatments performed in Ca-containing de-ionized water.

Fig. 4. SEM micrographs of the coatings on substrates pre-treated at 230 °C for (a) 1 week, (b) 2 weeks, (c) 3 weeks, (d) 4 weeks, and pre-treated at 200 °C for (e) 1 week, (f) 2 weeks, (g) 3 weeks, (h) 4 weeks. All pre-treatments performed in only de-ionized water. As the coating thickness increased with soaking time (especially for two weeks and longer time periods), crack formations in coatings were observed regardless of the chemistry of the hydrothermal environment. Substrates treated in Ca-containing hydrothermal environment and at 230 °C were, however, able to develop crack free and complete apatite coatings at two weeks of soaking time after drying (Fig. 3(b)).

4. Conclusions The selected SBF for the present investigation was able to produce bio-mimetic apatite coatings on each of the substrates pre-treated under all hydrothermal conditions. All the pre-treated titanium substrates were coated completely with the mineral layer if reasonable amount of soaking time allowed. However, cracks are formed after drying. It was found that the crack formation was more severe for the samples that were subjected to hydrothermal pre-treatment in the absence of Ca ions, resulting in de-bonding of the coatings. It was generally observed that the integrity between the coatings and the Ca-ion-implanted titanium substrates was reasonably good although cracks formed in thick coatings. Within the scope of this investigation, the applied hydrothermal treatment

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

407

in Ca-containing de-ionized water environment at 230 oC and two weeks of SBF immersion yielded completely covering and crack-free apatite coatings after drying. It is concluded that the presence of Ca ions on the titanium metal substrates has favorable effects in bio-mimetic mineralization of apatite from the SBF. This result could be attributed to a more controlled mineralization of apatite from the SBF in terms of slower accumulation rates, resulting in better quality coatings. Acknowledgements We kindly acknowledge the support for this work from The Scientific and Technological Research Council of Turkey – TUBITAK, Project 107M374. We also wish to thank Gebze Institute of Technology Nanotechnology Research Center and Dr. Osman ÖZTÜRK for their help with XPS analysis.

References [1] X. Chen, Y. Li, P. D. Hodgson, C. Wen, Mater. Sci. Eng.. C 29 (2009) 165. [2] L. Jonasova, F. A. Müller, A. Helebrant, J. Strnad, P. Greil, Biomaterials 25 (2004) 1187. [3] M. Nakagawa, L. Zhang, K. Udoh, S. Matsuya, K. Ishikawa, J. Mater. Sci., Mater. Med. 16 (2005) 985. [4] De Andrade MC, Sader MS, Filgueiras MR, Ogasawara T, J. Mater. Sci., Mater. Med. 11 (2000) 751. [5] R. Sultana, M. Kon, L. M. Hirakata, E. Fujihara, K. Asaoka, T. Ichikawa, Dent. Mater. J. 25 (2006) 470. [6] X.-B. Chen, Y.-C. Li, J. D. Plessis, P. D. Hodgson, C. Wen, Acta Biomater. 5 (2009) 1808. [7] B. Feng, J. Y. Chen, S. K. Qi, L. He, J. Z. Zhao, X. D. Zhang, Biomaterials 23 (2002) 173. [8] H. S. Ryu, W.-H. Song, S.-H. Hong, Curr. Appl. Phys. 5 (2005) 512. [9] R. Narayanan, S. K. Seshadri, T. Y. Kwon, K. H. Kim, J. Biomed. Mater. Res. B 85B (2008) 279. [10] S. Bharati, M. K. Sinha, D. Basu, B. Mater. Sci. 28 (2005) 617. [11] T. Kokubo, Acta Mater. 46 (1998) 2519. [12] T. Kokubo, T. Matsushita, H. Takadama, J. Eur. Ceram. Soc. 27 (2007) 1553. [13] S. Jalota, S. B. Bhaduri, A. C. Tas, Mater. Sci. Eng.. C 28 (2008) 129. [14] S. Jalota, S. B. Bhaduri, A. C. Tas, J. Mater. Sci., Mater. Med. 17 (2006) 697. [15] K. Asami, N. Ohtsu, K. Saito, T. Hanawa, Surf. Coat. Tech. 200 (2005) 1005. [16] H. U. V. Gerth, T. Dammaschke, E. Schäfer, H. Züchner, Dent. Mater. 23 (2007) 1521.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.408

Highly porous hydroxyapatite ceramics for engineering applications Hrvoje Ivankovic1,a, Sebastijan Orlic1,b, Dajana Kranzelic1,c and Emilija Tkalcec1,d 1

University of Zagreb, Faculty of Chemical Engineering and Technology, Zagreb, Croatia a

b

c

d

[email protected], [email protected], [email protected], [email protected]

Keywords: Hydroxyapatite, Cuttlefish bone, DTA-TG-EGA-FTIR, Microstructure Abstract Highly porous hydroxyapatite (Ca10(PO4)6(OH)2, HA) was prepared through hydrothermal (HT) transformation of aragonitic cuttlefish bones (Seppia Officinalis L. Adriatic Sea) in the temperature range from 140°C to 220°C for 20 minutes to 48 hours. Mechanism of hydrothermal transformation of bones was investigated by DTA/TG analyzer coupled online with FTIR spectrometric gas cell equipment (DTA-TG-EGA-FTIR analysis), X-ray diffraction analysis (XRD) and scanning electron microscopy (SEM). DTA-TG-EGA-FTIR analysis have shown the release of CO2 at about 400°C, 680°C and 990°C. The first release could be attributed to organics not completely removed from the heat treated bones, and the second release to decomposition of unconverted aragonite, whereas, the third one could be attributed to CO32– groups incorporated in the structure of HA. The interconnecting porous morphology of the starting material (aragonite) was maintained during the HT treatment. The formation of dandelion-like HA spheres with diameter from 3 to 8 µm were observed, which further transformed into nanoplates and nanorods with an average diameter of about 200-300 nm and an average length of about 8-10 µm. 1. Introduction Hydroxyapatite (HA) is being extensively used as bone grafting material in hard tissue implants and as material for bone-tissue engineering applications, due to its excellent biocompatibility and osteoconductivity [1]. Non-medical applications of porous HA ceramics include packing media for column chromatography, gas sensors, catalyst and host materials [2]. The most of the synthetic HA is stoichiometric with chemical composition Ca10(PO4)6(OH)2. By contrast, HA prepared from natural sources, which primarily include corals, nacres, animal bones, exoskeletons etc. is non stoichiometric, and have other ions incorporated, mainly CO32–, trace of Na+, Mg2+, Fe2+, F–, Cl– [3]. CO32– containing HA has gained much attention as it can be more easily resorbed by the living cells in comparison with stoichiometric HA, and therefore it leads to faster bone regeneration. CO32– can be substituted for either OH– (A-type) or PO43– (B-type) groups in the structure of hydroxyapatite. Sometimes, both A-type and B-type substitutions can also occur [4]. Several attempts to convert natural aragonite structures (e.g. corals, nacres, etc.) hydrothermally (HT) to hydroxyapatite have been reported [5-9]. The reaction sequences in hydrothermal systems are complex and in most cases the information regarding the course of reactions is only partial. Various mechanisms of transformation of CaCO3 into HA are assumed in literature. The ability of fast transformation of natural aragonitic (CaCO3) structure into HA, even at room temperature, has been shown by Ni et al. [10]. Rocha et al. [7-9] were the first one who performed the hydrothermal

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

409

transformation of aragonitic cuttlefish bones into HA. The inorganic part of cuttlefish bone (also called cuttlebone) is a lamellar mineralized porous structure of aragonite. Its highly channeled structure favors the diffusion of the reaction solution towards the aragonite and its fast transformation into HA. Zaremba et al. [11] studying the aragonite transformation in gastropod (abalone) nacres suggested dissolution-recrystallization mechanism of the HA growth, whereas Yoshimura et al. [12], proposed dissolution-precipitation mechanism followed by nucleation and growth of HA on the surface of calcite (CaCO3). According to Jinawath et al. [13], aragonite in porites was initially transformed into intermediate CaHPO4 (DCPA) at pH~2-4, which at pH>6 transformed into HA. The authors also proposed dissolution-recrystallization mechanism as driving force for hydroxyapatite growth. The aim of this work was: to study the transformation mechanism of aragonitic cuttlefish bones into HA in the dependence of temperature and time of hydrothermal treatment. Incorporation of CO32– groups into HA structure has been studied by DTA-TG-EGA-FTIR analysis.

2. Materials and methods The starting materials were pieces of native cuttlefish bones (~2cm3), Sepia Officinalis L., from the Adriatic Sea, heated at 350°C for 3 h, to remove the organic part of cuttlefish bones. The bones were poured with the required volume of an aqueous solution of 0.6 M NH4H2PO4 (Ca/P=1.67) in teflon lined stainless steel pressure vessel and sealed from 140°C to 220°C in the step of 20°C for various times (20 minutes to 48 hours) in the electric furnace. The converted HA was washed with boiling water and dried at 110°C for further characterization. The conversion of HT transformation was followed by X-ray diffraction analysis (Philips PW 1820 counter diffractometer with Cu Kα radiation). To quantify the HA transformed by hydrothermal treatment, Rietveld structure refinement approach was used [14]. Fourier transform infrared spectra (FTIR) were performed by attenuated total reflectance (ATR) spectroscopy for solids with a diamond crystal. DTA-TG–EGA– FTIR analysis were monitored on DTA/TG analyzer Netzsch STA 409 with an online coupled Fourier transform infrared attenuated total reflectance spectrometer (ATR-FTIR Bruker Vertex 70). The samples were heated in DTA from room temperature to 1350°C with a heating rate of 10°C min-1 in flowing nitrogen (30 cm3 min-1). DTA was coupled to FTIR via heated transfer line connected to an interface that consisted of gas cell heated up to 200°C to prevent condensation on the windows. The FTIR spectrometer acquired 996 complete IR spectra with measuring resolution of 8 cm-1 and iteration was performed for 32 times in the range 4000–650 cm-1.

410

12th INTERNATIONAL CERAMICS CONGRESS PART B

3. Results and discussion The conversion of aragonite was followed by quantitative XRD analysis. XRD patterns of the samples HT treated for 20 minutes at various temperatures are given in Fig.1. Poorly crystallized hydroxyapatite

well

crystalline

CaHPO4·2H2O,

were

determined in the samples heat treated at 140°C and 160°C. With the increase of

a

Intensity (a.u.)

brushite,

 −brushite
− hydroxyapatite



and

 






b

HT temperature, the amount of brushite

c

decreases, so at 180°C just discernible

d

amount of brushite is detected, and at

e

200°C brushite was not observed. XRD patterns for samples HT treated at 180°C 10

20

30

40

50

2θ (°)

are shown in Fig.2. as an example. The sample HT treated at 180°C for 48 hours

Fig.1. XRD patterns of samples HT treated at: (a)-140°C, (b)-160°C (c)-180°C, (d)-200°C and (e)-220°C for 20 minutes

contained 95.4 wt% of HA and 4.6 wt%

of untransformed aragonite, while in the sample treated at 200°C for 24 hours aragonite transformed completely into hydroxyapatite. On the other hand, the sample treated at 220°C for 24 hours contained 97.9 wt % of HA, and 2.1 wt% of untransformed aragonite, while for longer treatment time (48 h), the amount of hydroxyapatite decreased on account of monetite, CaHPO4, which was formed in quantity of 3.2 wt %. The FTIR spectra of HT treated samples at 180°C are given in Fig.3. The formation of HA by HT

-silicon (002)

(030)



treatment

-aragonite -calcite

evident

according

characteristic bands of PO4

180°C/48h

Intensity (a.u.)

is

3–

to

tetrahedra:

(ν3 1042 and 1088 cm-1; ν4 602 and 563 180°C/24h

cm-1; ν1 960 cm-1 and ν2 470 cm-1) [15].

180°C/8h

With the increase of HT treatment time,

180°C/4h

the intensities and resolutions of PO43-



180°C/2h

bands are increased and OH– band at

 



3570 cm-1 appears. The small intensity of





180°C/1h

OH– bands is also characteristic for 20

24

28

ο

2θ ( )

32

36

Fig.2. XRD pattern of HT samples treated at 180°C

40

nanocrystalline

biological

apatites;

probably because of a greater amount of

carbonate ion incorporated in HA structure [16]. Accordingly, the same explanation could be valid for our results too. In general, there are two types of CO32– substitutions in apatites; the substitution at the OH– site (A-type) and at the PO43– site (B-type) [17] which is reflected on the FTIR spectra.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

411

The broad bimodal peaks at 1412 and 1446 cm-1 and the peak at 872 cm-1 confirms carbonate ion in the B-site of HA structure. When the cuttlefish bone is treated at 200°C, the OH– stretching vibrations are visible already in (B)

(A)

ν3−PO4

the samples treated for 1 hour and ν4−PO4

ν1−PO4

Absorbance /a.u.

CO3

with

unconverted aragonite has been

-1

180°C/48h

analyzed by DTA/TG analysis,

180°C/8h

and simultaneously the amount of

180°C/4h

incorporation of CO32– in the HA

180°C/2h 180°C/1h

structure has been followed by the

cuttlefish bone 3500

intense

ν2−PO4

180°C/24h

4000

more

duration of HT treatment. The

CO3

OH

became

3000

2500 2000

in situ online coupled DTA- TG-

1500

1000

500

-1

Wavenumbers / (cm )

EGA-FTIR system. The results obtained

Fig.3. FTIR spectra of hydrothermally converted HA at 180°C

FTIR

from

DTA-TG-EGA-

analysis

generally are

presented as (1) a Gram-Schmidt plot, which shows information related with the total IR absorbance of the evolved components in whole spectral range; (2) a three dimensional spectra (a stacked plot) of evolved gases; and (3) IR spectra obtained at the maximum evolution rate for each decomposition stage. DTA-TG curves are shown in comparison with the Gram-Schmidt plot for the sample HT treated at 180°C (Fig. 4.). Three evolved gases regions at ~400°C, 690°C and 990°C can be distinguished on

0,1 100

DSC

plot

characterized with the mass loss on TG curve. 3D stacked plot,

98

96

-0,1

94

TG (mas.%)

-1

DSC (µVmg )

0,0

Gram-Schmidt

Gram-Schmidt 92 TG

-0,3 0

400

800

1200

gas in the region between 320°C and 1200°C suggest that CO2 is

3.96 mas.%

-0,2

and the IR spectra of evolved

90 1600

Temperature (°C)

the sole gas evolved during the DTA-TG-EGA-FTIR

analysis.

The first release of CO2 at about 400°C could be attributed to

Fig.4. DTA/TG scans and Gram-Schmidt plot of HT treated sample at 180°C for 24 h

oxidation

of

organics

not

completely decomposed at the pretreatment of cuttlefish bone, and CO2 release at about 690°C should be attributed to decomposition of aragonite. Namely, unconverted

aragonite was also

412

12th INTERNATIONAL CERAMICS CONGRESS PART B

determined by XRD analysis. The release of gaseous CO2 between 800°C and 1060°C should be attributed to CO32– groups incorporated in HA structure. The absorption bands at about 2400 and 2270 cm-1 in Fig.5. indicate that CO32– incorporated into HA is 1193°C 1121°C 1105°C

released at about 1000°C. SEM micrographs have shown that the

993°C 937°C

general image of cuttlefish bones

873°C

was preserved after hydrothermal

745°C 689°C 609°C

treatment and the cuttlefish bones retained its form with the same

512°C 328°C 1000

1500

2000

2500

3000

3500

4000

4500

-1

Wavenumbers /(cm )

Fig.5. Evolution of gaseous species (CO, CO2) from the sample HT treated at 180°C as observed in in situ DTA-TG-EGA-FTIR measurement

channel size (~80x300 µm). The enlarged view of transformed HA (Fig.6.) indicates the existence

of

many uniform, dandelion-like HA nanostructures with diameter from 3 to 8 µm, formed on the surface of lamellae and pillars. With the prolonged

HT

treatment,

the

dandelion-like structures [18] are transformed

into

various

nanostructure branches which later form radially oriented nanoplates and nanorods with an average diameter of about 200-300 nm and an

average

length

of

about

8-10 µm.

4. Conclusions Transformation Fig.6. SEM micrographs converted cuttlefish bone at 180°C: a) general image, b) dandelion-like nanostructures, c) nanoplates and d) nanorods of HA.

of

aragonitic

cuttlefish bones into hydroxyapatite using hydrothermal treatment at temperatures between 140°C and

220°C for various times (1-48 h) has been investigated. In the initial reaction step at 140°C and 160°C small amount of brushite (CaHPO4·2H2O) was crystallized, due to acidic conditions of the suspension in the pressure vessel.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

413

DTA-TG-EGA-FTIR analysis has shown that CO32– groups incorporated into HA are released from the structure at about 990°C. The interconnecting porous morphology of the starting material (aragonite) was maintained during the HT treatment. Dandelion-like HA nanostructures were formed. Maintained 3D architecture of the natural cuttlefish bones offer promising alternatives for bone tissue engineering, bulk catalyst and host materials application.

Acknowledgment The financial support of the Ministry of Science, Education and Sports of the Republic of Croatia in the framework of the project “Bioceramic, Polymer and Composite Nanostructured Materials”, (No.125-1252970-3005) and Universidad Politecnica de Valencia, (Centro de Biomateriales) Spain is gratefully acknowledged. References 1. L.L. Hench, J. Wilson: Introduction to bioceramics, Singapore World Scientific; 1993. 2. K.Yamashita, T. Kazanawa: Inorganic phosphate Materials, ed. T. Kazanawa, Elsevier, Amsterdam; 1989 3. L.Z. LeGeros in P.W. Brown, B. Constantz (Eds. Hydroxyapatite and Related Materials, CRC Press, Boca Raton 4. R. Murugan, S. Ramakrishna: Acta Biomaterialia 2 (2006) 201 5. J. Hu, J.J. Russell, B. Ben Nissan, R. Vago: J. Mater. Letters (2001) 20; 85 6. K.S. Vecchio, X. Zhang, J.B. Massie, M. Wang, CW. King, Conversion of bulk seashells to biocompatible hydroxyapatite for bone implants, Acta Materialia 3 (2007) 910 7. J.H.G. Rocha, A.F. Lemos, S. Agathopoulos, P. Valério, S. Kannan, F.N. Oktar, J.M.F. Ferreira: Bone 37 (2005) 850 8. J.H.G. Rocha, A.F. Lemos, S. Agatholopoulos, S. Kannan, P.Valerio, J.M.F. Ferreira: J. Biomedical Mater. Res., Part A 77A [1] (2006) 160 9. J.H.G. Rocha, AF. Lemos, S. Kannan, S. Agatholopoulos, JMF. Ferreira: J. Mater. Chem. 15 [47] (2005), 5007 10. M. Ni, B.D. Ratner: Biomaterials 24 (2003) 4323 11. C.M. Zaremba, D.E. Morse, S. Mann, P.K. Hansma, G.D. Stucky: Chem. Mater. 10 (1998) 3813 12. M. Yoshimura, P. Suyaridworakun, F. Koh, F. Fujiwara, D. Pongkau, A. Ahniyaz: Mater. Sci. & Eng. C 24 (2004) 521 13. S. Jinawath, D. Polchai, M.Yoshimura: Mater. Sci. & Eng. C 22 (2002) 35 14. R.A Young, The Rietveld method International Union of Crystallography. Monographs on Crystalography, Vol 5, Oxford Press, Oxford, 1993 15. A.F. Lemos, J.H.G. Rocha, S.S.F. Quaresma, S. Kannan, S. Agathopoulos, J.M.F. Ferreira: J. Eur. Ceram. Soc. 26 (2006) 3639 16. J.D. Pasteris, B. Wopenka, J.J. Freeman, K. Rogers, E. Valsami-Jones, J.A.M. van der Houwen M.J. Silva: Biomaterials 25 (2004) 229 17. E. Landi, G. Celotti, G. A. Tampieri: J. Eur. Ceram. Soc. 23 (2003) 2931 18. J. Liu, K. Li, H. Wang, M. Zhu, H. Yan: Chem. Phys. Letters 396 (2004) 429

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.414

Lithium Disilicate Glass-Ceramic Obtained from Rice Husk-Based Silica Felipe Antunes Santos1,a, Claudinei dos Santos1,2,b, Durval Rodrigues Jr1,c, Dolores Ribeiro Lazar3,d, Dayane Faviero de Castro1,e, Daltro Garcia Pinatti1,f, Rosa Ana Conte1,g 1

EEL - USP (Escola de Engenharia de Lorena da Universidade de São Paulo) - Polo UrboIndustrial, Gleba AI-6, s/n, Mondesir, PC 116, Lorena, SP, Brazil 2 UNIFOA- MeMAT (Centro Universitário de Volta Redonda - Pró-Reitoria de Pesquisa e Extensão) - V. Redonda (RJ), Brazil 3 IPEN (Instituto de Pesquisas Energéticas e Nucleares) - Av. Lineu Prestes, 2242, Cidade Universitária, CEP: 05508-000, São Paulo, SP, Brazil a

[email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected]

d

Keywords: Lithium disilicate, silica (SiO2) from rice husk, bioceramic, dental ceramic, glassceramic, prosthesis, dental material.

Abstract. In this work lithium disilicate glass-ceramic and glasses based on the Li2O-SiO2 system have been investigated by replacing the high-purity SiO2 starting powder by silica obtained from rice husk. Glasses were developed at the stoichiometric composition of 66%.molSiO2:33%.molLiO2 using SiO2 obtained by thermochemical treatment of rice husk. The influence of rice husk-SiO2 on phase formation, microstructure, hardness and fracture toughness was determined and discussed. Investigations were carried out by means of differential thermal analysis, X-ray fluorescence, X-ray diffractometry and scanning electron microscopy. Amorphous and transparent glasses were obtained after melting. The glasses presented Tg near to 480 0C, crystallization peak at 660 0C in both glasses from different silica sources and Li2Si2O5 as the crystalline phase after heat treatment. The hardness (HV300gF) presented average values near to 430 HV for both high-purity and rice husk silica powders. Fracture toughness measurements present results near to 1.7 MP am1/2 for both compositions. Introduction Full exploitation of materials, including the use of agricultural and industrial waste in production processes, either as feedstock or as energy source, is a growing and developing alternative because of the serious environmental and economic problems that arose with industrialization. The use of alternative sources of raw materials for manufacturing of glass-ceramics has been subject of different studies [1,2]. By definition glass-ceramics are obtained by a process of controlled crystallization of suitable glass system (common glasses are amorphous, in other words, have no ordered interatomic structure). The material is first formed as a glass using the same procedure as conventional glasses, giving the desired shape, and controllably cooling to room temperature for reheating later or, instead, directly leading to the temperature at which crystal nucleation occurs at a characteristic rate [3]. These crystallization process provide superior resistance to the glass, can be produced with uniform and very small grain size and with no porosity [4]. The materials use in certain applications is made possible due to their characteristic properties: low values of thermal expansion, high levels of translucency, and in some cases high transparency, chemical stability and relatively high values of mechanical strength. Glass-ceramics based on lithium disilicate glass ceramics were first developed by Stookey [5, 6], taking as an initial basis for the development of materials on this system some compositions

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

415

derived from the stoichiometric composition of phyllosilicates Li2Si2O5 crystals. The glass formation and crystallization of phases in the binary system SiO2-LiO2 have been subject of many studies in recent years [7, 8, 9], with special attention to the stoichiometric composition of 33.33 mol% LiO2 and 66.67% mol SiO2. The nucleation of glasses based on lithium disilicate with exactly stoichiometric composition was studied in detail by several authors [9, 10, 11, 12]. They determined that the mechanism of embryo crystals near the melting temperature of glasses with stoichiometric composition was pointed out as the potential mechanism of nucleation of the crystalline phase. In addition to studies that determined the major mechanisms of nucleation of SiO2-LiO2-phase in the glass-ceramic system in the 80s and 90s, great attention was paid to microstructural analysis and improvement in chemical durability [13]. Considerable improvements were obtained by using Al2O3 and K2O to stoichiometric glasses aiming to boost the use of ceramic as a biomaterial in medicine and especially as restorative material in dental prostheses systems [14, 15]. It is worth noting that major advances in chemical durability have also been reached by the development of glass-ceramic in non stoichiometric composition. The aim of this study is to evaluate the properties of glasses based on lithium silicate obtained from two sources of silica: commercial high purity and the silica obtained from rice husk. Experimental Procedure Powder mixtures based on stoichiometric composition of 66SiO2:33LiO2 (in mol%), using both silicas, commercial and extracted from rice husk, were melted and the glasses were obtained. The effect of the SiO2 substitution on phase formation, microstructure, mechanical properties and chemical property were investigated by means of differential thermal analysis; X-ray diffractometry, X-ray fluorescence composition analysis, Vickers hardness and scanning electron microscopy. Figure 1 shows the processing route used in this work.

Figure 1 – Processing route used in this work. Melted samples with subsequent annealing were characterized. After that the samples were submitted to the crystallization heat treatment, which has the following conditions of temperature:

416

12th INTERNATIONAL CERAMICS CONGRESS PART B

475°C for 10 h for nucleation and then 563°C for 3 h for grain growth [16, 17]. The crystallized samples were mechanical characterized for Vickers Hardness (HV) and fracture toughness (KIC) using Vickers indentation method. Results and Discussion Table 1 shows X-ray fluorescence results for commercial silica and rice husk silica. Tabela 1 – XRF results of commercial and rice husk-sílica starting powders Element / Compound Rice husk silica Commercial silica (wt %) (wt %) SiO2 CaO SO3 Fe2O3 MgO K2O P2O5 MnO Al2O3 NiO CuO SrO ZrO2

99 0,4 0,3 0,2 0,1 0,1 < 0,05 < 0,05 < 0,05 < 0,05 < 0,05 < 0,05

98 < 0,05 0,07 0,09 0,5 0,09 0,7 < 0,05 < 0,05

According to the XRF analysis it was possible to check that both powders have a good purity, with a few oxide impurities. Probably because the higher oxide iron concentration in the rice husk silica, the glass obtained from this silica before heat treatment had a light green color. Figure 2 shows the XRD patterns of the powder mixtures, fused samples and heat-treated samples developed with commercial- and rice husk- silica. Comparatively, Fig. 2a and Fig 2b indicate that the mixture with rice husk-silica present an tendency to amorphization caused probably by the processing route to obtain the rice husk-silica. On the other hand, this behavior is not important because the powder mixture will be melted. Comparing XRD patterns of the glasses, Fig 2c and Fig.2d, no crystalline peaks are present in both compositions, indicating total amorphous glass-matrix of the samples. Heat treated samples, Fig. 2e and Fig. 2f, indicate high crystallization degree in both samples, with lithium disilicate peaks, which proves that the heat treatment procedure was suitable.

Li2CO3 (202) Li2CO3 (002)

400

200

0

Li2CO3 (204)

SiO2 (023) SiO2 (301)

SiO2 (211)

SiO2 (112)

SiO2 (102) SiO2 (200)

10000

Li2CO3 (111) Li2CO3 (202) Li2CO3 (002) Li2CO3 (112) SiO2 (110)

20000

600

417

Pure Mixture Li2CO3 + SiO2 (Rice husk)

Li2CO3 (311)

30000

Li2CO3 (200)

Intensity (a.u.)

800

SiO2 (100) Li2CO3 (110) Li2CO3 (111)

Intensity (a.u.)

40000

1000

Li2CO3 (-112) Li2CO3 (020) Li2CO3 (-311) Li2CO3 (021) Li2CO3 (310)

Pure Mixture Li2CO3 + SiO2 (Commercial)

Li2CO3 (111)

50000

Li2CO3 (110)

SiO2 (101)

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

0

0

10

20

30

40

50

60

70

80

0

90

10

20

30

40

50

60

70

80

90

2θ ( º )

2θ ( º )

(b)

(a) 1200 1000

Glass fused using rice husk SiO2

800

Intensity (a.u.)

800

600

400

200

600

400

200

0

0 20

30

40

50

60

70

80

0

90

10

20

30

40

2θ ( º )

25000

70

80

90

Lithium Disilicate obtained from rice husk silica - Crystallized

Li2Si2O5 (110)

10000

Li2Si2O5 (221)

15000

Li2Si2O5 (130) Li2Si2O5 (040) Li2Si2O5 (111)

20000

5000

Li2Si2O5 (223)

Li2Si2O5 (113)

Li2Si2O5 (221)

Li2Si2O5 (241) Li2Si2O5 (132) Li2Si2O5 (170) Li2Si2O5 (202) Li2Si2O5 (330)

Li2Si2O5 (200)

2000

Li2Si2O5 (002)

30000

Intensity (a.u.)

Li2Si2O5 (002) \c2

35000

Lithium Disilicate obtained from commercial silica - Crystallized

Li2Si2O5 (130) Li2Si2O5 (040) Li2Si2O5 (111)

4000

Li2Si2O5 (110)

Intensity (a.u.)

6000

60

(d)

(c) 8000

50

2θ ( º )

Li2Si2O5 (223)

10

Li2Si2O5 (113)

0

Li2Si2O5 (241) Li2Si2O5 (132) Li2Si2O5 (170) Li2Si2O5 (202) Li2Si2O5 (330)

Intensity (a.u.)

1000

Glass fused using commercial SiO2

0

0

-5000 10

20

30

40

50

2θ (º)

(e)

60

70

80

90

10

20

30

40

50

60

70

80

90

2θ (º)

(f)

Figure 2 - XRD patterns (a) powder mixture containing lithium carbonate and commercial silica; (b) powder mixture containing lithium carbonate and rice husk silica; (c) Glass obtained from commercial silica; (d) glass obtained from rice husk silica; (e) Lithium disilicate glass-ceramic obtained from commercial silica; (f) Lithium disilicate glass-ceramic obtained from rice husk silica.

Figure 3 shows the thermal behavior of Li-Si-O-glass through differential thermal analysis (DTA), for both glasses (obtained with commercial silica - S1 and with rice husk silica SA1).

418

12th INTERNATIONAL CERAMICS CONGRESS PART B

35

50

DTA - lithium disilicate - S1

30

DTA - lithium disilicate - SA1

40

25

Heat Flow (µv)

Heat Flow (µv)

30 20 15 10 5 0

20

10

0

-10

-5

-20 0

200

400

600

800

1000

1200

1400

1600

0

Temperature (°C)

200

400

600

800

1000

1200

1400

1600

Temperature (°C)

(a) (b) Figure 3 – Differential thermal analysis (DTA) of lithium disilicate obtained from: (a) commercial silica (S1); (b) rice husk silica (SA1). Thermal analysis allows to verify some important transformations in materials. The thermal variation of the sample during the test provides approximate information as glassy transformation temperature (TG) and approximate crystallization temperature, which are important to optimize heat treatment parameters. It can be observed that for both Li-Si-O glasses the thermal variations appear in similar positions. Table 2 shows the mechanical behavior of the Li-Si-O glasses and lithium disilicate glassceramics obtained from both silica-matrixes. Table 2 – Mechanical properties of the glasses and glass-ceramics Composition Vickers Hardness Fracture Toughness (HV300gf) (MPa.m1/2)

Li2O-SiO2 (Commercial silica) Li2O-SiO2 (Rice husk silica)

Before heat-treatment (glass)

After heat treatment (glass –ceramic)

Before heattreatment (glass)

After heat treatment (glass –ceramic)

425 ± 8

430 ± 5

0.75 ± 0.08

1.81 ± 0.15

435 ± 7

437 ±5

0.81 ± 0.13

1.84 ± 0.12

A significant improvement of the fracture toughness (near to 100%) has been observed in samples after heat treatment. Independently on the starting powder the crystallization heat treatment presented an important contribution for toughness increase. It is related to the crystallization of the Li2Si2O5 phase, which leads to the organization in the atomic-scale structure with grains formation and activation of the toughening mechanisms. Hardness is not significantly influenced for heat treatment. Furthermore, the different silica powders are not influenced the results. The results confirm that mechanical properties of heat-treated samples are superior than the samples before heat treatment. Heat treatment conditions were mainly chosen within bibliographic references information. BRAUN, 2008, held different treatment conditions, involving nucleation and grain growth temperature levels. One of the bests conditions to guarantee the maximum of phase crystallization was 475°C for 10 h for nucleation followed by 563°C for 3 h for grain growth, which were used in this work. The most important data collected by the mechanical characterization of the present work were the comparison of glass-ceramics obtained from different sources. The results were very similar, with HV and KIC values of glass-ceramic obtained from rice husk silica about 1,6% better than the glass ceramic obtained from commercial silica.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

419

Conclusions Li2O-SiO2-based glasses and glass-ceramics were prepared with rice-husk silica as starting powder and compared with commercial high-purity silica. Thermal behavior and crystalline phases presented for both glasses and glass-ceramics were are similar. Hardness and fracture toughness of the glasses presented results of 430 HV and 0.8 MPam1/2, respectively. After heat treatment, Li2Si2O5 was the unique crystalline phase present in both compositions. Furthermore, improvement of 100% in fracture toughness was observed for devitrification of the structure. These partial results are totally positive with respect to the replacement of commercial silica by rice husk silica. It is then possible to reduce environment impacts caused by inappropriate disposal of rice husk at the same time that glass-ceramic materials are obtained with similar mechanical properties as the ones preparared with commercial silica powder. Acknowledgment The authors acknowledge the FAPESP, CNPq, CAPES and University of São Paulo PostGraduation for the financial support. References [1] C. T. Kniess, N. C., Kuhnen, H. . Riella, E. Neves, C. D. G., Borba in: Estudo do efeito da quantidade de óxido de ferro em cinzas pesadas de carvão mineral na obtenção de vitrocerâmicos, vol. 25 of Química Nova nº6, p. 926-930 (2000). [2] E. J. Siqueira, I. Yoshida, L. Pardini and M. Schiavonet: Ceramics International, Vol. 35 (2007), p. 213-220 [3] Z. Strnad in: Glass-Ceramic Materials: Glass Science and Techonology. Vol. 8, Elsevier Science Publishing Company, New York (1986) [4] W. Holand, G. Beall in: Glass-Ceramic Technology. (American Ceramic Society, Ohio 2002) [5] S. D. Stookey: Ind. Eng. Chem., Vol. 45 (1953), p. 115-118 [6] S. D. Stookey: Ind. Eng. Chem., Vol. 51 (1959), p. 805-808 [7] X. Zheng, G. Wen, L. Song, X. X. Huang: Acta Materialia, Vol. 56 (2008), p. 549–558 [8] W. Holand, V. Rheinberger, E. Apel, C. Hoen: Journal of the European Ceramic Society, Vol. 27 (2007), p. 1521–1526 [9] E. D. Zanotto: J. Non-Cryst. Solids, Vol. 219 (1997), p. 42-48 [10] P. W. Mcmillan in: Glass-Ceramics, Ed. 2, Academic Press, NY (1979) [11] P. F., James: J. Non-Cryst. Solids, Vol. 73 (1985), p. 517-540 [12] R. Ota, N. Mashima, T. Wakasugi, Fukunaga: J. Non-Cryst. Solids, Vol. 219 (1997) p. 70-74 [13] A. Schmidt, G. H. Frischat: Phys. Chem. Glasses, Vol. 38 (1997), p. 161-166 [14] J. M., Barrett, D. E. Clark, L. L. Hench, U.S. Patent 4, 189-325. (1980) [15] J. M. Wu, W. R. Cannon, Panzera, U.S. Patent 4, 515-634. (1985) [16] S. E. Braun in: Efeito do Grau de Cristalização nas Propriedades Mecânicas de Vitrocerâmicas de Dissilicato de Lítio, Master Dissertation of Universidade Federal do Paraná, Curitiba (2008) [17] P. C. Soares Jr., E. D. Zanotto, V. M. Fokin, H. Jain: J. Non-Cryst. Solids, Vol. 331 (2003), p. 217-227

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.420

Reactive Milling and Mechanical Alloying in Electroceramics Roberto Rivas-Marquez1, Carlos Gomez-Yanez1, Ivan Velasco-Davalos1, Jesus Cruz-Rivera2 1

Department of Metallurgical Engineering, ESIQIE-IPN, Ed. 7, Unidad Zacatenco, D.F., MEXICO 2 UASLP, Av. Sierra Leona 550, Lomas, 78210, San Luis Potosi, SLP, MEXICO [email protected], [email protected], [email protected], [email protected]

Keywords: Mechanical activation, varistors, ZnO

Abstract:Using Mechanical Activation it is possible to obtain small grain size and good homogeneity in a ceramic piece. For ZnO varistor devices Mechanical Activation appears to be a good fabrication technique, since good homogeneity and small grain sizes are advantageous microstructural features. The typical formulation is composed by ZnO, Bi2O3, Sb2O3, CoO, MnO2 and Cr2O3 as raw materials, and during sintering, several dissolutions and reactions to form pyrochlore and spinel phases occur. When Mechanical Activation is applied to the entire formulation, it is difficult to know what processes are being mechanically activated due to the complexity of the system. The aim of the present work was to clarify how the mechanical activation is taking place in a typical ZnO varistor formulation. The methodology consisted in the formation of all possible combinations of two out of the five oxides above mentioned and to apply mechanical activation on the mixture of each pair of powders. The results showed that systems containing Bi2O3 are prone to react during mechanical activation. Also, reduction reactions were observed in MnO2. In addition, the powder mixture corresponding to the whole formulation was milled in a planetary mill, pressed and sintered, and varistor devices were fabricated. Improvement in the non-linearity coefficient and breakdown voltage was observed. Introduction Varistors are protecting devices against electrical surges. Several materials have been used to fabricate varistors including SiC but the most common material is ZnO. The commercial formulation developed by Matsuoka is [1,2]: 97 % ZnO+ 1 % Bi2O3+ 0.5 % Sb2O3 +0.5 % CoO+ 0.5 % MnO+ 0.5 % Cr2O3

(1)

Some additives such Sb2O3 have the purpose to control the grain growth while other additives aim to increment the non linearity in the current-voltage curve (I-V) which is the main characteristic of a varistor. The non linearity in the I-V curve is quantified by the non linearity coefficient α; the higher the value of α the better is the varistor. Other important parameter that qualifies the performance of a varistor is the breakdown voltage, VR, which is related to the grain boundaries presented in between the device electrodes by the next expression [1,2]: VR= n Vg

(2)

Where n is the number of grain boundaries in between the device electrodes, and Vg is the breakdown voltage per intergranular barrier. The expression 2, implies that the smaller the grain size the higher the breakdown voltage of the device. It can be concluded hence, that small grain size and good homogeneity in the mixture are two key microstructural features seeked in a varistor in order to obtain good performance. These microstructural characteristics can be accomplished by using Mechanical Activation which is a processing technique consisting in the high energy milling of the powder mixture. During milling, an important particle size reduction might occur and good

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

421

homogeneity is produced in the mixture. It is common moreover, that dissolutions and chemical reactions, among the raw phases, take place during milling [3]. Some economical advantages could result from this process because the milled powder would need either shorter sintering time and/or lower sintering temperature as compared to the conventional process since the phases that normally are formed during sintering could be produced during milling or since the milled powders present larger surface area. Some researching works have been focused on the production of varistors with superior performance as compared to the commercial analog using mechanical activation [4,5]. Encouraging results have been reported. However, since the formulation (1) is complex, not much efforts have been made to determine how the mechanical activation is carried out in the commercial varistor formulation (1). Our group analyzed the mechanical activation on the system ZnO – Bi2O3 – Sb2O3 finding the formation of Bi3SbO7 during milling [6]. Other group [7], has found the formation of ZnCr2O4 while using high-energy milling on the system ZnO-Cr2O3. N. Nicolic et al. [5] found the formation of Zn2SnO4 by milling in the system ZnO-SnO. The kinectics of the reactions occurring during sintering, that is , by thermal activation has been the issue in several investigations. Two important phases formed during sintering are the pyrochlore (Zn2Bi3Sb3O14) and the spinel (Zn7Sb2O12). M. Inada et al. [8,9] found the dissolution of Co and Mn in ZnO and that Cr forms Bi rich phases with Bi2O3. Moreover, there is a strong interaction among the pyrochlore, spinel and liquid (Bi rich) phases and the other components. The aim of this work was to analyze the mechanical activation of the mixture (1). Experimental All the reactives used during this work were reagent grade. The reactives composing the commercial formulation (1) were combined by pairs of reactives resulting 15 pairs. Each pair was milled in a planetary mill (Fritch, Pulverissete 5.0) during 1, 3 and 6 hours using ZrO2 balls (3 mm of diameter) and a nylamide M container using a balls/sample ratio of 10:1 and a speed of 180 RPM. Three of the 15 pairs were not milled because these pairs were previously studied and the results were published elsewhere [6]. The mixtures studied are the following: (2) ZnO + 3 wt% MnO2, (3) ZnO + 3 wt% Cr2O3, (4) ZnO + 3 wt% Co3O4, (5) Sb2O3 + 50 % MnO2, (6) Sb2O3 + 50 % Cr2O3, (7) Sb2O3 + 50 %Co3O4, (8) Bi2O3 + 50 % MnO2, (9) Bi2O3 + 50 % Cr2O3, (10) Bi2O3 + 50 % Co3O4, (11) MnO2 + 50 % Cr2O3, (12) MnO2 + 50 % Co3O4, and (13) Co3O4 + 50 % Cr2O3. The concentrations of the mixtures above enlisted, were chosen to keep the relative concentrations present in the formulation (1). In mixtures involving ZnO (2-4), was chosen 3 wt% of the other component in order to have enough reactant to be detected by X-Ray Diffraction (XRD) and, at the same time, to keep approximately the relative concentration present in formulation (1). Transformations and reactions were monitored by X-Ray Diffraction. The particle size was analyzed by Scanning Electron Microscopy and Image Analysis. The whole formulation (1) was milled during 3 hours, uniaxially pressed at 300 MPa and sintered at 1200 °C for 2 hours. Discs of 1 cm of diameter and 1 mm thick were sliced by a slow cutting saw. On both sides of the discs, electrods of Au-Pd were deposited through a mask by Sputtering and leads were cold solded using Silver paint. The I-V curve was obtained by a D.C Voltage Source (Hewlett Packard model 6521A, 1 KV) and an Electrometer (Keithley 6517A). Results and discussion Under the milling conditions, no mechanical activation was found in the most of the cases. Therefore, in sake of economy only the diffraction patterns where some reaction was detected will be shown. For mixtures containing Bi2O3 mechanochemical reactions were found. In case of the mixture (8) Bi2O2CO3 and Bi2Mn4O10 were detected as products (Fig. 1). In case of mixture (9) only the carbonate Bi2O2CO3 was detected as compound mechanochemically formed (Fig. 2). In case of mixture (10) the compounds Bi2O2CO3 and Bi24CoO37 were produced during milling (Fig. 3).

422

12th INTERNATIONAL CERAMICS CONGRESS PART B

Fig. 1.. XRD patterns for the mixture Bi2O3 + 50% MnO2 milled at different times.

Fig. 2. XRD patterns for the mixture Bi2O3 + 50% Cr2O3 milled at different times. In the mixtures 2, 5, 8, 11 and 12 where MnO2 was involved, it was observed a rather unusual feature: the longer the milling time, the higher the intensity of the MnO2 diffraction peaks. So, to clarify this strange observation, pure MnO2 was milled (Fig. 4). The production of Mn2O3, Mn3O4 and Mn5O8 were ere found during milling. This fact suggests that milling is reducing MnO2. In order to observe the effect of the products formed during mechanical activation in the final electrical properties the milling conditions above detailed were used to mill the whole formulation (1) for different milling times. With these mixtures, varistors were constructed. A typical typic I-V current is shown in Fig. 5 and the electrical electric parameters are presented in Table able 1.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

423

Fig. 3. XRD patterns for the mixture Bi2O3 + 50% Co3O4 milled at different times. M

M

O

+

+ *

M MnO2 O Mn3O4 * Mn2O3 + Mn5O8

Intensity (a.u.)

6 Hrs 3 Hrs 1 Hrs

0 Hrs

25

30

35

40

2θ

Fig. 4. XRD patterns for MnO2 milled at different times. 4.0 3.5

Log E [V/cm]

3.0 2.5 2.0 1.5 1.0 0.5 -10

-8

-6

-4

-2

0

2

Log J [A/cm ]

Fig. 5. I-V V curve for a sample of the formulation fo mulation (1) milled during 3 hours, pressed at 300 MPa and sintered at 1200 °C C for 2 hours.

424

12th INTERNATIONAL CERAMICS CONGRESS PART B

Table 1 Non linearity coefficient, α, and breakdown voltage for samples of powders milled (formulation 1) at different times, pressed at 300 MPa and sintered at 1200 °C for 2 hours. All the devices have the same geometry; contact area of 0.196 cm2 and 1 mm thick. Milling Time [hours] 0 1 3

Breakdown Field(EB) [v] 1684 1000 3250

α 21.55 10.15 30.03

6

2615

25.57

Conclusions Mechanical activation was carried out in the systems containing B2O3 to form Bi2Mn4O10 and Bi24CoO37. Besides, it is apparent that Bi2O3 reacts with the CO2 from the atmosphere to produce Bi2O2CO3. On the other hand, it appears that the MnO2 used during this work as a raw material, it is subjected to a reducing process by milling to produce Mn2O3, Mn3O4 and Mn5O8. Therefore it is probable that MnO would be the final product if further milling is applied. References [1] F. D. Martzloff and L. M. Levinson, in: (Editor), Electronic Ceramics, edited by L. M. Levinson, Marcel Dekker, New York, (1988). [2] D. R. Clarke, J. Amer. Ceram. Soc., Vol. 82(3) (1999), p 485 [3] M. S. El-Eskandarany, Mechanical Alloying for fabrication of advanced engineering materials (William Andrew Publishing, Norwich, NY., 2001). [4] C. P. Fah and J. Wang, Solid State Ionics, Vol. 132 (2000), p. 107 [5] N. Nicolic, T. Sreckovic and M. M. Ristic, J. Euro. Ceram. Soc., Vol. 21 (2001), p. 2071 [6] C. Gomez-Yanez, J. Velazquez-Morales and E.G. Palacios, J. Electroceram., Vol. 13 (2004), p. 745 [7] Z. V. Marinkovic, L. Mancic, P. Vulic and O. Milosevic, J. Euro. Ceram. Soc., Vol. 25 (2005), p. 2081 [8] M. Inada, Jpn. J. Appl. Phys., Vol. 17(1) (1978), p. 1 [9] M. Inada and M. Matsuoka in: Additives and Interfaces in Electronic Ceramics, edited by: M. F.Yan and A. H. Heuer, The American Ceramic Society, Columbus, Ohio, 1983.

© (2010) Trans Tech Publications, Switzerland doi:10.4028/www.scientific.net/AST.63.425

Crystal Growth of Calcite Nano-plate by Alternate Soaking Method, using CDS Crystals K.HAYASHIa, M.Tomoharab, K.Fujino, G.Sakanec, Y.Katayama LSSC, Department of Chemistry Okayama University of Science a

[email protected], [email protected], [email protected]

Keywords: Crystal growth, Surfactant, Calcium dodecyl sulfate, Calcium carbonate, Calcite, Alternate soaking.

Abstract Hetero-epitaxial growth of calcite crystals on calcium dodecyl sulfate (Ca(DS)2 crystals was studied by alternate soaking method.

= CDS) single

The calcite (006) oriented single crystals grow on the (001) surface of the CDS. The hetero-epitaxial growth mechanism is discussed by the lattice matching of the a-c planes of calcite and CDS according to the structure data of the CDS single crystal. Introduction The two main subjects of calcium carbonate researches are in shape control and polymorph control [1]. These subjects sometimes cannot separate each other. However, the aim of the research has been focused separately. The aim of the present investigation is the shape control of calcium carbonate crystals. The three principle methods, (A) inhibitor method [2], (B) impurity method [3], (C) template method [4], have been performed for the shape control of calcium carbonate. Surfactants [5], amino acids [6], and proteins [7] have been used for the inhibitors of crystal growth. The surfactant monolayer was used as the template or substrate for calcium carbonate single crystal growth by Mann et al. [8]. They found that the (110) oriented calcite crystals were grown on the stearic acid monolayer and the (001) oriented calcite crystals were grown on the n-eicosyl sulfate monolayer. Nan group used the CDS to prepare CaCO3 crystals. They did not use the CDS for the template but for the surfactant solution [9]. The CDS single crystals were used as the template in the present experiment. The single crystal can be more rigid and less defective than the monolayer. We expect that the single crystal template is better than the monolayer template for the growth of calcite single crystal films to be used for alternate soaking experiment. The crystal structure of the CDS was just determined by Sakane in our laboratory [10]. We will discuss the crystal growth of calcium carbonate film on the basis of the structure data of the CDS as the template of hetero-epitaxial growth. Experimental The CDS single crystals were grown in the aqueous solution of sodium dodecyl sulfate (SDS) and

426

12th INTERNATIONAL CERAMICS CONGRESS PART B

calcium chloride at 50oC. The CDS single crystals were fixed on the plastic net, and alternately soaked in the calcium chloride aqueous solution and the ammonium carbonate aqueous solution. At 25cycles, 50 cycles, and 100 cycles, a part of the sample was cut out and examined by powder X-ray diffraction method and optical microscope observation. Results and Discussion The morphology and the habit of the CDS single crystals are shown in Fig.1. The crystals are (001) - oriented, thin and soft. There are many steps on (001) surface. The steps of screw growth are often observed on (001) plate (shown in Fig.1 B.). The thickness of the CDS crystal is about 10µm and the largest dimension is below 0.1mm. The crystal structure of the CDS is similar with a CdI2-type structure. Two dodecyl sulfate (DS) chains and a calcium atom correspond to two iodine atoms and a cadmium atom of CdI2 structure, respectively.

A Fig.1

The (110) projection of the CDS crystal structure is shown in Fig.2.

B

Morphology and habit of CDS single crystal. A: (001) oriented single crystal plate of 100×250×10µm3 B: Trace of screw growth on (001) plate

The DS chains tilt to the calcium atom plane. As the DS chain is very long, the interlayer Ca-Ca distance is 2.92nm long. On the other hand, the intralayer Ca-Ca distance is 0.54nm long and the intralayer distance is 11% larger than that of the calcite. The intralayer calcium atom arrangement of the CDS is slightly distorted from a hexagon plane lattice. The axes cross angle is 118o. The Ca-Ca distance matching diagram of the CDS and the calcite structures is shown in Fig.2. In an epitaxial growth point of view, we expected the possibility of hetero-epitaxial growth of the calcite (006) oriented film on the CDS (001) surface. Calcite crystal is placed on the CDS crystal. CO32- ions are represented as triangles. On this section, the CO3 triangles seem short lines.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

a

xy

b

xy z

z

c

Fig.2

a

b

427

c

(010) section of calcium dodecyl sulfate (CDS) crystal structure and fitting of Ca-Ca

distances of CDS and Calcite The CDS structure is the lower and the calcite structure is the upper. Concerning the CDS structure, Blue large circle: Ca, Yellow medium circle: S, Red small circle: O, Zigzag chain:C12H25 Vertical Ca-Ca vector: c-axis, Horizontal Ca-Ca vector: a axis, c=2.92nm, a=0.54nm. Concerning the calcite structure, Yellow circle: Ca, Black triangle (seems short line in this section): CO3. The powder XRD patterns of the calcite crystals grown by the alternate crystal growth method on the CDS template after 0 cycle, 25 cycles, 50 cycles, and 100 cycles are shown in Fig3. The strong peaks at low angle are assigned to the CDS 00l reflections. Hence the CDS template crystals are oriented to the 001 direction. The small peak at 31.4o and 65.8o can be assigned to the 006 and 0012 reflections of the calcite. The intensities of these small peaks become stronger, as the number of soaking cycle increases. Moreover, the low angle peak heights of the CDS markedly fall down as

428

12th INTERNATIONAL CERAMICS CONGRESS PART B

the number of cycle increases. This phenomenon is resulted from the X-ray beam absorption by the calcite film. The intensity is proportional to the exponential of minus inverse sine theta. Hence the extinction effect of low angle X-ray intensity is huge. These experimental results suggest that the calcite (006)-oriented films grow on the CDS (001) surface and the increment of the alternate soaking cycle makes the calcite (006)-oriented films thick. The theoretical thickness of the calcite film after 25cycles alternate soaking is 7.1nm, and after 50cycles, 14.2nm.

16000 I Intensity/cps

12000 8000 II 4000 III 0 0

20

40 2θ/o

60

80

Fig.3 X-ray diffraction patterns of calcite film grown on CDS (001) surface. I: 0 cycle, II: 25 cycles, III: 50 cycles. The red arrows indicate 006 reflection of calcite. Mann et al observed the 006-oriented triangle pyramidal island crystals on the n-eicosyl sulfate (ES) monolayer template [9]. The shape difference can be explained by the effect of the alternate soaking method. The alternate soaking process can control the level growth on the CDS crystal surface. Both results suggest that the orientation of the calcite crystals can be controlled by the functional group, sulfate. Calcite island crystals were observed on the calcite film after 25 cycles with the optical microscope. The photographs of the island crystals are shown in Fig.4. The small spherical calcite island crystals are randomly scattered on the calcite film grown on the CDS template. The calcite island crystals grow and change the crystal shape from sphere to parallelepiped as the number of cycle increases. The facet of the parallepiped calcite crystal may be the (01-12) surface. The orientation of the island calcite crystals is different from that of the calcite film. The crystal shape of the present island crystals is different from that of Mann’s result. The both growth mechanisms may be different.

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE

429

(2) (1)

(3) Fig.4 Optical microscope photographs of calcite film and calcite island crystals on CDS (001) platelets. The blue bar at the right bottom in each photograph indicates the 100nm scale. (1): 0 cycle, (2): 25 cycles, (3): 50 cycles. The template crystals of the CDS are the different at each run. The CDS template is formed with many oriented thin crystals. The line contrast on the photograph is resulted from the crystal boundaries. The small island crystals are also calcite single crystals oriented in arbitrary direction (possibly (01-12) direction). The island crystals of the present investigation may grow in the solution or the surface of the solution and deposit on the template surface. Hence the orientation of island crystals are independent of the orientation of the template crystal. Conclusions The hetero-epitaxial growth of calcite nano-film on the calcium dodecyl sulfate template was observed. The alternate soaking method was effective to avoid the island crystal growth. The orientation of the island calcite crystals was different from that of the calcite film. Acknowledgement The present investigation is partially supported by the financial aid of JET Corporation.

430

12th INTERNATIONAL CERAMICS CONGRESS PART B

References [1] Stephen Mann, “Biomineralization, principles and Concept Bioinorganic Materials Chemistry”, Oxford Univ. Press, (2001) Oxford, pp6-191. [2] J.Yu, X. Zhao, B. Cheng, Q. Zhang, J.Solid State Chem., 178 (2005) 861-867. [3] W.K.Park, S.-J. Ko, S.W. Lee, K.-H. Cho, J.-W. Ahn, C.Han, J.Cryst.Growth 310 (2008) 2593-2601. [4] T.Serizawa, T.Tateishi, D.Ogomi, M.Akashi, J.Cryst.Growth 292 (2006) 67-73. [5] A.Szczes, J.Cryst.Growth 311( 2009) 1129-1135. [6] A.-J. Xie, Y.-H. Shen, C.-Y. Zhang, Z.-W. Yuan, X.-M. Zhu, Y.-M. Yang, J.Cryst.Growth 285 (2005) 436-443. [7]Stephen Mann, “Biomineralization, principles and Concept Bioinorganic Materials Chemistry”, Oxford Univ. Press, (2001) Oxford, pp117-119. [8] Stephen Mann, “Biomineralization, principles and Concept Bioinorganic Materials Chemistry”, Oxford Univ. Press, (2001) Oxford, pp171-174. [9] Z.Nan, Q. Yang, Z. Chen, J.Cryst.Growth 312 (2009) 705-713. [10] submitted to Acta Cryst. Section E Structure Report, on may 19 (2010). G.Sakane, M.Tomohara, Y.Katayama, K.Hayashi, “Calcium dodecyl sulfate”.

Keywords Index 3-D Al2O3 3-D Ti 3-D Ti-6Al-4V FGM

322 322 322

A α -Al2O3 Absorbency Advanced Ceramic Aerogel Ageing Duration Al-Based Alloy Al-O-N System Al2O3 Alternate Soaking Alumina Alumina Ceramic Aluminum Nitride (AlN) Atomic Structure

91 85 236 147 152 348 263 79, 158 425 175, 332 74, 364 24 358

B B4C B4C-TiB2 Band Gap Formation Battery Biliquid Foam Bioceramic Biomimetic Bismaleimide

68, 79 68 141 14 97 414 402 120

C C10TAB Calcite Calcium Carbonate Calcium Dodecyl Sulfate Carbide Carbon Combustion Synthesis Carbothermal Nitridation Catalyst Catalyst Support Catalytic Property CaWO4 Centrifugal Force Ceramic Ceramic Laminates

152 425 425 425 273 236 24 187 370 287 246 302 228, 322 364

Ceramic Preform 164 Ceramic System 203 Ceria Nanoparticle 107 Cermet 273 Circular Wave-Guide Fixture 85 Clay 147 Co-Doping 36 Coating 197, 383, 402 Coefficient of Thermal Expansion 120 CoFGM 322 Combustion Synthesis (CS) 187, 197, 228, 263 Comfort 107 Composite 147, 263, 332 Composite-Layered Material 302 Composites Processing 263 Conversion 297 Copper Pyrovanadate 358 Corn Starch (CS) 364 Crystal Growth 425 Cuttlefish Bone 408

D Defect Crystal Structure Densification Mechanism Dental Ceramic Dental Material Die Pressing Dielectric Material Doping Double SHS DTA-TG-EGA-FTIR

352 62 414 414 1 141 74 246 408

E Electrode Electromagnetic Energy Localization Electronic Structure Electrophoretic Deposition (EPD) Epoxy

14 141 358 340 114

F Fast Reaction Ferroelectric Material Fiber

203 7 85

432

12th INTERNATIONAL CERAMICS CONGRESS PART B

Finite Element Model (FEM) Flow Forming Fly Ash (FA) Freeze Casting Full Dense Functional Graded Functionally Graded Material (FGM) Furnon Fusion Reactor

58 348 181 340 374 340 322, 332, 370, 383 312 383

G Gasless Combustion Glas Ceramics Glassy Matrix Gold Nanorod Good Shape Capability Gradient Porosity

213 131, 181, 414 131 126 374 175

H Hardness HCI Leaching High Disperse Filler High-Strength High Temperature High Temperature Laser Confocal Microscopy High-Temperature Synthesis (SHS) High-Temperature TEM Hot-Pressing HPHT Hybrid Hybrid Foam Hybrid Material Hydrothermal Modification Hydroxyapatite (HA)

392 251 131 374 85 181 297 47 282 396 120 114 97 402 408

I Image Analysis In Situ Observation Insulation Insulation Coating Integrative Chemistry Intermetallic Intermetallic Compound (IMC) Ion Beam Sputtering

364 47 41 131 97 197, 228 370 392

J Joining

197

K Kinetics

203

L Lanthanum Hexaaluminate Layered Double Hydroxide (LDH) Leaching Li2SNO3 Li2TiO3 Light-Weight Lining Lithium Disilicate Lithium Niobate Loss on Ignition Low Processing Temperature

332 352 246 352 352 374 312 414 7 181 374

M Macrostructural Transformation MAX Phase Measurement Mechanical Activation (MA) Mechanical Property Mechanically Activated Solid Powder Mixture Mesoporous Hydroxyapatite Mesoporous Silica Metal Metal Alkoxide Metal Ion Metal Matrix Composite (MMC) Metalloceramic Micro-Channel Micro Pattern Micro-Wave Microballon Microgravity Microreactor Microstructure Microwave Hybrid Heating Microwave Sintering Mirror Substrate Modeling Monodisperse

222 282 85 420 348, 396 213 152 126 187, 322 7 36 273 297 370 141 85, 197 114 302 370 47, 364, 392, 408 332 91 374 58 30

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE Mortar Mullite Multi-Layer Municipal Solid Waste (MSW)

312 91 392 181

N Nano-Composite Nano Grain Nano Structure Nanoclay Nanoparticle Nanopowder NbC Negative Thermal Expansion Niobium Co-Doped SrTiO3 Nitrogen Co-Doped SrTiO3 Numerical Simulation

257, 396 47 41 114 7 1, 187 257 358 52 52 58, 213

O Open Porosity Oxide Oxide Nanoparticle

164 187 236

P Panoscopic Assembling Particle Size Control PH Adjustment Photocatalytic Photonic Crystal Photonic S-Catalysis Plasma Facing Material (PFM) Plastic Deformation Plate-Like Ceria Plate-Like Titanate Polymer Polymeric Sponge Porous Ceramics Porous Film Potassium Niobate Powder Powder Metallurgy (PM) Prosthesis

107 30 152 36 141 97 383 348 30 107 147 164 97, 170, 297, 364 370 7 85 370 414

R Reaction Sintering Reactive High-Energy Milling Refractory

374 257 312

Rolling

433 348

S Sacrificial Template Self-Oscillatory Combustion Self-Propagating High Temperature Synthesis (SHS) SEM Microscopy SHS SHS Carrier SHS Catalyst SHS Process Si-C-N System Silica (SiO2) from Rice Husk Silica-Based Aerogel Silica Based Xerogel Silicon Carbide (SiC) Silicon Nitride Simulated Body Fluid (SBF) Sintering SiO2 Skin Layer Slip Casting Sodium Niobate Sol-Crystal Method Sol-Gel Sol-Gel Process Sol-Gel Processing Solution Combustion Synthesis (SCS) Solution Phase Synthesis Solution Synthesis Solvothermal Synthesis Space Application Spark Plasma Sintering (SPS) Spinel Spray Freeze Drying Stabilized Zirconia Stacking Fault Stainless Steel FGM Starch Starch Consolidation Starch Consolidation Casting (SCC) Stereolithography Strontium Titanate Structure Surface Oxide Layer

170 222 187, 203, 213, 222, 228, 273, 287 158 257, 263, 282, 302 287 287 251 263 414 41 41 374, 396 170, 396 402 263, 282 79 131 175, 364 7 7 41, 131 24, 97 14 287 126 30 52 41 58, 62, 68, 74, 79, 322 62 1 85 352 322 170 158, 175 364 141 52 297 374

434

12th INTERNATIONAL CERAMICS CONGRESS PART B

Surface Plasmon Resonance (SPR) Surfactant Synthesis

126 425 7

T Technology Terahertz Wave Thermal Treatment Thermodynamic Ti-Al-C-N System Ti-Al-C System Ti2AlC TiB2 TiO(OH)2 TiO2

312 141 14 24 263 282 282 68 352 36

U UV-Shielding

107

V Varistor Vector Network Analyzer Visible Light Photocatalyst Visible Light Responsive Vitrous Material

420 85 52 36 181

W W2B5 Water Content Water-Thinnable Binder WC Wear Resistance

246 152 164 322 322

X X-Ray Diffraction (XRD) X-Ray Powder Diffraction

246 352

Y Y2O3

79

Z Zinc Pyrovanadate Zirconia Zirconia-Toughened Alumina (ZTA) Zirconium Diboride Powder Zirconium Silicate

358 85, 91 340 251 91

Zirconium Tungstate ZnO ZrN/TiN Zro2

114, 120 420 392 322

Authors Index A Agnolon Pallone, E.M.J. Akkas, B. Akpinar, S. Algatti, M.A. Alhassan, S. Alkan, M. Alkoy, S. Annapoorani, K. Aoi, Y. Apak, B.

257 251 91 158, 175 147 251 402 1 392 74

97 297 352 358 203 1 197 170, 257 97 14 263, 282

C Carn, F. Casalegno, V. Çelik, Y. Chen, C.G. Chen, M.Y. Chlubny, L. Chmelíčková, M. Chmielewski, M. Chrysanthou, A. Collins, S. Conte, R.A. Cruz-Rivera, J.

97 197 79 114, 120 114, 120 263, 282 364 164 273 340 414 420

D de Campos, É. Degraw, A. Denisova, T.A. Deptula, A.

246, 251 41 402 414 383 41 41

E Ellison, P. Ertugrul, O.

340 91

F

B Backov, R. Baev, V.K. Baklanova, Y.V. Bamburov, V. Berlin, A.A. Binner, J. Bonamartini Corradi, A. Bressiani, A.H.A. Brun, N. Brykala, M. Bućko, M.M.

Derin, B. Dias, J.P. Dinçer, M. dos Santos, C. Du, J. Duarte, N. Durães, L.

158, 175 228 352 14

Faviero de Castro, D. Feng, Y.B. Ferraris, M. Ferreira de Lucena, E. Fitrah, R.A. Fontaine, F. Fujino, K. Furuhata, S.

414 383 197 158, 175 126 24 425 392

G Garcia, C. Gawryla, M.D. Ge, C.C. Ghouleh, Z. Göller, G. Gomez-Yanez, C. Goretta, K.C. Gregorová, E. Guette, A. Guo, S.Q. Gushin, A.N. Guthrie, R.I.L.

58 147 383 181 68, 74, 79 420 14 364 24 383 297 181

H Hamagami, J. Han, S.H. Hayashi, I. Hayashi, K. Hernandez, J. Hiraga, K. Hirano, Y.

126 197 126 425 41 62 7

436 Hoos, K.H.

12th INTERNATIONAL CERAMICS CONGRESS PART B 120

Mukasyan, A.S. Muto, H.

187 126

I Iguchi, M. Isac, M. Itoh, Y. Ivankovic, H.

370 181 374 408

J Jach, K. Julián-López, B.

164 97

164 263 425 85 126 62 297 141 396 141 408 358 91

L Łada, W. Lazar, D.R.R. Leonelli, C. Lew, K.S. Lis, J. Liu, X.W.

14 414 197 152 263, 282 107

M Maksimov, Y.M. Maksimova, L.G. Mansurov, Z.A. Martirosyan, K.S. Matsuda, A. Matsuura, K. Medvedeva, N. Mesquita, R.M. Minamidate, Y. Morita, K. Mota, R.P.

Nakano, H. Neder, R.B. Negahdari, Z.

47, 392 352 332

O

K Kalinski, D. Kata, D. Katayama, Y. Kawakami, S. Kawamura, G. Kim, B.N. Kirdyashkin, A.I. Kirihara, S. Klimczyk, P. Komori, N. Kranzelic, D. Krasnenko, T. Kusoglu, I.M.

N

297 352 312 236 126 348 358 170 30 62 158, 175

Ochoa, M. Odawara, O. Ohmi, T. Ohno, M. Ohta, N. Olczak, T. Olevsky, E. Onel, K. Onüralp, Y. Orlic, S. Öztürk, K.

41 302 370 348 141 14 58 91 68, 74, 79, 251 408 402

P Pabst, W. Pailler, R. Pietrzak, K. Pinatti, D.G. Portugal, A. Preiss, A. Prokofiev, V. Puszynski, J.A.

364 24 164 414 41 340 222 228

R Radzali, O. Rashkovskiy, S. Rivas-Marquez, R. Rocha, N. Rodrigues Júnior, D. Rosa, R.

152 213 420 41 414 197

S Sağ, C.P. Şahin, F.Ç. Sakane, G. Sakka, Y. Salvo, M. Sampaio Fernandes, R. Sano, S.

402 68, 74, 79 425 62 197 175 85

Pietro VINCENZINI, Ralf RIEDEL, Alexander G. MERZHANOV and ChangChun GE Santana, J.G.A. Santos, F.A. Sato, M. Sato, T. Schiraldi, D.A. Shilova, O.A. Shteinberg, A.S. Smolyakov, V. Su, B. Sulaeman, U. Suyama, S. Suyama, Y.

158 414 85 30, 36, 52, 107 147 131 203 222 340 52 374 7, 47

T Takahashi, K. Takamura, K. Takao, Y. Takayama, S. Tarakina, N.V. Teker, D. Tkalcec, E. Tokita, M. Tomasi, R. Tomohara, M. Trombini, V.

7 7 85 85 352 402 408 322 257 425 257

U Ungureanu, S. Uygun, B.

97 68

V Vaidhyanathan, B. Velasco-Davalos, I. Veronesi, P.

1 420 197

W Wang, C. Wawszczak, D. Willert-Porada, M.

383 14 332

X Xanthopoulou, G.

287

Y Yamada, T. Yazici, S.

7 246

Yeoh, F.Y. Yin, S. Yoshida, H.

437 152 30, 36, 52, 107 62

Z Zaza, F. Zhang, P.L. Zhou, H.B. Zhou, Z.J. Zientara, D. Zubkov, V.G.

14 36 383 383 263 352