Read April 2005 JAM text version

Published by Society for the Advancement of Material and Process Engineering

Journal of

Advanced Materials

An International Journal of Processing, Science, Characterization and Application of Advanced Materials

ISSN 1070-9789 Special Edition No. 1, 2006

Micrograph of two polyurethane materials illustrating structure changes due to processing differences.

Special Edition No. 1, 2006

Documenting Creative Advances in Advanced Materials and Processing

1

Editor-In-Chief Dr. A. Brent Strong, Lorin Farr Professor of Entrepreneurial Technology, in Manufacturing Engineering Technology, Brigham Young University Editors

Dr. Dan Adams, Associate Professor; Mechanical Engineering Department, University of Utah Dr. Scott Beckwith, Technical Director and Consultant; SAMPE, BTG Composites, Inc. Dr. Tia BensonTolle, Chief, Structural Materials Branch; Air Force Research Laboratory, Wright Patterson Air Force Base Dr. Will McCarvill, Manager; Commercial Chemistries Dr. Michael Miles, Assistant Professor; Manufacturing Engineering Technology, Brigham Young University Dr. Don Radford, Associate Professor, Director; Mechanical Engineering, Composite Materials, Manufacture and Structures Laboratory, Colorado State University Dr. Uday Vaidya, Professor; Department of Materials Science & Engineering, The University of Alabama at Birmingham Dr. Anthony Vizzini, Bill and Carolyn Cobb Chair, Head; Aerospace Engineering, Mississippi State University

Reviewers

Stanley Abkowitz, President; Dynamet Technology, Inc. Dr. Don Adams, President; Wyoming Test Fixtures, Inc. Dr. Alexander Bogdanovich, Vice President of Research & Development; 3TEX, Inc. Jack Burroughs, President; Seecor, Inc. Dr. Shive Chaturvedi, Associate Professor; Department of Civil and Environmental Engineering and Geodetic Science, Ohio State University Dr. Linda Clements, Director of Materials R&D; 2Phase Technologies, Inc. Ray Cull, Technical Director; Henkel Consumer Adhesives North America Eugene Dan-Jumbo, Sr. Principal Engineer/ Scientist; Northrop Grumman Corp. Dr. Ron Eby, Robert C. Musson Professor and Ohio Eminent Scholar; University of Akron George Epstein, Editor; Composites and Adhesives Newsletter Dr. William Frazier, Chief Technology Officer/ Chief Scientist; Naval Air Systems Command Dr. F. H. (Sam) Froes, Director/Dept. Chair; Institute for Materials and Advanced Processing, Department of Materials Science and Engineering, University of Idaho Dr. Randall German, CAVS Chair, Professor, Director; Mechanical Engineering, Center for Advanced Vehicular Systems, Mississippi State University Dr. Nick Gianaris, Senior Engineering Specialist; General Dynamics Land Systems, Wayne State University Dr. Heidi L. Schreuder-Gibson, Scientist; US Army Research Development and Engineering Command, Natick Soldier Center, Macromolecular Science Team Dr. Faramarz Gordaninejad, Foundation Professor, Director; Mechanical Engineering; Composite and Intelligent Materials Laboratory, University of Nevada, Reno Dana Granville, Chair; DoD ManTech Composites Fabrication & Processing, U.S. Army Research Laboratory Dr. Pelin Gundes Bakir, Associate Professor; Mechanics Division Department of Civil Engineering, Istanbul Technical University Dr. Richard Hale, Associate Professor; Aerospace Engineering, University of Kansas Dr. Rikard Heslehurst, CPEng., DIEAust, FRAS; School of Aerospace, Civil and Mechanical Engineering, University College, UNSW Australian Defence Force Academy Dr. Nao Igata, Professor Emeritus; University of Tokyo Dr. Brian Jensen, Senior Scientist; NASA Langley Research Center Dr. Howard Katzman, Senior Scientist; Space Materials Laboratory, The Aerospace Corporation Dr. Frank Ko, Professor; Fibrous Materials Lab, Department of Materials Science and Engineering, Drexel University Dr. Ellen Lackey, Associate Professor; Mechanical Engineering, University of Mississippi Bob Lacovara, Technical Director; ACMA Dr. Xiaodong Li, Associate Professor; Department of Mechanical Engineering, University of South Carolina Dr. Jia-Horng Lin, Associate Professor; Laboratory of Fiber Applications and Manufacturing, Department of Fiber and Composite Materials, Feng Chia University Dr. Robert Lopez-Anido, Associate Professor/ Fulbright Scholar; Department of Civil Engineering, University of Maine and IDIEM, Universidad de Chile Lee McKague, President; Composites Consulting, Inc. Dr. Joey Mead, Professor; Department of Plastics Engineering, University of Massachusetts Lowell Dr. Daniel Melo, Professor and Chairman; Department of Mechanical Engineering, Universidada Federal do Rio Grande do Norte, Brazil Dr. Robert Messler, Jr., Professor; Materials Science & Engineering, Rensselaer Polytechnic Institute Dr. Amod Ogale, Professor; Chemical and Biomolecular Engineering, Clemson University Dr. Christopher Pastore, Professor and Co-Director; Engineering and Design Institute, Philadelphia University Stan Peters, General Manager; Process Research, Inc. Dr. Kim Pickering, Professor; Department of Materials and Process Engineering, The University of Waikato,Hamilton, New Zealand Leonard Poveromo, Director; Technology Development Integrated Systems/AEW/EW, Northrop Grumman Dr. Pizhong Qiao, Associate Professor; Department of Civil & Environmental Engineering, Washington State University Dr. Anatol Rabinkin, Group Leader; Materials and Applications Development, METGLAS Solutions, Honeywell International Dr. Gary Roberts, Researcher; NASA Glenn Research Center Dr. Jack Roberts, Principal Professional Staff; Applied Physics Lab, The Johns Hopkins University George Schmitt, Director; International Programs AFRL Materials and Manufacturing Directorates Dr. Cecil Schneider, President; CEC Technologies, Marietta GA Dr. James Seferis, First Alumni Award and Research Professor; Center for Composite Materials, University of Delaware Dr. Howard Siegel, Consultant Dr. Jeffrey Waldman, Sr. Materials Engineer; Navmar Applied Sciences Corporation Charles Watson, Fellow; Organic Matrix Composites, Pratt & Whitney Yongbo Xu, Professor; Shenyang National Laboratory for Materials, Science Institute of Metal Research, Chinese Academy of Sciences

2

Journal of Advanced Materials

Journal of Advanced Materials

An International Journal of Processing, Science, Characterization and Application of Advanced Materials ISSN 1070-9789 Special Edition No. 1

5

Porous Biodegradable Polyurethane Scaffolds Prepared by Thermally Induced Phase Separation C.A. Martínez-Pérez, P. E. Garcia-Casillas, Partido Romero, A. Martínez-Villafañe, A. Duarte Moller, J. Romero-García Synthesizing Research of Nanometer Zinc Oxide by Evaporation Method Zhai Xiujing, Fu Yan, Bai Bin, and Wang Jie Statistical Approach for the Optimal Deposition of AlN Preferential Orientation Films Xiao-Hong Xu, Hai-Shun Wu, Fu-Qiang Zhang Preparation of Nickel Aluminate Spinel by Solid State Reaction Yong-Sheng Han, Jian-Bao Li, Xiao-Shan Ning, Bo Chi Use of Cermets and Ceramics Instead of Tungsten Carbide in Saws and Other Brazed Applications Thomas J. Walz Sol-Gel Synthesis and Characterisation of Alumina-Strontium Hexaluminate Composites K. Vishista, F.D. Gnanam, and H. Awaji Effect of the Composition of Rare Earth Elements on the Microstructure and Electrochemical Properties of RE(NiCoMnAl)5 Hydrogen Storage Electrode Alloys Hongge Pan, Jianxin Ma, Yongfeng Liu, Mingxia Gao, Rui Li, Changpin Chen Investigation and Fabrication of Nano-sized SnO2 Powder and Its Gas Sensing Properties Wei Yinghui, Yao Minqi, Guo Hongli, Hu Lanqing, Hou Lifeng, Xu Bingshe Nanometer CeO2 Material: Sol-gel Synthesis with Different Precursors and Strong Ultraviolet Absorption Bing Yan and Wengang Zhao Characteristics of Y2O3:Eu Phosphor Thin Films by Post-deposition Annealing Xiaosong Zhang, Kaishun Zou, and Yi Tao, Lan Li, Zheng Xu Low Temperature Polarized-dependent Spectrum Studies of ZnO Film Q. Cao and X.-Y. Li Mode-I and Mixed-Mode I/II Fracture Behaviour of Sintered Alumina at Ambient and Elevated Temperatures Sweety Kumari, N. Eswara Prasad, G. Malakondaiah, B. Bhaskar, and B. Naga Prasad Rao Pressureless Sintering of a Magnesia Doped Gelcast Alumina Ceramic C. T. Bodur and F. Cinar Sahin Study of the Damage Induced by Ar+ on the Highly Oriented Pyrolitic Graphite (HOPG ) Surface Using Extended Electron Energy Loss Fine Structure in Reflection Mode C. González-Valenzuela and A. Duarte-Moller Synthesis and Mechanical Properties of La2O3-WSi2/MoSi2 Composites Houan Zhang, Ping Chen, Siwen Tang Manufacturing and Mechanical Properties of Grids Braided from Stainless Steel/PP Functional Ply Yarn Jia-Horng Lin, Ching-Wen Lou, Shih-Hua Chiang

12

18

23

26

30

36

40

44

48

52

56

66

68

73

78

Journal of Advanced Materials, (ISSN 1070-9789), Special Edition No. 1, is published by the Society for the Advancement of Material and Process Engineering, 1161 Park View Dr., Suite 200, Covina, CA 91724-3751. Subscription rates for members (including postage) $30 US (US), $38 Canada/Mexico; for non-members (including postage) $70/year US; $78/year Canada and Mexico; $85/year outside US/Canada/Mexico; Institutions $150. Europe 30--Europe Institutions, Agencies, non-members--180. Copyright 2006­The Society for the Advancement of Material and Process Engineering. All rights, including translation, are reserved by SAMPE® 1161 Park View Drive, Suite 200, Covina, CA 91724-3751. Published by the Society for the Advancement of Material and Process Engineering. Responsibility for the contents rest upon the authors and not upon SAMPE®.

Special Edition No. 1, 2006

3

Dr. A. Brent Strong, Editor-in-Chief

Lorin Farr Professor of Entrepreneurial Technology, in Manufacturing Engineering Technology, Brigham Young University

The Journal of Advanced Materials (JAM) has been a popular journal for publishing articles concerning research into a wide variety of advanced materials and the processes which convert them into useful products. Over the last several years, this popularity has led to an accumulation of excellent papers that have all been thoroughly peer reviewed but could not be published in a timely manner because of space limitations of the journal. As I assumed the editorship of JAM, this backlog of papers was a major problem. We wanted to publish these papers promptly to ensure the highest relevancy of the research. Therefore, I asked the SAMPE executive cabinet to authorize the publication of special editions of JAM, thus allowing the backlog to be reduced. Because of the high costs of printing and distribution associated with traditional paper editions, we have elected to publish these special editions electronically. As far as possible, the authors of the papers were contacted to obtain their permission to publish the papers in this electronic format. We thank them for their understanding. It seems that this solution is the best for them (currency of publication) and for the JAM (fulfilling an obligation to honor commitments to publish in a timely manner). The special edition can be found on the SAMPE website under the JAM heading. The special editions have been roughly grouped by subject and this first special edition is focused on ceramic materials and nanomaterials, although many other related subjects are also included. I hope that you will take the time to peruse the articles. There are several gems to be found.

4

Journal of Advanced Materials

Porous Biodegradable Polyurethane Scaffolds Prepared by Thermally Induced Phase Separation

C.A. Martínez-Pérez, P. E. Garcia-Casillas Departamento de Ciencias Básicas, Instituto de Ingeniería y Tecnología, Universidad Autónoma de Ciudad Juárez Partido Romero, Cd. Juárez, Chih A. Martínez-Villafañe, A. Duarte Moller Centro de Investigación de Materiales Avanzados S.C. Miguel de Cervantes, Chihuahua, Chih. México J. Romero-García Departamento de Biopolímeros, Centro de Investigación en Química Aplicada, Saltillo, Coah. México Original Manuscript Received 04/05/03; Revised Manuscript Received 06/01/04

Abstract

Guide regeneration techniques have been recently used to heal soft and hard tissue defects. In this approach, scaffolding plays an important role. Hydroxypatite (HA) resembles the natural bone mineral and has shown good bone bonding properties. In this work, porous polyurethanes have been prepared by a thermally induced phase separation technique. Freeze drying of the separated polymer/solvent phase produced foams with co-continuous structure of interconnected pores. The microstructure can be controlled by varying polymer concentration, quenching temperature, and co-solvent utilized. This homogenizing technique can lead to the preparation of porous materials with controllable and reproducible morphology. SEM analysis showed that the pore size range varied from a few microns to a few hundred microns. Due to the interconnected pores, and their biocompatibility and bioactivity; they are promising scaffolds for bone ­ tissue engineering.

Introduction

Recently, biodegradable materials have been proposed in temporary orthopaedic implants1-5. A considerable amount of work has been focused in the elaboration of porous materials for guide bone regeneration (GBR), like polyglicolic acid6, polylactic acid7 and polyurethane8. These materials should be porous enough to induce and guide cell attachment growth, and tissue regeneration in three dimensions9,10. For implantation, the porous materials should fulfill a number of requirements among others, high porosity, a large surface area, a large pore size (between 100 mm300 mm), and a uniformly distributed and highly interconnected pore structure throughout the matrix11,12. Several methods have been reported to fabricate porous scaffold polymers. Porogen leaching13,14, emulsion freeze drying15, expansion in high-pressure gas16,17, and phase separation technique18, 19 have been reported. The porogen leaching method has been the best known method, and involves the casting of a polymer/porogen composite membrane followed by aqueous washing out of the incorporated porogen. Various porogens such as salts, carbohydrates, and polymers can be used to produce porous materials. The salt-leaching technique was suitable for controlling pore sizes by changing the size of particulates, but residual salts remaining in the scaffolds caused an irregularity problem for cell seeding and culture. Although the freeze drying/salt-leaching techniques have been used to prepare low-density polyurethane materials with a macroporous structure (100-300 mm), the large pores were due to the removal of the salt crystals, the pores Special Edition No. 1, 2006

resulted in an irregular shape, residual salts could remain in the scaffolds, and the use of TIPS was not discussed5. The emulsion freeze-drying method often results in a closed cellular structure in the matrix. The expansion technique using a high-pressured CO2 gas also resulted in a closed pore structure inadequate for cell seeding. The phase separation technique is based on thermodynamic demixing of a homogeneous polymer solvent solution into a polymer rich phase and polymer poor phase, usually by either exposure of the solution to another immiscible solvent or cooling the solution below a binodal solubility curve. In the thermally induced phase separation (TIPS), thermal energy is used as a latent solvent to induce phase separation19, 20.

Figure 1. A schematic representation of a binary phase diagram of a polymer solution showing the expected morphological variations from liquid ­ liquid phase separation. 5

The quenched polymer solution below the freezing point of solvent is subsequently freeze-dried to produce a porous structure. Figure 1 shows a schematic temperaturecomposition phase diagram for a binary polymer/solvent system. Above the binodal curve, a single polymer solution phase is formed. Cooling below the curve, polymer-rich and polymer-poor phases are separated in a thermodynamic equilibrium state to be quenched. For example, when the polymer solution is quenched from point A to point B in Figure 1, the arrested morphology from the position in the metastable zone between the spinodal and binodal curves displays a poor connected stringy or beady structure, which results from a nucleation and growth mechanism.21 On the other hand, if the system is quenched into the metastable region, the phase separation takes place in a spinodal mechanism, resulting in a microporous interconnected structure.19-21 The spinodal curve is defined as the line at which the Gibbs free energy of mixing second derivative is equal to zero, and it divides the two-phase region into unstable and metastable regions. Depending upon the location of the quenching end point, two distinctive morphologies can be obtained i.e.; located in the metastable region between binodal and spinodal curves or in the unstable region below the spinodal curve. It is preferable to use the spinodal decomposition for the production of open-pore microcelular foams. Pore size distribution and their interconnectivity are determined by a delicate balance of various parameters such as polymer concentration, quenching route, and solvent/nonsolvent composition19-21. These events occur at the early stage of phase separation, but in the later stage, the coalescence of phase separated droplets continuously proceeds minimizing the interfacial free energy associated with the interfacial area, which is called the coarsening process. This effect is induced by a differential interfacial tension exerted between the two phase separated domains. It was demonstrated that the coarsening process results in pore size enlargement primarily via Ostwald ripening, coalescense, or a hydrodynamic flow mechanism 16,18. Thus, the coarsening process should be carefully considered as a kinetic parameter to control the pore morphology of the resulting foams. This effect has been scrutinized in the fabrication of synthetic membranes, because it is of paramount importance in determining the final membrane morphology22,23. To attain macroporous scaffolds, it is desirable to use the coarsening effect, which induces the pore enlargement. The coarsening process, however, concomitantly tends to generate more closed pores; thus, it is important to optimize various TIPS parameters to achieve a macroporous open cellular structure. Large pore size and open cellular structure are critical parameters for cell seeding and neovascularization when implanted in vivo. Polyurethane (PU) has been used in many biomedical applications like vascular prostheses, meniscus reconstruction, and catheters, to mention a few. This versatility is due to high biocompatibility and a wide range of mechanical and physical properties23-25. 6

In this study, we investigated the TIPS technique to produce 3D porous biodegradable polyurethane scaffolds by adjusting parameters involved in the TIPS process. 1,4 dioxane was used for solid-liquid phase separation, water was added as a nonsolvent to induce a liquid ­ liquid phase by lowering the degree of polymer-diluent interaction. Several authors have reported that Ca, P formation on a material's surface in SBF is a indicator of their bioactivity, since bioactive materials bond to bone in vivo through similar surface layer26,27. Therefore, in the present study, we also present preliminary results of the calcium phosphate growth on the surface polyurethane previously treated with tetraethoxysilane (TEOS) immersed in 1.5 SBF solution.

Experimental Procedures

Porous Polyurethane and Composites Preparation Six equivalents of polycaprolactone diol (m.w. 1250) and two equivalents polycaprolactone triol (m.w. 900) were dissolved in 1,4 dioxane and n-hexane as co-solvent followed by the addition of nine equivalents of 1,6 diisocyanatehexane and, as catalyst, 0.5 wt.% of dibutyltin dilaurate was used. Water was added to encourage phase separation. To compare the structures resulting from the variation of solvents, the volume relation of dioxane, nhexane and water were varied as 87:13:0, 93:7:0, 87:0:13, 93:0:7. The homogeneous solutions were quenched under the following conditions: mixture CO2/acetone bath (-78°C), and two freezers at -25°C and -15°C, respectively, and then incubated overnight at the quenching temperature. Afterwards, the solutions were placed in a freeze drying apparatus (Labconco Freezone 4.5) connected to a vacuum pressure (0.15 milibar) and the solvent was removed for 48 hours. After freeze-drying, the polymer was cured under reduced atmosphere (26 inHg) at 60°C for 48 h. HA 5% wt. was added to a polymeric solution before freezing to prepare the composites. The mixture was stirred magnetically for 15 min. Preparation of SBF Solution The simulated body fluid (1.5 SBF) which had approximately 1.5 times higher ionic concentrations than human blood plasma and was prepared according to literature28 by dissolving reagent-grade NaCl, NaHCO3, KCl, K2HPO4.3H2O, MgCl2.6H2O, CaCl2 and Na2SO4 in ionexchange distilled water. The solution was buffered at pH 7.25 with tris(hydroximethyl) aminomethane ((CH2OH)3CNH2) and 1 M hydrochloric acid (HCl) at 37°C. TEOS and SBF Treatment Rectangular substrates (2 x 2 x 1 cm3) of polyurethane were cut and then soaked in 20 ml of 1M HCl for three min. in order to increase the number of polar groups and increase the affinity of the silicate ions to the substrate28. After the HCl treatment, substrates were washed with distilled water and dried at room temperature, each substrate was soaked in 40 ml. of (TEOS, 99%) the flask

Journal of Advanced Materials

Figure 2. FT-IR of a polyurethane sample. was kept at 60°C for 2 h. After cooling the flask, substrates were filtered out and washed with ethyl alcohol and then each substrate was soaked in 1.5 SBF at 37°C in the polystyrene bottle for different periods for making the apatite nuclei grow on the surface of the substrate in situ. The 1.5 SBF solution was renewed every day. After that, the substrate was washed moderately with distilled water and dried at room temperature.

Results and Discussion

A scaffold material for tissue engineering should have a high porosity and an appropriate pore size. In this work, the solid-liquid phase separation of PU/dioxane/nonsolvent and subsequent sublimation of the solvent has been used to obtain highly porous polyurethane. The FT-IR spectra of the PU developed is presented in Figure 2. The spectrum shows the characteristic peaks for polyurethane like 3330 cm-1 which are attributed to the vibration stretching mode

of N-H; at 1730 cm-1 non-hydrogen bonded urethane C=O stretch, at 1535 cm-1 corresponding to the stretching mode of urethane N-H bonding + C-N; and at 1222 cm-1 the peak is attributed to C-N stretching. Freeze-drying of a polymer solution is a process in which the solvent is removed by sublimation from the frozen material so that it leaves a porous structure. The density of the resulting porous polymer was determined by concentration of the polymer in the solution; and the morphology of the foam is determined by phase separation. Phase separation can be divided into liquid-liquid phase separation (which may occur prior to freezing of the solvent) and liquid-solid phase separation (which occurs when the solvent freezes). Adding a co-solvent or non-solvent to the solution may induce liquid-liquid phase separation19. Figure 3 shows SEM pictures of different morphologies of a 35% (w/v) (polymer/solvent) PU prepared in pure dioxane and quenched at different temperatures. When the polymer solution was quenched at ­78°C, a microcellular pore structure was formed, suggesting that the quenched state of the polymer solution was located within the unstable region, see Figure 1. The cell sizes in the scaffolds increased at ­25°C. The microstructure was bead-like and ladder-like, respectively. The anisotropic morphologies were likely developed by a solid-liquid phase separation, in which preferential crystallization of pure dioxane predominantly occurred in the direction of heat transfer. The morphology of polyurethane 35% (w/v) prepared in varying solvent volume ratios of dioxane and n-hexane as co-solvent is shown in Figure 4 and 5. We can see that increases in the ratio of n-hexane and the quenching temperature, results in an increase of the spaces,. The nhexane improves a solid ­liquid phase separation and the morphologies obtained were a bead-string like structure. The formation of this structure suggests that a nucleation and growth mechanism proceeded at this polymer concentration as a major TIPS pathway.

(a)

(b)

Figure 3. SEM images of the cross section of polyurethane produced by quenching 35%(w/v) polymer solution in pure dioxane at ­78°C(a) and ­25°C. Special Edition No. 1, 2006 7

(a) (b) Figure 4. SEM images of cross section of polyurethanes as a function of the quenching temperature prepared by 35%(w/v) polymer in dioxane/c-hexane (86:14 v/v) quenched at a) ­78°C; b) ­15°C. In order to improve liquid-liquid phase separation, water was added as non-solvent in different ratios. Figure 6 shows SEM pictures of different morphologies of 35% (w/ v) PU prepared in varying volume ratios of dioxane and water at different quenched temperatures. In general, the morphologies showed an open cellular microporous structure. The increasing amount of water in the solvent mixture tended to generate large cellular pore sizes. This was probably caused by the fact that as the non-solvent volume fraction increased, a weaker polymer-diluent interaction might induce the formation of polymer poor phase with greater droplet domains. This can be explained by the above mentioned coarsening effect, in which a phase-separated polymer-poor (solventrich) droplets coalesce to form larger droplet domains at the later stage of phase separation. Since the coarsening effect reflects a kinetic behaviour towards the thermodynamic minimization of interfacial energy, one can manipulate the cellular structure by arresting the phase separation process. In this study, the coarsening effect decreased with the decrease in the quenching temperature. It is worth to noting that even though the solid CO2/ acetone quenched scaffolds exhibited a microcellular morphology, their closed cell structures were somewhat different from the typical bicontinuous morphology produced by a spinodal decomposition mechanism, as shown in Figure 1. As the polymer solution was incorporated into the solid CO2/acetone mixture, it is conceivable that an instantaneous heat gradient, although presumably of short duration, developed in the direction from the contacting mold surface to the core region of the quenched polymer sample. This would create different phase-separated morphologies from the centre towards the surface in the resultant scaffolds by exerting locally different phase-

(a)

(b) Figure 5. SEM images of cross section of polyurethanes as a function of quenched temperature prepared by 35%(w/ v) polymer in dioxanec-hexane (70:30 v/v).quenched at a) ­25°C; b) ­15°C. 8 Journal of Advanced Materials

(a)

(b)

(c)

(d)

(e) Figure 6. SEM images or the cross section of PU scaffolds as a function of quenched temperature and solvent/ nonsolvent volume ratio. The scaffolds were prepared by quenching a 35%(w/v) polymer solution under the following conditions: -78°C(a,d); -25 °C (b,e); -15°C(c,f). The volume ratios of dioxane and water were 84/16 (a,b,c); 93/07 (d,e,f).

Special Edition No. 1, 2006

9

Figure 7. (a) SEM picture of TEOS treated polyurethane after immersion in 1.5 SBF solution for 2336 h. arresting time scales on the spinodal decomposition. Thus, until the final phase arresting occurred, the coarsening effect was likely to play a critical role in the core region and the surface. This effect was expected to be more pronounced in a large dimensional sample. From this reasoning, the prepared scaffolds developed microporous surface skin structure, whereas in the core region macroporous structure was developed. In order to have an indicator of the PU biocompatibility; the samples were treated with tetraethoxysilane and then were immersed in 1.5SBF. A SEM picture of TEOS treated polyurethane after immersion in SBF during 336 h is shown in Figure 7. It can be form apatite nuclei onto functionalized polyurethane after 24 h. after the nucleation of sufficient number of apatite nuclei the surface is covered by calcium phosphate particles accompanied by a secondary nucleation over the initial layer. The Ca-P layer

formed on a polymer's surface is composed of spherulites with very fine crystallites, suggesting a high Ca,P nucleation rate. Some spherulites were formed directly on the surface of other growing spherulites or their interface. This type of occurrence suggests that the front of a growing layer is also preferential nucleation site for other spherulites. Figure 8 shows the set of EDAX patterns of the SBF-treated samples indicate the presence of some sodium and chlorine, from the SBF solution. Magnesium and potassium were also detected. These elements especially magnesium, could instead, be incorporated in the apatite structure and replace calcium. The silicon peak disappears as the time of immersion increases; it was almost fully dissolved due to mechanism based on the hydration, formation of silanol groups and consequent nucleation of the Ca-P particles. The Ca/P value of the coatings were determined by ICP and EDAX and was closely 1.6 after 336 h, less than 1.67 Ca/P value of pure hydroxyapatite due to the incorporation of other iones. The formation of apatite in the polymer surface is an indicator of its bioactivity.

Conclusions

Highly porous poly(urethanes) can be fabricated by using a thermally induced phase technique. It is possible to control porosity and the morphology by varying the polymer concentration, quenching temperature and the co-solvent utilized. The better structure according to the mentioned characteristics was obtained using water as a non solvent in the ratio of 86:14 dioxane/water and quenched temperatures of ­25 and ­15°C. These biodegradable scaffolds materials coated with bonelike apatite have a great potential as bone ­ repairing materials, due to high bioactivity favourable to bond to bone chemically and to allow the growth of connective tissue through the porous of the material.

(a)

(b)

Figure 8. EDAX patterns of TEOS treated samples immersed in 1.5SBF after: (a) 24 h immersion; (b) 336 h. immersion. 10 Journal of Advanced Materials

References

1. C. Chaput, E.A. DesRosiers, M. Assad, M. Brochu, L.H. Yahia, A. Selmani, C.-H. Rivard, Advances in Materials Science and Implant Orthopaedic Surgery , Luwer Academic Publishers, Dordrecht, The Netherlands, p. 229,1995. 2. A. G. Mikos, Y. Bao, L.M. Cima, D. Ingber, J.P. Vacanti, R.J. Langer, J. of Biomed. Mat. Res., 27, p. 183, 1993. 3. A.G. Mikos, Sarakinos, G.S.M. Leite, J.P. Vacanti, R. Langer, Biomaterials, 14, p. 323, 1993. 4. G.J. Beumer, C.A. van Blitterwijk, D. Bakker, M. Ponce, Biomaterials, 14, p. 598, 1993. 5. J.H. de Groot, A.J. Nijenhuis, P. Bruin, A.J. Pennings, R.P.H. Veth, J. Klompmaker, and H.W.B. Jansen, Colloid and Polymer Science, 268 n. 12, p. 1073, 1990. 6. K.A. Athanasiou, G.G. Nidederauer, C.M. Agrawal, Biomaterials, 17, p. 93, 1996 7. B. Immirzi, M. Malinconico, G. Orsello, S. Portofino, M.G. Volpe, J. Mater. Sc. 34, p. 1625, 1999. 8. M. Sheth, R.A. Kumar, D. Vipul, R.A. Gross, S.P. Mccarthy, J. Appl. Polym. Sci., 66, p. 1495, 1997. 9. N. Ignjatovíc, S. Tomic, M. Dakic, M. Miljkovic, M. Plavsic, D. Uskokovic, Biomaterials, 20, p. 809, 1999. 10. S. Jin, K.J. Gonsalves, J. Mater. Sc. Mater. Med., 10, p. 363,1999. 11. T. C. Lindholm, T.J. Gao, T.S. Lindholm, Int. J. Oral Maxillofac. Surg., 23, p. 306, 1994. 12. A.C.A. Wan, E. Khor, G.W.J. Hastings, J. Biomed. Mater. Res. (Appl. Biomater), 38, p. 235, 1997. 13. L. Freed, J.C. Marquis, A. Nohria, J. Emmanue, A.G. Mikos, J. Biomed. Mater. Res., 1993, 27:11-23. 14. A.G. Mikos, G. Sarakinos, S.M. Leite, J.P. Vacante, R. Lauger, Biomaterials, 1993; 14: 323-330. 15. K. Whang, C.H. Thomas, K.E. Healy, G. Nuber, Polymer, 1995; 36, 837-842. 16. D.J. Mooney, D.F. Baldwin, N.P. Such, J.P. Vacanti, R. Lauger, Biomaterials, 1996; 17, 1417-1422. 17. L.D. Harris, B.S. Kim, D.J. Mooney, J. Biomed. Mater. Res. 1998; 42, 396-402.

18. H. Lo, S. Kadiyala, E. Guggino, K.W., Leong, J. Biomed. Mater. Res., 1996; 30, 475-484. 19. N. Young Soung, P. Tae Gwan, J. Biomed. Mater. Res., 1999; 47, 8-17. 20. S.W. Song, J.M. Torkelson, Macromolecules, 1994; 27:6389-6397. 21. F. Jun Hua, G. Eun Kim, J. Doo Lee, Y. Keun Son, D. Sung Lee, J. Biomed Mater. Res., 2002; 63, 161-167. 22. C. Schugen, V. Maguet, C. Grandfils, R. Jerome, P. Teyssie, J. Biomed. Mater. Res. 1996, 30 449-461. 23. M. Szycher, A.A. Siciliano, and A.M. Reed, "Polyurhetane Elastomers in Medicine," Polymeric Biomaterials, ed. S. Dumitriu, Marcel Decker, Inc., New York, NY., p. 233, 1994. 24. S. Worley, R. Marchand, C. Lavallée, Biomaterials, 11, p. 97, 1990. 25. T.V. Chirilia, I.J. Constable, G.J. Crawford, S. Vijayasekaran, D.E. Thompson, Y.C. Chen, W.A. Fletcher, B.J. Griffin, Biomaterials, 14, p. 26, 1993. 26. M. Neo, T. Nakamura, T. Yamamuro, C. Ohtsuki, T. Kokubo, and Y. Bando, J. Biomed. Mater. Res., 26, p. 1419, 1992. 27. M. Neo, S. Kotani, Y. Fujita, T. Yamamuro, Y. Bando C. Ohtsuki, and T. Kokubo, "Differences in Ceramic-bone Interface Between Surface-active Ceramics and Resorbable Ceramics: A Study by Scanning and Transmission Electron Microscopy," J. Biomed. Mater. Res., 26, p. 452, 1992. 28. M. Tanahashi, T. Yao, T. Kokubo, T. Miyamoto, M. Minnoda, T. Nakamura, T. Yamamuro, J. Mater. Sc. Mater. Med., 6, p. 319, 1995.

Special Edition No. 1, 2006

11

Synthesizing Research of Nanometer Zinc Oxide by Evaporation Method

Zhai Xiujing, Fu Yan, Bai Bin, and Wang Jie Northeastern University, Shenyang, P.R. China Email:[email protected] Original Manuscript Received 01/10/05; Revised Manuscript Received 08/13/05

Abstract

Nanometer ZnO particles with mean diameters 70-90 nm were synthesized by evaporating of zinc powders of average particle size of 370 mm and were studied by XRD,TEM and electron diffraction. The particles were formed by the oxidation of evaporated zinc vapor in the air. The effect of experimental parameters was examined, the increase of the air flow-rates reduced the average particle size, while increasing the evaporation temperature and the amount of metal charged increased the average particle size. It was found that the crystal habits of particles had a tetrapod-like or wurtrite structure consisting of four needle crystals by TEM and all particles were single crystal by electron diffraction.

Introduction

Recent progress in the preparation and characterization of materials on the nanometer size scale has introduced a new point of view for physics and chemistry in the world of reduced dimensions with grain sizes which is from 1nm to 100nm because of the expectation of their unique properties1,2. ZnO powders are very important materials because of interesting semiconductive properties and many technological applications in science and technology, such as based varistors, dielectric, and other functional devices and also could be used as enforcement phase, wear resistant phase and anti-sliding phase in composites in consequence of its high elastic modulus and strength3,6. It has been reported that ZnO particles formed by the oxidation of Zn vapor typically showed two kinds of shapes: granular and tetrapod-like. The crystal structure of ZnO was of the wurtzite type (hcp) (a0=0.3249nm, c0=0.5205nm) at ordinary temperatures and under atmospheric pressure7,8. In the previous works, the preparation of nanometer ZnO by evaporating the metallic component had been achieved in the presence of inert gas (argon or helium) so that a nanometer-sized was obtained9. The loose metallic powder was subsequently oxidized by introducing air or oxygen into the chamber at different annealing temperatures of zinc metal10. In the present study, Zn powder is evaporated directly using an electric furnace as the evaporation source with air flow-rate pressure. This paper describes the preparation method and presents results of the formation behavior of nanometer ZnO at different evaporation conditions. In addition, the crystal structure and morphology of nanometer ZnO powders were discussed. Evaporating method was perhaps the cleanest of all the nanoceramic synthesis route in a well-controlled atmosphere within a work chamber. On the other hand, the need to evaporate in a low-pressure environment translated directly to work chamber11,12. 12

Experimental

The metal zinc powders of mean grain size 370 mm and purity 99.99% used in this study. The evaporation of Zn was carried out in a large stainless steel work chamber (300 mm in diameter, 500 mm in height) by well-controlled evaporation from an electric furnace. The temperature sensor was a Pt-PtRh thermocouple connected to the perature-regulating unit of a voltage supply maximum 220 V. Air flow-rate was determined by monitoring the average pressure in the apparatus during particle production. When the heating current was switched on, the temperature rose up to certain value, the pressure was supplied from the bottom and then the Zn powders were inserted into an alumina crucible, which was supported inside an outer alumina tube. The temperature transferred into inner crucible via radiation. After five minutes, the vapor rose from the crucible and then condensed on the inner wall and on the lid of the stainless steel work chamber generating nanometer ZnO particles. Water was used to cool inner wall of work chamber to make deposition easy. The time from putting the zinc powder to the completion of the formation of vapor was less than 15 minutes. Nanometer samples were first examined by X-ray diffraction using monochromatic Cu Ka radiation. The shape and the detailed study of morphology of nanometer ZnO was observed by transmission electron microscopy (TEM) of a Philips type EM400T working with electron diffraction system. The average size of nanoparticles ZnO was calculated by the equation Which can be described by: D=kl/Bcosq, Which k=0.9; l­the wavelength of Cu Ka radiation, 0.154060nm; q­the angle of the strongest intensity in XRD patter; B--half-height width of the strongest intensity in XRD patters. Journal of Advanced Materials

Results and Discussion

As soon as the zinc powder was heated the smoke of the particles was observed to start from a vent of the lid of the work chamber. The color of the smoke is white and did not vary with the evaporation conditions; this means that the color of the condensed particles remained the same white, the normal color of nanometer ZnO particles did not depend upon the size of the particles. It was found that the particle size depended on the air pressure, evaporation temperature, and amount of metal charged. The average particle size could be controlled by adjusting these factors. The particles range in the mean size from 70 to 90 nm. The different sizes mean that the particles have formed in various parts and followed different trajectories before final collection. Moreover, the difference of mean particle size is most likely due to increased collisions between particles. It was showen that the particles size was proportional to the source temperature in Figure 1a. At high temperature, the time of stationary evaporation almost vanished and the density of particles was larger with growth layers, so larger particles were formed. The size of particles was strongly influenced by the air flow-rate pressure. In Figure 1b the particle sizes of ZnO decreased almost linearly with increasing air flow-rate pressure. Figure 1c definitely showed that the mean size of ZnO particles plotted against amount of metal charged at which they were prepared at 1100° and air flow rate 0.6m3/h. The size of particles increased with the amount of metal charged. The mean size of ZnO particles became about 92.7 nm or grain size 82.6 nm. This fact could be explained that the size depended on the vapor density for the following reason. The surface area of molten metal usually increased with the charged amount and accordingly the evaporation rate increased13,14. Since, however, the microscopic shape of smoke remained almost unchanged independent of the charge amount, the increase in vapor density resulted in the increase in crystal size. The crystalline phase of the collected ZnO particles was determined by X-ray diffraction and the details of preparation have been mentioned. On the basis of Figures 2-4, it seems that there are no significant deference in the XRD patterns between all of the synthesized ZnO powders prepared at various evaporation conditions, no observed peaks of Zn crystallite structure appear and the only observation is due to the nano-ZnO formed or all the peaks could be completely identified as reflections of a single phase of ZnO. The specimens have almost similar microstructures as wurtzite structure but have different grain sizes as determined by XRD. The color of the samples were white, the normal color of nanometer ZnO particles. Basically, the crystal structure is the same in the small (70 nm) or large (109 nm) particles. Nano-ZnO particles prepared by the evaporated zinc powder using air showed 5 peaks in the range of 31-57° for all samples. These peaks are intense and narrower with a good crystalline. Overall 13

(a) amount of metal charged = 5g air flow rate 0.4m3/h

(b) Temperature: 1100°C amount of metal charged = 5g

(c) temperature : 1100°C air flow rate 0.6m3/h Figure 1. Effect of synthesizing conditions by evaporation method.

Special Edition No. 1, 2006

(a) (a)

(b)

(b)

(c) Figure 3. XRD patterns of nano-ZnO by different flow-rate (a) 0.4 m3h-1 (b) 0.5 m3h-1 (c) 0.6 m3h-1; temperature = 1100°C, amount of metal charged = 5g. The resulted grain sizes were (a) 86nm, (b) 70nm, (c) 97nm in diameter.

(c) Figure 2. XRD patterns of nano-ZnO by different temperature (a) 1000°C (b) 1100°C (c) 1200°C; air flow-rate = 0.2 m3h-1 and amount of metal charged = 5g. The grain sizes resulted were (a) 95nm, (b) 80nm, (c ) 88nm.

peak positions width and their relative intensities were similar and ZnO particles did not exhibit change in crystallinity, although they were prepared at different conditions. The interplanar spacing was in a good agreement with the standard diffraction patterns (5-0664) of ZnO and hence the wurtzite ZnO phase is well established. Corresponding to Figure 5, the peak positions and phase structure for all of those samples had similar microstructures but a little different grain sizes was determined by analysis of XRD peak broadening. The difference in grain sizes may ascribe to the samples collected from whole inner wall of the work chamber and due to mixing particles at different locations. All samples were identified as wurtzite structure. The crystalline size of the samples is estimated to be 90, 96 and 97 nm for a, b and c respectively being close to those obtained from TEM observation. As a result, the evaporations of nanoZnO powders are almost the same which illustrates a good crystalline structure belong to ZnO phase. 14

ZnO peaks represent pure nanometer wurtzite ZnO. The reflection peaks used for particle size determination are labeled. Figure 6 shows X-ray diffraction of nano-ZnO particles prepared by oxidizing Zn powder at evaporation temperature of 1100°C and amount of metal charged 3g shows that the stable ZnO phase is the only phase present, and no evidence for other Zn phase. For our samples, all the diffraction peaks are sharper. The strong third peak is the (101) reflection and the (100), (002), and (110) reflections peaks are also more intense and narrower, while the (102) reflection has very little intensity but this weak peak is clearly visible. So, the observed patterns can be unambiguously attributed to the presence of hexagonal wurtzite crystallites. No metallic zinc crystallite structure was detected. This observation provides clear evidence that the single crystals are present and shows a wellknown wurtzite (hexagonal) diffraction pattern. Wurtzite ZnO is a key material in a variety of industrial fields due to its excellent electrical and optical properties. Figure 7a-b are represented XRD patterns of ZnO particles, the intensities of the ZnO peaks (1, 2, 3, 4 and 5) corresponding to the planes (100), (002), (101), (102) and (101) (2q = 31.7o, 34.5o, 36.2o, 47.6o, 56.5o respectively. Journal of Advanced Materials

(a)

(b)

Figure 5. X-ray diffracting patterns for three samples of nano-ZnO particles were synthesized under the same conditions. Evaporation temperature = 1100°, air flow-rate = 0.6 m3h-1 and amount of metal charged = 5g.

(c) Figure 4. XRD patterns of nano-ZnO by different metal charged (a) 5g (b) 10g (c) 15g; temperature = 1100°C, air flow-rate = 0.6 m3h-1. The diameters were (a) 81nm, (b) 96nm, (c) 93nm. The peak spacing planes are showed in Figure 9 (b). It is interesting to note that the positions of the peaks due to ZnO phase. The crystal structure of ZnO is known to be the wurtzite structure in which the oxygen atoms are arranged in a hexagonal closed-packed type (hcp) with lattice with zinc atoms occupying half the tetrahedral sites. The tow types of atoms, Zn and O are tetrahedrally coordinated to each other and are therefore, equivalent position. The zinc structure is thus relatively open with all the octahedral and half the tetrahedral sites empty. According to the following equations, the position of these peaks yields a wurtzite lattice constants of a = 0.3249 nm and c = 0.5205 nm. These lattice parameters are in very good agreement with the lattice reported for the tetrapod-like value that corresponds to a particle diameter of 97 nm. The third peak should give a reasonable indication of crystalline size.

Figure 6. X-ray diffraction pattern of nano-ZnO particles prepared at evaporation temperature 1100°C, air flow-rate = 0.6 m3h-1 and amount of metal charged = 3g.

[3.1]

Figure 7. XRD of nano-ZnO particles was prepared at evaporation temperature = 1300°C, air flow-rate = 0.6 m3h-1 and amount of metal charged = 5g and collected directly on the inner wall of the work chamber. (a) reflection peaks, (b) spacing planes. 15

Special Edition No. 1, 2006

[3.2]

The morphology of the ZnO nanoparticles depends on air pressure in the chamber in the range of 0.2-0.6m3/h; oxygen can reach the middle zone of the vapor where a large amount of embryos exist, and absorbs the oxygen atoms. The oxidation of the surface releases energy that accelerates the growth of ZnO. Protrusion of ZnO continues to grow by coalescing with definite orientations so as to minimize their interface energy. Meanwhile, the center of the nanoparticles is completely oxidized by diffusion and the tetrapod-like ZnO forms in the final stage. Figure 9 showed the structure examined by TEM of nanometer ZnO particles viewed along {1120}. The particles had a tetrapodlike shape known as `fourlings' with four legs of crystalline needles united at a common juncture. Each needle leg extends along the c-axis of the wurtzite structure. Figure 9 shows the selected area diffraction pattern from the tetrapod-like of particle; it consists of strong spots of zinc oxide. From the appearance of the equal thickness fringes in the electron micrographs and the arrangement of the electron diffractions spots relative to the profiles of the particles, it was concluded that the particles had the shape of a tetrapod like. Nanoparticles with a tetrapod -shaped crystal are expected to possess properties having applications in shock-resistance, sound insulation, photosensitization, fluorescence, gas sensitization and catalysis. Throughout the present study the information offered by the shape of the particles has proved to be very useful that the particles have very cut crystal habit and a definite crystal structure.

Figure 9. TEM micrographs of tetrepod-like particles and an electron diffraction pattern. found to exist in nanometer ZnO particles and was consisted of four crystalline needles or tetrapod-like united at a common juncture .Such particles are called "fourlings".

References

1. Z.R. Dai, Z.W. Pan, Z.L. Wang, "Novel Nanostructures of Functional Oxides Synthesized by Thermal Evaporation," Advanced Functional Materials, 13(1), 9, (2003). 2. M.E. Greene, C. Reagan Kinser, D.E. Kramer, L.S.C. Pingree, M.C. Hersam, "Application of Scanning Probe Microscopy to the Characterization and Fabrication of Hybrid Nanomaterials," Microscopy Research and Technique, 64(5-6), 415, (2004). 3. J. Xu, W. Ji, X.B. Wang, H. Shu, Z.X. Shen, S.H. Tang, "Temperature Dependence of the Raman Scattering Spectra of Zn/ZnO Nanoparticles," Journal of Raman Spectroscopy, 29(7), 613, (1998). 4. J.B. Butt, "Progress Toward the a Priori Determination of Catalytic Properties," AIChE Journal, 22(1), 1, (1976). 5. Y.C. Kang, S.B Park, "Preparation of Zinc Oxidedispersed Silver Particles by Spray Pyrolysis of Colloidal Solution," Materials Letters, 40(3), 129, (1999). 6. J. Xu, Q. Pan, Y. Shun, Z. Tian, "Grain Size Control and Gas Sensing Properties of ZnO Gas Sensor," Sensors and Actuators B: Chemical, 76(3), 277, (2003). 7. U. Pal, S.G. Casarrubias, C. Zarate, "Preparation of Ge/ZnO Nanocomposites by Radio Frequency Alternate Sputtering," Solar Energy Materials and Solar Cells, 76(3), 305, (2003).

Conclusions

Nanometer ZnO powders were synthesized directly by the evaporation method which can be well controlled by evaporation from an electric furnace as Zn powder is heated under ordinary laboratory conditions. Most collection times were from 2 to 4 minutes with shorter times to the lid. The metal atoms diffused radically outwards and condensed on the inner wall and on the lid of the work chamber. The particles probably grew by coalescence. The particles were affected by the air flow rate, pressure, amount of metal charged, and evaporation temperature. The air had a dominant effect on the oxidation of ZnO. Larger particles were formed when the pressure was smaller. With increasing evaporation temperature and amount of metal charged, the mean diameter of nanometer ZnO particles was increased. The particles ranged between 70 and 100 mm in size. XRD indicated that the nanoparticles were single crystal of wurtzite structure. TEM and electron diffraction showed that the only wurtzite structure was

16

Journal of Advanced Materials

8. L.F. Dong, Z.L. Cui, Z.K. Zhang, "Gas Sensing Properties of Nano-ZnO Prepared by Arc Plasma Method," Nanostructured Materials, 8(7), 815, (1997). 9. H.J. Fan, F. Fleischer, W. Lee, K. Nielsch, R. Scholz, M. Zacharias, U. Gösele, A. Dadgar, et.al.. "Patterned Growth of Aligned ZnO Nanowire Arrays on Sapphire and GaN Layers," Superlattices and Microstructures, 36(1-3), 95, (2004). 10. K. Yamamoto, K. Nagasawa, T. Ohmori, "Preparation and Characterization of ZnO Nanowires," Physica E, 24(12), 129, (2004). 11. W. Jun, X. Changsheng, B. Zikui, Z. Bailin, H. Kaijin, W. Run, "Preparation of ZnO-glass Varistor from Tetrapod ZnO Nanopowders," Materials Science and Engineering, 95(2), 157, (2002). 12. N. Oleynik, M. Adam, A. Krtschil, J. Bläsing, A. Dadgar, F. Bertram, D. Forster, A. Diez, et.al., "Metalorganic Chemical Vapor Phase Deposition of ZnO with Different O-precursors, Journal of Crystal Growth, 248(1), 1,4 (2003). 13. C.C. Lin, K.H. Liu, S.Y. Chen, "Growth and Characterization of Zn­ZnO Core-shell Polygon Prismatic Nanocrystals on Si," Journal of Crystal Growth, 269(2-4), 425, (2004). 14. S. Nishida, Y. Funabashi, A. Ikai, "Combination of AFM with an Objective-type Total Internal Reflection Fluorescence Microscope (TIRFM) for Nanomanipulation of Single Cells," Ultramicroscopy, 91(1-4), 269, (2002).

Special Edition No. 1, 2006

17

Statistical Approach for the Optimal Deposition of AlN Preferential Orientation Films

Xiao-Hong Xu, Hai-Shun Wu, Fu-Qiang Zhang School of Chemistry and Materials Science, Shanxi Normal University, Linfen, China Original Manuscript Received 06/19/03; Revised Manuscript Received 12/17/03

Abstract

It is common practice to perform a number of deposition experiments by varying the controllable parameters to determine the optimal film growth conditions. The preferential orientation intensity of AlN films strongly depends on several sputtering parameters. Based on a Gauss distribution function, the relationship between preferential orientation intensity of the films and sputtering parameters (such as sputtering pressure, target power and the distance from target to substrate) was simulated. It is obvious that the calculated results are in good agreement with experimental ones. This simulation method is of great significance for further designing of new experimental schemes, finding the optimum sputtering condition of oriented films, or preparing good preferential orientation films.

Introduction

Experiment

AlN thin films were prepared by the direct-current (DC) Aluminum nitride (AlN) piezoelectric films feature a wide band gap, high electrical resistivity, high resistance to reactive magnetron sputtering from an aluminum target of breakdown voltage, high acoustic propagation rate and 99.999% purity in high purity argon (99.995%) and nitrogen low transmission loss, and have a wide application (99.995%) mixture gas. The magnetron cathode was a perspective in the microelectronic devices, especially in water-cooled target with a diameter of 6cm. The distance the surface acoustic wave (SAW) and bulk acoustic wave from target to substrate ranges from 3 to 15cm. In the (BAW) devices1-3. When applying the AlN films to the SAW vacuum chamber, eight substrates have been placed on devices, the structure is required to have the the turntable at one time. The sputtering chamber pressure polycrystalline preferential orientation, because the electro- was reduced to 5×10 -5 Pa before deposition. The mechanical coefficient of the films strongly depends on spontaneous heating by DC plasma and radiation from the crystal orientation4-6. Some researchers reported the the target resulted in a substrate temperature of about oriented structure of AlN thin film was strongly affected by 100°C without a substrate heater. Si (111) substrates were the deposition factors, such as sputtering pressure, the chosen for deposition of AlN films. In addition, X-ray distance from target to substrates, target power, nitrogen diffraction (XRD) was used to characterize the structure of concentration and substrate types7-11. Musil et al reported the AlN films. the orientation films is controlled by plasma chemistry, physics, kinetics12,13. In our previous work9, AlN films Results and Discussions oriented with (002) and (100) plane were prepared Experimental Design successfully on the Si (111) substrates by controlling the When preparing the AlN film using the magnetron sputtering factors. Then, in this paper, the relationship sputtering method, sputtering parameters such as between the oriented intensity AlN films and the sputtering sputtering pressure, target power, the distance from target parameters was simulated, and the optimum parameters to substrate, nitrogen concentration and substrate types for the excellent preferential orientation AlN films were all affect the preferential orientation of AlN films to different found. extents. Among these parameters, sputtering pressure Scientific experiments typically have a twofold purpose: (P ), target power (P ) and the distance from target to r w to determine and quantify the relationship between the substrate (D) are more important than others. In Table 1, values of one and more measurable responses and the shows 3 factors (P , P and D), each were set at 8 different r w values of a set of factors presumed to affect the responses, levels. Table 2 shows the designs of 50 experiments for and to determine the values of the factors that produce the 3 factors. Since there are many possible combinations the optimal values of the responses. Up to now, only the of sputtering factors and their level, it is necessary to find relation model of (002) oriented AlN films has been reported a suitable one for specific application. in several articles 14,15. When applying the AlN film to the SAW Table 1. Sputtering parameters and their levels. devices, the AlN (100) structure is required, thus, the (100) and (002) oriented films were investigated using simulated method in this work. 18 Journal of Advanced Materials

Table 2. Arrangement of parameters and results of calculation

Special Edition No. 1, 2006

19

Figure 1. The relation plot between Pw and F(100).

Relationship Between the Oriented Intensity and Sputtering Parameters

Preferential orientation of AlN films strongly depends on the selection of various sputtering parameters. When preparing the AlN film using the magnetron sputtering method, AlN film can be in the crystalline or amorphous state. The grains in the crystalline film are likely to grow in a directional or non-directional way. Here, Gauss distribution function was introduced in simulating the relationship between oriented intensity and sputtering parameters, and expressed using the following formula: [1]

where, x is the deviation between the random value and the average value, f(x) is probability16. Here, F is used to express "the oriented intensity of AlN films". When "0.75<F<1.00", the films are considered as good oriented

films with (100) or (002); when "0<F<0.75", the films are considered as mixed films or amorphous films. First, the effects of Pw factor on (100) oriented films was investigated, the results show that when Pw ranges from 35 to 50W, the orientation of films with (100) plane are all good, their oriented intensity (F(100)) are all approximately 1. Otherwise, when the power is higher or lower than the above optimal values, the F(100) rapidly become bad, the results are shown in Figure 1. The next step is to study the effects of Pr on F(100). It is obvious that the optimal range of Pr is not isolated, which strongly depends on the selection of D values. The plots of Pr vs F(100) are shown in Figure 2a-b, which corresponds to D=4 and 6 cm, respectively. Finally, we investigated the effects of D on F(100). Figure 3a-b shown the relation between D and F(100). From Figure 3, we can clearly see when Pr=0.5Pa, the optimal values of D range from 5 to 9cm, as Pr=0.3Pa, D range from 7 to 10cm. Note that these factors together affect F(100) value. Using the same method, we investigated how the important three factors influence the oriented intensity of (002) plane (F(002)). In order to shorten the length of article, we don't depict the simulated processes of (002) plane in detail. The simulated graph of the relationship between the oriented intensity of AlN thin films and the three important parameters are shown in Figure 4, in which, Pr values were given in each 0.01Pa, D in 0.1cm, or Pw is fixed at 35-50w. In planar drawing (a), distinct I, II, III and IV represent (002) oriented films, mixed films, (100) oriented films and amorphous films, respectively. This simulated graph is in good agreement with the experimental one (which is our previous work reported in9). From the threedimension draw (b), it is not difficult to find the Pr and D have an important effect for the oriented intensity (F). Suppose that Pw =30W, the simulating results are shown in Figure 5a, in which, F(100)=0.95 higher than 0.75, and F(002)=0.53 lower than 0.75, so the films are oriented with (100) plane. Whereas, if Pw =65W [see Figure 5b], then

Figure 2. The relation plot between Pr and F(100). 20 Journal of Advanced Materials

Figure 3. The relation plot between D and F(100). F(002)=0.93 higher than 0.75; F(100)=0.55 lower than 0.75, therefore the films are oriented with (002) plane. This result denotes the power range for formation of the (002) plane films was greater than that for the (100) plane. In a word, shorter distance D and lower sputtering pressure are conducive to the growth of the AlN (002) films, whose c-axis is perpendicular to the substrate. On the contrary, longer distance D and higher sputtering pressure are advantageous to the growth of the AlN (100) films, whose c-axis is parallel with the substrate. Relatively speaking, the sputtering power for the (002) plane is greater than that for the (100) plane. 31. When Pw, D, and Pr were given the certain values, the films oriented conditions have been determined by using Gauss distribution function. In addition, this simulated method is very important to obtain the optimal sputtering parameters corresponding (100) or (002) plane oriented AlN films, and has an important signification to prepare the good preferential orientation films.

Conclusions

In this paper is studied the simulation of the oriented AlN film is studied with the help of the Gauss distribution function method. It was found that the sputtering pressure, the distance from target to substrate and the target power are very important factors for the crystalline preferential orientation of AlN films. It is obviously that the calculated results are in good agreement with experimental ones. This simulation method is of great significance for further designing of a new experimental scheme, finding the optimum sputtering condition of oriented films, or preparing the good preferential orientation films.

A Comparison of the Results for Experiment and Calculation A comparison of experimental and calculated results listed in Table 2 should be done to guarantee the correctness of the above-simulated process. It is easy to find that the results of calculation are in good agreement with that of experiment, only except for the No. 6, 27 and

Figure 4. The simulative drawing of orientation distribution of AlN films, (a) Planar drawing (b) Three-dimensional drawing. Special Edition No. 1, 2006 21

Figure 5. The simulative three-dimensional drawing of orientation distribution of AlN films, (a) Pw=30w (b) Pw=65w.

Acknowledgment

The authors would like to express thanks to Natural Science Foundation of China (No. 20341005) and Natural Science Foundation of Shanxi Province for their support to this research.

8. A. Mahmood, N. Rakov, M. Xiao, Mater. Lett., 2003, 57:1925-1933 9. X.H. Xu, H.S. Wu, C.J. Zhang, Z.H. Jin, Thin Solid Films 2001; 388: 62-67. 10. X.H. Xu, H.S.Wu, C.J. Zhang, Z.H. Jin, Chin. J. Appl. Chem., 2000; 17: 411-413 (in Chinese) . 11. X.H. Xu, H.S.Wu, C.J. Zhang, Z.H. Jin, Piezoelectricity and Acoustooptics 2000; 22: 256-258. (in Chinese). 12. H. Polakova, M. Kubasek, R. Cerstvy, J. Musil, Surf. Coat. Tech., 2001; 142-144: 201-205 13. J. Musil, J. Vlcek, Mater. Chem. Phys., 1998; 54:116122 14. F.J. Hickernell, R.X. Yue, F.S. Hickernell, IEEE Trans. Ultrason. Ferroelect., Freq. Contr. 1997; 44: 615-621. 15. M. Akiyama, C.N. Xu, K. Nonaka, K. Shobu, T. Watanabe, Thin Solid Films, 1998; 315: 62-65. 16. H.F. Shen, "Probability and Mathematic Statistics," Beijing: High Education Press, 1995. p. 113 (in Chinese).

References

1. D. Liufu, K.C. Kao, J. Vac. Sci. Technol. A 1998; 16: 2360-2365. 2. M.A. Dubois, P. Muralt, Appl. Phys. Lett. 1999; 74: 3032-3034. 3. M.T. Wauk, D.K. Winslow, Appl. Phys. Lett. 1968;13 : 286-288. 4. R.F. Davis, Proc. IEEE 1991; 79: 702-705. 5. R.R. Clemente, B. Aspar, N. Azema, B. Armas, C. Combescure, J. Durand, A. Figueras, J Cryst. Growth 1993; 133: 59-65. 6. M.B. Assouar, O. Elmazria, L.L. Brizoual, P. Alnot, Diamond and Related Mater., 2002; 11:413-417 7. H. Cheng, Y. Sun, P. Hing, Thin Solid Films, 2003; 434: 112-120

22

Journal of Advanced Materials

Preparation of Nickel Aluminate Spinel by Solid State Reaction

Yong-Sheng Han, Jian-Bao Li, Xiao-Shan Ning, Bo Chi State Key Laboratory of New Ceramic and Fine Processing, Department of Materials Science and Engineering, Tsinghua University, Beijing, China Original Manuscript Received 12/27/02; Revised Manuscript Received 06/11/03

Abstract

The nickel aluminate spinel with a BET surface area of 7.437m2/g was prepared by solid state reaction using Ni2O3 and g-Al2O3 as the starting materials. The suitable sintering temperature was determined as 1200°C, because the formation reaction of NiAl2O4 has completed at this temperature and the sample sintered at 1200°C possesses relatively small particles and high surface area.

Introduction

Nickel aluminate spinel is an important material in various fields due to its high thermal stability and specific catalytic properties. It has been used in sensors and heterogeneous catalysis as well as in high temperature fuel cells1-3. Nickel aluminate spinel has also been proposed as a candidate anode for aluminium electrolysis because it exhibits high temperature stability and resistance to alkalis and melting aluminum. It can be prepared by many methods, such as solid state reaction4, sol-gel method1,2, ion exchanged zeolites methods5,6 and microwave-induced method7. The solid state reaction have the advantages of low cost and high productivity. It is applicable in industry. The solid state reaction involves the mixture of metal oxides followed by sintering in air. Some authors have focused and reported their study on the formation of NiAl2O4 by solid state reaction. Some of them prepared NiAl2O4 at high temperature, such as 1637K8 and 1600°C9,10, while others prepared it at low temperature, such as 1100°C11. All of them had prepared nickel aluminate spinel, but the purity and quality (such as particle size) of samples had not discussed. In this paper, the mixture of Ni2O3 and g-Al2O3 are sintered at different temperatures. Then the fired samples are analyzed by XRD, SEM, BET and particle size. And the effect of sintering temperature on the formation and property of NiAl2O4 powder is discussed in detail.

pan) with filtered CuKa radiation of wavelength 0.15418. The voltage and current settings of the diffractometer were 40Kv and 120mA, respectively. The scan angle was from 5° to 90° with a step size of 0.02° and a scan speed of 4°/ min. The quantitative phase analysis was carried out using the peak area of the respective phase in XRD patterns. The microstructure of the fired samples were investigated by a JSM-6301(JEOL, Japan) scanning electron microscope(SEM). And the samples were coated with Au particles. The BET surface areas of samples were measured by a NOVA400 gas sorption analyzer (Quantachrome Corporation, Canada) using nitrogen as the adsorbate. The bath temperature is 77.4K. The degass temperature is 200°C and the degass time is 2h. The particles size of samples were measured by BI-XDC particle sizer (Brookhaven Instruments Corporation, America).

Results and Discussions

As a component of anode, nickel aluminate spinel should possess the properties of high purity and high sinter ability. The impurities unknown in anode would change the electrochemical behavior of cell greatly. And high sinter ability of NiAl2O4 would make it easy to prepare and process a cermet anode which contains nickel aluminate spienl. Hence, two requirements for high quality nickel aluminate powder were proposed in this paper. Firstly, complete reaction is needed and the product phase should be nearly pure NiAl2O4. Secondly, the product should possess small particle size and high surface area, which lead to the high sinter ability of product. Figure 1 showed XRD patterns of samples sintered at different temperatures. When the sample was sintered at 1000°C, the product contains three phase, namely NiAl2O4, Al2O3, NiO. When the sintering temperature increased to 1100°C, the Al2O3 phase disappeared the product is composed of NiAl2O4 and NiO. When the sintering temperature increased to 1200°C, only the NiAl2O4 phase was checked by XRD. The products sintered at 1300°C and 1400°C contained no crystal phase other than NiAl2O4 phase. Therefore, to attain a relatively pure NiAl2O4 powder, the sample should be sintered at 1200°C or above. The Ni2O3 and g-Al2O3 powder were mixed in 1:2 molar 23

Experimental Details

Powders of Ni2O3 (>99%) and g-Al2O3(>99%) with particle sizes of about 10 to 20um were mixed in 1:2 molar ratio. The mixture was ball-milled in a plastic container using ZrO2 balls and ethanol as the mixing medium for about four days. The dried powder was then placed in an alumina crucible and contacted tightly by pressing. Afterwards, the crucible was heated at various temperatures for two hours in air to produce nickel aluminate spinel. The formation reaction can be described as the following: Ni2O3 M 2NiO + 1/2 O2 NiO + Al2O3 M NiAl2O4 The fired samples were checked by powder X-ray diffraction, using a D/max-RB X-ray diffracometer (RIGAKU, JaSpecial Edition No. 1, 2006

Figure 1. XRD partners of products sintered at different temperature. ratio. If they react completely without loss, the mixture should transform to spinel completely. However, the mechanically mixing make it is difficult to get a absolutely uniform mixture. Some Al2O3 particles could not transform simultaneously at high temperature because there is not enough NiO around them. When the temperature increase to above 1100°C, the residual Al2O3 dissolve into NiAl2O4 and form a Al2O3-NiAl2O4 solid solution2, which had been proved by the change of lattice parameter a0. The lattice parameter of NiAl2O4 is 0.8050nm12. When some Al2O3 dissolved into NiAl2O4, the lattice parameter would become small because the alumina polymorph has a lattice parameter a0 =0.7905nm and is known to dissolve in many spinel-type aluminate at a temperature on the order of 1450K13. In our experiments, the product sintered at 1100°C has a lattice parameter of a0=0.8030nm. The change of a0 and the fact that no diffraction lines other than those corresponding to a single-phase spinel were observed, provide strong evidence for solid solution formation between Al2O3 and NiAl2O4. Because a long milling time was carried out in our experiments, the uniformly mixing was under consideration and there should be a few residual Al2O3 dissolving into spinel. So the excess of NiO which resulted from the formation of solid solution is a few and below the measurement of XRD. So we could not observe the trace of NiO above 1200°C. But the product had a green-blue color, which indicates that NiO still existed in products. Figure 2 showed the NiAl2O4 weight percentage in products sintered at different temperatures. These results were calculated from the peaks area of respective phase in XRD patterns. The sample sintered at 1000°C contains 86.0wt% NiAl2O4 while the sample sintered at 1100°C contains 98.4wt% NiAl2O4. When the sintering temperature increases to 1200°C, only nickel aluminate spinel was observed in XRD patterns. From above results, we can conclude that the formation of NiAl2O4 had happened at 1000°C while the reaction completed at 1200°C. Therefore, the sin24

Figure 2. NiAl2O4 content in products sintered at different temperatures. tering temperature should be higher than 1200°C to prepare the nickel aluminate spinel. The BET surface areas of samples sintered at various temperatures were shown in Figure 3. All the samples were measured at the same condition, using the nitrogen absorption method. The sample sintered at 1200°C has a BET surface area of 7.437m2/g. With the increase of sintering temperature, the surface area decreased. The samples sintered at 1500°C has the surface area of only 2.115m2/g which is below one-third of that sintered at 1200°C. Figure 4 showed the particles size (D50) of samples sintered at different temperatures. We can observe from Figure 4 that the D50 increased with the rise of sintering temperature. And the D50 value of sample sintered at 1500°C was about four times larger than that sintered at 1200°C. Figure 5 showed the SEM images of samples sintered at 1200°C (a) and 1500°C (b). The particles in Figure 5a are irregular and loosely contacting while the particles in Figure 5b has grown up with obvious sinter-

Figure 3. BET surface areas of products sintered at different temperature. Journal of Advanced Materials

(a)

Figure 4. D50 values of products sintered at different temperatures. ing characterization. The particle sintered at 1500°C are larger than that sintered at 1200°C, which agrees well with the measurement of D50. The smallest D50 value was found for the sample sintered at 1200°C which also has the highest surface area. To attain high sinter ability, the small particle and high surface area are needed. Therefore, the sample sintered at 1200°C has a high sinter ability and this temperature was suggested in this paper.

(b)

Conclusions

A high quality nickel aluminate spinel was prepared by solid state reaction using Ni2O3 and g-Al2O3 as the starting materials. The effect of sintering temperature on the formation of NiAl2O4 was investigated. The formation of NiAl2O4 had happened at 1000°C and completed at 1200°C. When the sintering temperature changed from 1200°C to 1500°C, the particles size of products increased while their BET surface areas decreased. The sample sintered at 1200°C has the relatively smallest particle size and the highest surface area, which are necessary for high quality powder. Therefore, the sintering temperature of 1200°C was suggested in this paper.

Figure 5. SEM of products sintered at 1200°C (a) and 1500°C (b).

Metal Spinels with High Surface Areas from Zeolite Precursors," Chem. Mater., 13 (2001) 607-612. 6. W. Schmidt and C. Weidenthaler, "A Novel Synthesis Route for High Surface Area Spinels Using Ion Exchanged Zeolites as Precursors," Microporous and Mesoporous Materials, 48 (2001) 89-94. 7. R.D. Peelamedu, R. Roy, D.K. Agrawal, "Micowave-Induced Reaction Sintering of NiAl2O4," Materials Letters 55, (2002) 234-240. 8. J.M. Fernandez Colinas and C.O. Arrean, "Kinetics of Solidstate Spinel Formation: Effect of Cation Coordination Preference," Journal of Solid State Chemistry, 109 (1994) 43-46. 9. R. Subramanian, M. Higuchi, R. Dieckman, "Growth of Nickel Aluminate Single Crystal by the Floating Zone Method," Journal of Crystal Growth, 143 (1994) 311-316. 10. K.D. Becker and J. Backermann, "Kinetics of Order-disorder Process in Spinels," Phase Transitions 55 ,(1995) 181197. 11. P.G. Kotula and C.B. Carter, "Interfacial Control of Reaction Kinetics in Oxides," Physical Review Letters 77, (1996) 33673370. 12. C. Otero Arean, J.S. Diez Vinuela, Journal of Solid State Chemistry, 60 (1985) 1. 13. A. Navrotsky, B.A. Wechsler, K. Geisinger, F. Seifert, Journal of American Ceramic Society, 69 (1986) 418.

References

1. P. Jeevanandam, Yu. Koltypin, A. Gedanken, "Preparation of Nanosized Nickel Aluminate Spinel by A Sonochemical Method," Materials Science and Engineering, B 90 (2002) 125132. 2. C.O. Arean, M.P. Mentruit, A.J.L. Lopez, J.B. Parra, "High Surface Area Nickel Aluminate Spinels Prepared by A Sol-gel Method," Colloids and Surfaces A: Physicochemical and Engineering Aspects, 180 (2001) 253-258. 3. L. Kou and J.R. Selman, "Electrical Conductivity and Chemical Diffusivity of NiAl2O4 Spinel Under Internal Reforming Fuel Cell Conditions," Journal of Applied Electrochemistry, 30 (2000) 1433-1437. 4. F.S. Pettit, E.H. Randklev, E.J. Felten, "Formation of NiAl2O4 by Solid State Reaction," Journal of the American Ceramic Society, 49 (1996) 199-203. 5. W. Schmidt and C. Weidenthaler, "Nanosized Transition

Special Edition No. 1, 2006

25

Use of Cermets and Ceramics Instead of Tungsten Carbide in Saws and Other Brazed Applications

Thomas J. Walz NW Research Institute, Inc. / Carbide Processors, Inc. Tacoma, WA Original Manuscript Received 05/06/02; Revised Manuscript Received 09/23/02

Abstract

Cermets and ceramics have proven advantages over tungsten carbide in many applications. Some of the advantages include longer wear, reduced weight, much lower thermal conductivity and increased productivity while reducing downtown time. The history of cermet and ceramic advantages has been almost entirely in mechanically held (indexable) applications. It has not previously been possible to braze them reliably with the high strengths and low cost many applications require. In the present study, a new technique was developed to braze cermets and ceramics using standard tungsten carbide brazing tools and techniques. Cermet and ceramic parts were treated and assessed for wettability. Then the cermets were used as saw tips in various cutting applications. It was shown that cermets and ceramics could be successfully brazed and that they had definite advantages over tungsten carbide.

Introduction

The value of ceramics has long been recognized in metal machining. They exceed tungsten carbide or coated carbides by cutting cleaner, faster and maintaining the cutting edge much longer than tungsten carbide. They work well in applications where they can be mechanically held. A logical extension is to use the same material in other applications where they must be brazed. However, the very qualities of wear resistance and corrosion resistance that give ceramics their value in cutting also makes them hard to braze. There have been several attempts to make cermet and ceramic tipped saws. Earlier attempts failed as the tips fell off as soon as any stress was applied. In some instances, the tip even fell off the saw while in the packing crate as it was shipped to the customer. In these previous attempts, materials for the saw tips were selected primarily for characteristics that allowed for brazing. Other techniques required the use of expensive "active" braze alloys and brazing methods such as high temperature, high pressure, vacuum brazing or a combination. These techniques proved to be successful only in applications where required tensile strength was 210 Kg / sq.m. (3,000 psi) or under and where the temperature did not exceed 150°C (300°F). "Active" braze alloys do not provide the strength necessary for sawmill, cabinet shop, masonry drills and other saws, drills and tools. In addition, the preferred materials for these applications are not readily brazeable. Ceramics have high temperature properties that make their use desirable in applications such as saws, tools, jet engine exhausts, and other applications where similar conditions exist. The qualities that make ceramics desirable for use in 26

these applications also make them hard to use. The ability to resist wear, heat and corrosion make them very difficult to braze or weld successfully. Currently it is possible to braze ceramics using special alloys in special atmospheres and with elaborate preparation. Preparing ceramics for plating and brazing is well known in industries such as electronics and aerospace. The drawbacks have been the expense in terms of materials, equipment and overall processing costs. As an example, some common processes are based on palladium at $700 US per tr. oz. In addition the epoxies and alloys suitable for brazing ceramics are typically too weak and break down at temperatures too low to make their use practical in many applications. Active metal alloys, catalytic surface treatment, special atmospheres and multi-step processing all add to the cost of the processing and effectively prohibit their use in extremely price sensitive industries. Thus, when ceramics (including cermets) have been used, it is commonly in indexable or mechanically held applications in tools. For example, they can be used in a rotary cutting tool where the cutter inserts are placed in a pocket at the perimeter of the tool and mechanically held in place.

Materials and Methods

Saw blades were selected as the test application for this process. Saw blades are probably the most punishing common application for brazed parts. Cermets were selected since they were available sintered to shape while the only ceramic shapes available were standard machining inserts. The drawback to using cermets has been that they were not readily brazeable with the necessary strength and temperature tolerance. Cermets are now brazeable with Journal of Advanced Materials

palladium catalysts, and brazing in a vacuum furnace. The overall process of the process for preparing the ceramic body is as follows:

Step 1 Cleaning Cleaning is a cathodic electro-cleaning process in alkaline solution as discussed in patent U.S. Patent No. 5,624,626. The Figure 1. Feeds and speeds in sfm Machinery's Handbook. ceramic part is connected to the negative terminal of the rectifier and a stainless steel rod is used as an anode tensile strengths of up to 100,000 psi (7,000 kg/ sq cm) connected to the positive charge. and can tolerate temperatures up to 1400°F (750°C). New developments permit cermets to be brazed with the same Step 2 Surface etching alloys, at the same cost, using the same equipment and The part is then treated to prepare the surface to bond techniques as are used for tungsten carbide. with an intermediate material. Typically this is a chemical It was the object of this research to provide a bath and may or may not use electric current. The surface commercially feasible process for joining ceramic parts is roughened and chemically activated. The part is to a substrate in a viable manner so that the process has immersed in a solution for an etching effect. This process a desirable balance of advantageous features. Important may be accelerated or enhanced by the use of electric features that were considered included high bond strength current, increased temperature or altering chemical at high temperatures, resistance to hostile environments, combinations. corrosive materials, high impact and vibration. In addition the process would desirably be low cost, fast, simple, Step 3 Deposition of cobalt or other metal and readily meet all environmental considerations. The The surface is then plated with a metal or other material processes developed use chemicals which are commonly that will bond to the ceramic as well as forming a layer available, allow for brazing using standard alloys and allow suitable for brazing with standard alloys or welding. for processing in an aqueous solution rather than a high Typically this is also a chemical bath and may or may not temperature (e.g. 2200°F, 1200°C) salt bath. Ceramics use electric current. It is beneficial if non-cyanide nitrogen can be prepared so that high temperature/high strength/ is introduced to create intermediate metallic compounds. low cost alloys (such as silver based alloys) can be used Cyanide compounds may be used but they are easily and effectively, thus allowing their use in applications unnecessary and add significantly to the cost in several such as cutting tools, wear parts and high temperature ways. linings such as in jet engine exhausts. The use of standard tools and techniques also allowed Step 4 for improved bond strength and provided increased impact A post cleaning and passivation of the parts can be resistance using techniques that were already developed desirable. for tungsten carbide. This process uses standard braze Once this process is performed the parts may then be alloys which allows it to compensate for a difference in brazed on saws using standard braze filler metals (such coefficients of expansion using common braze alloys in as AWS Bag-24 and similar) and will work successfully in common forms. sawing and other applications. The process will be described specifically as it applies to the brazing of ceramic (and more specifically cermet) saw tips to a saw blade. However, within the broader scope Results of the process, it will be recognized that the process could Cermet Tipped Tools in Industry TiCN cermets are now being used for brazed saws in also be applicable to other applications. In particular, applications which have more stringent operating plastics, particleboard, sawmills and other industrial environments such as tools (e.g. drills, routers, shapers) applications. In actual practice they can give a 5:1 or 6:1 wear parts and scrapers in high stress application, cotton advantage in tool life over tungsten carbide. They permit gins, ovens, kilns, jet exhausts and other similar feeding more material faster (Figure 1) and can give as much as a 4:1 advantage in throughput measured as parts structures. In order to meet industry requirements, we had to develop per hour (Figure 2 & 3). Some otherwise unbrazeable TiCN saw tips were treated. a process that specifically did not use high temperature baths of molten salt, active braze alloy, gold braze alloy, With cermet tipped saws, the feed rate of the saw could

Special Edition No. 1, 2006 27

lighter, less temperature sensitive, more resistant to various corrosive materials, much more wear resistant, and less thermally conductive. Ceramics also have stability at high temperature that make their use desirable in applications such as saws, tools and jet engine exhausts, and in other applications where similar conditions exist.

Discussion

Cermets are lighter in weight than tungsten carbide. Cermets have a density in the range of 6.4 g/cm^2 compared to 15.8 g/cm^2 for a typical tungsten carbide. This is advantageous in weight sensitive applications such as aerospace. Cermets have a typical thermal conductivity of 15 W/ mxK compared to 63 W/mxK for cobalt matrix tungsten carbide. For applications where both wear and thermal insulation are important such as plastic extrusion machinery, this is a great advantage. Cermets are now available with a transverse rupture strength equivalent to that of some tungsten carbide grades. Two commercially available grades have TRS's of 240,000 and 290,000 while typical C-3 carbide is in a range of 270,000 - 400,000. Tungsten carbide has much higher transverse rupture strength than cermet but this strength falls off more rapidly as the carbide gets hot. At 800° C cermets and carbide can have approximately the same strength. However, cermets stay much cooler so they have additional advantage in terms of tensile strength in high temperature applications. (Figure 4 ) Cermets can have greater wear resistance than tungsten carbide when used in high flow rate situations. For example, cermet tipped saws stay sharper much longer when fed at twice the speed of tungsten carbide. TiCN cermets can be made in the same facilities as tungsten carbide using the same tooling so there is a wide range of sizes and shapes available. TiCN is harder to make and porosity is more likely to be a significant problem; thus it needs to be a manufactured at a plant with stringent quality controls.

Figure 2. Wear resistance carbide vs. cermet. be doubled without increasing the rim speed and they could feed a full load of 30 slats at a time instead of their typical reduced load of 14 slats with carbide. This gave them a four to one improvement on their throughput since they cut about twice the number of slats in a bundle and they cut them in half the previous time (two times as much material in half the previous time = 4:1 throughput increase.) The cermet tipped saws ran four weeks before being replaced whereas the carbide tipped blades usually needed replacement after just two days. Cermet tipped saws are preferred over carbide tipped saws in operations where material is manually fed as in table saws and sliding panel saws. They feed smoother and with much less effort by the operator. This means much less operator fatigue and much better control. Cermets cut cooler than tungsten carbide, which gives them an advantage in metal cutting. They tend to run about 10 to 15 decibels quieter which helps to meet OSHA health standards in a manufacturing environment. In addition the TiCN cermets appear to have fewer health risks associated with elemental cobalt, which is often used as a binder in tungsten carbide saws and tool tips. TiN cermets have been successfully used in masonry drill bits. The steel shaft heated up but the tip stayed cool. The cool tips can be an advantage in cutting or drilling heat sensitive materials such as composites. Cermets (TiCN and others) and ceramics offer many design advantages over tungsten carbide. Cermets are

Figure 3. Mechanical and performance values. 28 Journal of Advanced Materials

(a)

(b) Figure 4. (a) Table-Effects of heat on transverse rupture strength; (b) Graph-Effects of heat on transverse rupture strength.

Brazed Alumina

It is possible to alter the basic techniques to use other materials such as alumina. The problem with alumina is that the saws have to be built using machining inserts. The inserts are first ground to thickness and then brazed onto the saw. Finally they are ground to shape. A tungsten carbide saw can be built for $100. A cermet tipped saw for $200 and an Alumina saw for $5,000 largely due to the huge amount of grinding required. The alumina parts used are machining inserts because they are well made and readily available. They also illustrate the ability of this process to deal with currently available commercial products. The Alumina parts are NTK insert DNMG433 ZhC6, a TiC ­ Al2O3 combination. The left part is an untreated part as purchased from a machine shop supply house. The second part is an untreated part we tried to braze. The braze alloy formed a nice sphere in the middle of the part. There was no part wetting. The third part has been treated with our surface treatment. This little dark spot at the top is where the part was held during treatment. The fourth part shows an insert coated with braze alloy. The lower edge shows wetting so strong that the capillary action overcame the edge of effect barrier. In other words, the alloy did not spill over the edge but migrated through the material surface. The alloy used was Bag-22 AWS silver braze alloy. It was brazed in an ambient atmosphere furnace using Englehard Black Flux.

Figure 5.

Conclusion

There is great promise for cermets and ceramics in brazed applications similar to the benefits seen in mechanically held applications. The brazing technology is well proven. The benefit is achieved through higher output with less downtime due to a decreased need to change worn parts thus improving overall profitability. The research now is to identify the properties of various materials and how those relate to particular applications. Special Edition No. 1, 2006 29

Sol-Gel Synthesis and Characterisation of AluminaStrontium Hexaluminate Composites

K. Vishista and F.D. Gnanam Department of Ceramic Technology, A.C. Tech, Anna University, Chennai, India H. Awaji Department of Environmental and Materials Engineering, Nagoya Institute of Technology, Gokisco-cho, Showa-ku, Nagoya, Japan Original Manuscript Received 05/10/04; Revised Manuscript Received 11/05/04

Abstract

Boehmite sol was prepared by hot water hydrolysis of aluminium iso-propoxide using nitric acid as the catalyst. Strontium nitrate to yield 0 to 20 vol% calcia was added to the boehmite sol. The boehmite with additives was calcined at 500°C for 3 h. The calcined powder was milled at 230 rpm for 6h and particle size was measured using laser particle size analyser. The powder samples were calcined at 1600°C for 3 h and the formation of strontium hexa ­aluminate was determined using phase diagram, transmission electron microscope, energy dispersive spectra and X-ray diffraction spectra. The powder samples were sintered at 1600°C for 6 h and the formation of hexaluminate (plate-like) grains was confirmed using scanning electron microscopy and optical microscopy.

Introduction

The mechanical properties of alumina ceramics depend on the microstructural properties such as grain size and grain size distribution1. These microstructural properties can be controlled by incorporating a second phase as dispersed particulates or platelets. The influence of grain growth on the alumina matrix is also dependent on the amount and nature of the additives as well as the presence or absence of the liquid phase during sintering2. The addition of metal long fibre or particulate dispersoid as a bonding phase, into ceramics, improves the mechanical properties especially the fracture toughness of the alumina ceramics3. The size of platelike grains in the sintered bodies is important in the manufacturing of toughened ceramics4. Composites with second phase addition such as whiskers, fibres and platelets are difficult to sinter to high density without hot pressing or hot isostatic pressing. The applications are limited because of the expensive processing routes for these composites. Alumina ceramic composites formed through the sol-gel route with particles or whiskers as reinforcement can exhibit fine-grained microstructure, which leads to high mechanical properties, combined with chemical stability5. The in situ formation of second phase (elongate grains) during sintering leads to novel pate-like microstructure. This type of microstructure is mostly observed in materials with anisotropic crystal structure6. Since the crystal structure of alumina is anisotropic, it is possible to form large platelets in alumina by applying this in situ-reaction techniques7. The morphology of sintered alumina varies significantly as a function of cation impurities such as magnesia, calcia, silica, soda, baria and strontia8. The addition of strontium oxide to alumina in the ratio of 30

1:6 yields strontium hexa-aluminate a compound of formula SrAl 12 O 19 (SA 6 ) and has an hexagonal cell 9,10 . A characteristic feature of SA6 is its strongly anisotropic growth which results in elongated grains and platelets11. Maschio et al. (1999)12 reported similar results when SrO was added to a system constituted by alumina-chromia/ zirconia (Al2O3-Cr2O3/ZrO2). The materials based on SA6 are interesting because SA6 forms platelets making the composites tougher13,14. The plate-like morphology of strontium hexa-aluminate increases the mechanical properties of alumina ceramics. Hence this paper deals with the sol-gel synthesis and characterization of insitu alumina-strontium hexaluminate composites.

Experimental Procedure

Boehmite sol was prepared by hot water hydrolysis of aluminium iso-propoxide using nitric acid as the catalyst. Hydrolysis was carried out using double distilled water by stirring for 1.0 h at 80°C. Then 0.3 mol of nitric acid/ mol of alkoxide was added as the peptizing agent. The peptization was carried out with vigorous stirring for 1 h at 80°C. Then strontium nitrate was added. Both hydrolysis and peptization were performed under reflux conditions resulting in no loss of the material. Strontium nitrate to yield 0 ­ 20 vol.-% SrO was added. Boehmite sol with additives was precipitated in ammonia, aged overnight, vacuum filtered, oven dried at 120°C for two days and calcined at 500°C for 3.0 h. The calcined powder was ground in a planetary mill using alumina balls and propanol medium at 230 rpm for 6 h. The particle size analysis was performed using laser particle size analyzer Shimadzu SALD model 1100. The powder samples were calcined at Journal of Advanced Materials

are formed because the aluminate liquids have randomnetwork structures built up of (AlO4) tetrahedra, analogous to the random network structures based on (SiO 4) tetrahedral that are encountered in silicate liquids and glasses.

Figure 1. Phase diagram of Al2O3 - SrO system. 1600°C/ 3 h and TEM studies were performed using JEOL 3000 FX high resolution Transmission Electron Microscope (HRTEM). X-ray powder diffraction spectrum was observed on Philips analytical diffractometer PW 3710 model using CuK radiation. The samples were scanned from 10 to 80° (2) (high angle XRD). The powder was compacted into cylindrical pellets of 20 mm in diameter and 15 mm in height using uni-axial press at 180 MPa ( 54% T.D. was obtained as the green density) and sintered at temperatures ranging from 1400-1600°C for six h in air atmosphere. The pellets were mirror polished using different grades of silicon carbide sheets. The samples were thermally etched, coated with gold using POLORON 500 sputter coating unit and then analyzed using Leica LEO Stereoscan 440 scanning electron microscope (SEM) under secondary electron (SE) mode.

XRD: Phase Determination The samples with different concentrations of strontia sintered at 1600°C for 6 h were ground and the phases were identified by XRD analysis. Figure 2 shows the XRD pattern for different concentrations of alumina/strontia composites. The presence of different strontium aluminates was confirmed by comparing the values of the observed peaks with their standard values as per JCPDS. It is found that with the increase in the SrO content, the strontium hexaluminate phase becomes prominent. Taking into consideration the phase diagram of the SrO-Al2O3 system (Figure 1) the presence of SA6, Al2O3 and SA2 phases were observed when the strontia content was varied from 10 to 20 vol.-%. The presence of SA6 and Al2O3 phases

Results and Discussion

SrO-Al2O3 Phase Diagram The SrO-Al2O3 system shown in Figure 1 ( Ref : Franco Massazza, chim. Ind. (Milan), 41, 114 (1959) fig. No. 294) has five binary phases: 3SrO.Al2O3, SrO.Al2O3, SrO.2Al2O3, 4SrO.Al2O3 and SrO.6Al2O3. Liquidus temperatures drop rapidly upon addition of Al2O3 to SrO. Thus, SrO.6Al2O3 melts incongruently to Al2O3 at 1920°C. The minimum melting compositions are the eutectics between i) SrO.2Al2O3 and SrO.6Al2O3 located at 1780°C.

ii) SrO.Al2O3 and SrO.2Al2O3 located at 1760°C iii) 3SrO.Al2O3 and SrO.Al2O3 located at 1505°C iv) -4SrO.Al2O3 and 3SrO.Al2O3 located at 1630°C It is noteworthy that phase relations in this area can be conveniently studied using the classical quenching techniques because the strontium aluminate liquids quench readily to yield glasses. The non-silicate glasses Special Edition No. 1, 2006

Figure 2. XRD pattern of alumina- 5 vol.-% SrO (b) 10 vol.-% SrO (c) 15 vol.-% SrO and (d) 20 vol.-% SrO. 31

were observed even below 10 vol.-% of strontia content. Okada and Hattori15 have reported the formation of strontium aluminates using the impregnation method. These samples showed the formation of SrAl2O4 upto 1150°C and co-existed with alumina. The formation of or alumina was not observed before the formation of alumina. The formation of SrAl12O19 (magnetoplumbite structure) occurred at 1230°C. The transformation to alumina was observed only at 12501260°C. The above report suggests that the aluminates containing Ca2+, Sr 2+ and Ba 2+ may not act as heterogeneous nucleation sites for Al 2 O 3 formation and are hence classified as additives which have a retarding effect on the phase transition of alumina.

(a)

(b) Figure 3. (a-b) TEM and EDS of Al2O3 - 10 vol.-% SrO composite powders. 32

Transmission Electron Microscope (TEM) and Energy Dispersive Spectra (EDS) Studies The size and location of the strontia particles in the composite powder were analyzed using HRTEM. Figure 3a-b shows the results of the TEM observations and the EDS-spectra of the alumina-10 vol% calcia composite powders calcined at 1600°C for 3 h and Figure 4a-b shows the results of the TEM observations and the EDSspectra of the alumina-15 vol%. strontia composite powders calcined at 1600°C/3h. It can be seen that the particles are in the range of 500 nm. The alumina-strontia composite powders consisted of fine particles and large columnar particles with length > 500 nm. The EDS analysis and the SAD pattern with Debyescherrer rings showed that the fine particles with a uniform size around 100 to 500 nm were alumina. The strontia crystallites are long in shape. The X-ray analysis indicated that the particles consisted of SrO.6Al2O3. This is consistent with the SAD patterns of these particles. The SAD pattern shows that the particles are crystalline in nature and grain growth is accelerated by crystallization. The crystalline nature of alumina and hexaluminate formed are confirmed through TEM. The spectra is taken on the composite powder, focused on the

Journal of Advanced Materials

alumina particle and strontia particles at its surroundings.

Particle Size Analysis A starting powder of high chemical purity, with small median particle size, narrow particle size distribution, controlled pore size and pore size distribution is an essential requisite for obtaining controlled microstructure and improved mechanical properties. The particle size analysis of the powders containing 0, 5, 10, 15 and 20 vol %. strontia was performed using laser diffraction particle size analyzer. Narrow particle size distribution was observed. The average particle size was found to increase with the amount of strontia, when the other processing parameters were unaltered. The results of the particle size analysis for various concentrations of strontia and calcined at 600°C/ 3 h are given in Table 1. From the XRD, the full width at half maximum was measured for each peak and the crystallite size t was calculated from the Scherrer equation16 (X-ray diffraction by Cullity). Table 2 gives the crystallite sizes of the alumina/strontia composite powders calcined at 1600°C for 3 h.

Microstructure

The samples were polished and thermally etched for SEM studies. The plate-like grains are clearly visible. The optical micrograph of alumina - 20vol% strontia composite is shown in Figure 5. The optical micrographs of alumina ­ 5 to 20 vol% strontia composites at lower magnification [Figures 6(a-d)] show that the composites have increasing tendency of forming platelike grains with increasing strontia additions. The aspect ratio of the grains increases with the concentration of SrO. This suggests that strontium hexaaluminate (SrO.6Al2O3) is formed in situ during sintering, as expected from the phase equilibria of SrO-Al 2O 3 system. The fracture surface of alumina -20 vol%. strontia composites viewed under SEM is shown in Figure 7a-b. The microstructure shows that the strontium aluminate grains are platelets. The SrAl12O19 platelets, are 33

(a)

(b) Figure 4. (a-b) TEM and EDS of Al2O3 - 15 vol.-% SrO composite powders. Special Edition No. 1, 2006

Table 1. Particle size analysis for various concentrations of strontia.

Particle size (mm) Concentration of strontia (vol%) d 10 d 50 d 90

Table 2. Crystallite size using XRD for different concentrations of strontia.

Crystallite size (nm) Vol. % of strontia Al2O3 SrO.6Al2O3

5 0 5 10 15 20 0.96 1 1.27 1.36 1.4 2.68 2.8 2.97 3.03 3.3 4.0 4.32 4.56 4.63 4.71 10 15 20

28.72 36.32 48.57 53.54

27.31 49.27 67.35 72.29

approximately 0.5 µm in thickness and 5 to 10 µm in length and width. The anisotropic grain growth was enhanced as seen in the microstructure with the increase in strontia content. The fine equiaxed microstructure observed for pure alumina was completely replaced by a new set of large faceted grains. The resultant microstructure was fine and uniform as many grains grow simultaneously.

Conclusions

(a) Alumina-strontium hexaluminate composites were synthesized by the sol-gel technique. (b) The powders were characterized using TEM and EDS the presence of alumina and strontium hexaluminate were confirmed. (c) X-ray diffraction confirmed the presence of the strontium hexa-aluminate, strontium di-aluminate, and alumina phases. (d) The microstructural analysis shows that the equiaxed alumina grains were replaced by elongated platelike strontium hexaluminate grains.

References

1. B. Koog and T. Kishi, "Effect of Nano Sized SiC and Micro Sized YAG Dispersion on the Microstructure of Alumina," J. Ceram. Jpn 106 (2)138-143 (1998). 2. T. Sekino, J-H. Yu, Y-H. Choa, J-S. Lee, and K. Niihara "Reduction and Sintering of Alumina/Tungsten Nano Composites," J. Ceram. Jpn 108 (6) 545-547 (2000). 3. C.W. Park and D.Y. Yoon "Effects of SiO2, CaO and MgO Additions on the Abnormal Grain Growth of Alumina," J. Am. Ceram. Soc., 83 (10) 2065-69 (2001). 4. C. Scheu, G. Dehm, and W.D. Kaplan "Microstructure of Alumina Composites Containing Niobium and Niobium Aluminides," J. Am. Ceram. Soc, 83(2) 6397-402 (2000). 5. Q. Yang and T. Troczynski, "Dispersion of Alumina and Silicon Carbide Powders in Alumina Sol," J. Am. Ceram. Soc, 82 (7) 1928-30 (1999). 6. L. An and H.M. Chan, "Control of Calcium Hexaluminate Grain Morphology in In-situ Toughened Ceramic Composites," J. Mater. Sci. 31 3223-29 (1996). 7. W.H. Tzing, W.H. Tuan, "Exaggerated Grain Growth of Fe Doped Alumina," J. Mater. Sci. Lett. 18 1115-1117 (1990). 8. A.P. Goswami, S. Roy, M.K. Mithra, and C.D. Gopes, "Impurity

Figure 5. Optical micrographs of alumina - 20 vol.% strontia composites. 34

Journal of Advanced Materials

Figure 6. (a-d) Optical micrographs of alumina - 5 to 20 vol.% strontia at lower magnification.

Dependent Morphology and Grain Growth in Liquid Phase Sintered Alumina," 84 (7) 1620-26 (2001). 9. P-L. Chen and I-W. Chen, "In Situ Alumina/Aluminate Platelet Composites," J. Am. Ceram. Soc., 75 (9) 2610-12 (1992). 10. E. Suvaci, K.S. Oh, and G.L. Messing "Kinetics of Template Growth in Alumina During the Process of Templated Grain Growth," Acta. Mater., 49 2075-2081 (2001). 11. O. Sbaizero, S. Maschio G. Pezzotti, and I.J. Davies, "Microprobe Fluorescence Spectroscopy Evaluation of Stress Fields Developed Along a Propagating Crackin an Al2O3/ CaO.6Al2O3 Ceramic Composites," J. Mater. Res. 16 (10) 2798-2802 Oct. (2001). 12. S. Maschio and G. Pezzotti "Microstructure Development and Mechanical Properties of Alumina-Hex-aluminate Composites As-sintered and After Aging in Aqueous Physiological Solution," J. Ceram. Soc. Japan 107 (3) 270274 (1999). 13. S. Maschio, E. Lucchuni, V. Seigo, "Piezospectroscopic Analysis of the Residual Stresses in the Strontium Hexaluminaate/Zirconia (SrAl12O19/ZrO2) System," J. Am. Ceram. Soc., 82 (11) 3145-49 (1999). 14. T­W. Sone, J-H. Han, S.H. Hong and D-Y. Kim, "Effect of Impurity on the Microstructure Development during Sintering of Alumina," J. Am. Ceram. Soc., 84 (6) 1386 ­ 88 (2001). 15. K. Okada, A. Hattori, T. Taniguchi, A. Nukui, R. N. Das, "Effect of Divalent Cation Additives on the - Alumina Phase Transition," J. Am. Ceram. Soc., 83 (4) 928-32 (2000). 16. B.D. Cullity, Elements of X-ray Diffraction, 2nd ed. AddisonWesley Publishing Company, Reading, MA, 102 ­ 103 (1978).

Figure 7. (a-b) Fracture surface (SEM) of alumina - 20 vol.% strontia composites. Special Edition No. 1, 2006 35

Effect of the Composition of Rare Earth Elements on the Microstructure and Electrochemical Properties of RE(NiCoMnAl)5 Hydrogen Storage Electrode Alloys

Hongge Pan, Jianxin Ma, Yongfeng Liu, Mingxia Gao, Rui Li, Changpin Chen Department of Materials Science and Engineering, Zhejiang University, Hangzhou, People's Republic of China E-mail: [email protected] Original Manuscript Received 12/25/02; Revised Manuscript Received 10/04/03

Abstract

In this paper, the microstructure and the electrochemical properties of RE(NiCoMnAl)5 hydrogen storage alloys with different composition of rare earth elements have been investigated. The results show that for all samples studied, the crystalline structure is hexagonal of CaCu5 type, the metallographic morphology structure is dendrite and the unitcell volume decreases after La being partially substituted by other rare earth elements. The effect of the composition of rare earth elements on the electrochemical properties of the RE(NiCoMnAl)5 is evident. As compared with La(NiCoMnAl)5 alloy, the electrochemical capacity of RE(NiCoMnAl)5 is decreased, the high-rate dischargeablity is enhanced and the cycling life is prolonged, when La is partially substituted by rare earth elements, Ce, Pr, and Nd. The effect of the composition of rare earth on the electrochemical properties is closely related to the change of the unit cell volume of the RE(NiCoMnAl)5 alloy.

Introduction

Rare earth-based AB5 type alloys are most widely used for commercialized Ni-MH batteries because of their prompt electrochemical activation, fair hydrogen storage capacity (the theoretical discharge capacity of LaNi5 being 372mAh/ g) and good charge and discharge kinetics1. LaNi5 itself can not be used as the negative electrode directly in commercial batteries because the electrochemical capacity of LaNi5 alloy degrades rapidly during charge and discharge cycling due to the rapid pulverization caused by the large volume expansion of alloy (23.5%) during hydriding2) and the subsequent serious corrosion in strong

alkaline electrolyte. As pointed out previously3, the high capacity degradation (over 70% decay after 150 cycles) and the high cost La have made the alloy inappropriate for the battery application. In 1984, Willems et al.4 found that multi-substitution, if properly executed, could effectively improve the charge­discharge cycling stability of AB5 type alloys. Currently, for commercial AB5 type hydrogen storage electrode alloys, inexpensive mischmetals are used in A side to substitution La, and Co together with Mn and Al are used to partially substitute Ni in B side. The effect of partial substitution of Ni with Co, Mn, Al, Si, Ti on the electrochemical properties have been investigated

Table 1. The serial number and the composition of the RE(NiCoMnAl)5 hybride alloy studied in this paper (wt%).

36

Journal of Advanced Materials

systematically. Nowadays, the B side composition of AB5 type ellectrode alloys in commercial batteries after extensively investigations is basic patterned5. Yet the study on the A side elements is few and inadequate. It is known that the A side elements in AB5 type alloys are the hydride forming elements and play the key sites in hydrogen storage6,7. The cost of the AB5 type alloys lowered when cheaper mischmetals (either cerium-rich michmetal Mm or lanthanim-rich michmetal Ml) are used to replace the more expensive La. However, due to the difference in composition of the mischmetals produced from different localities, a systematic study of the effect of the rare earth elements on the electrochemical properties of the alloys is impending. In this paper, the effect of the changes in Aside rare earth compositions, including different compositions of La, Ce, Pr, Nd purposely designed was studied, and different combinations of rare earths produced from different localities on the electrochemical properties of the hydride alloys were investigated.

Experimental Detail

RE(NiCoMnAl)5 hydride alloy with different composition of rare earth were prepared by inductive vacuum magnetic levitation melting of appropriate amounts of pure metals under argon atmosphere. The ingots were turned over and remelted three times to ensure higher homogeneity. The inductively coupled plasma atomic emission spectrometry (ICP-AES) was employed to determind the compsition of the alloys under investigation (see Table 1). Each alloy electrode was prepared by first mixing 0.1 g alloy powder with 0.2 g carbonyl nickel powder to uniformity and then the mixture was cold pressed in a metallic mold under a pressure of 20 MPa to form a pettet with a diameter of 10 mm and thickness of 1.5 mm. The pellet was then kept in a copper holder, which also acted as the current collector. The cell for electrochemical measurement contained a working electrode (the hydrogen storage alloy electrode for study), a sintered Ni(OH)2/NiOOH counter

electrode and a Hg/HgO reference electrode. A Luggin tube was used to reduce the ohmic drop during polarization measurements. The electrolyte employed was a 6 M KOH solution and the testing temperature was controlled at 30±1°C. The discharge capacity and the cycle life of the test electrode were determined by the galvanostatic method. The cut-off voltage for discharge was fixed at ­ 0.6 V vs. the Hg/HgO reference electrode. Each electrode was charged at 60 mA/g for 6h followed by a 10 min break and then discharged with the same current density to the cut-off potential. After the electrodes being completely activated by charging/discharging for 10 cycles, the charge/discharge cycle life studies were initiated by charging at 300 mA/g for 1.2h followed by a 10 min break and then discharged with the same current density to the cut-off potential. For investigating the high rate dischargeability, charging at 60 mA/g for 6h followed by a 10 min break and then the discharge capacities to the same cut-off potential were measured for several large discharge current densities. For metallographic studies, the alloy ingots were polished and etached in a solution of 30%HNO3 and 90%H2O (by volume). The crystal structures were determinded by Xray powder diffraction (XRD) by Philips' X'Per-MPD type X-ray diffractometer using Cu Ka radiation. Voltage=40 kV, Current=25 mA Scan angle=20°-90°Scan rate=0.02°/s.

Results and Discussion

RE(NiCoMnAl)5 Microstructure

From the studies for all the as-casted RE(NiCoMnAl)5 alloys, the metallographic structure is dendrite. There is no obvious effect of the composition of rare earth elements on the microstructure of RE(NiCoMnAl) 5 alloy. The microstructure of the serial number A-3 sample is shown in Figure 1. During solidification, the compositional segregation of alloy constituents, especially for Mn, leads to the inhomogeneity of the solidified alloy8. Moreover, the dendritic structures in definite directions were formed according to the directions of heat transfer in the alloy during solidification in the water-cooled copper crucible (from the center of the cavity to the water cooled wall of the mold). According to Li et al.9, through SRM/EDS observation, in the as-casted alloy, the segregation of Mn was high and pronounced, that of Co was comparatively low, and Al was segregated to the boundary of primary phase. Figure 2 shows the XRD patterns of RE(NiCoMnAl)5 hydrogen storage alloys with different compositions of RE. It can be seen that the main phase of each alloy is of the CaCu5-type hexagonal structure and the segregated phases in the alloy can not be observed. The lattice parameters and the unit-cell volumes of the RE(NiCoMnAl)5 alloys calculated on the basis of the XRD data, were listed in Table 2. The lattice parameters and volume all increase with the increasing content of La in the alloys, as the atomic radius of La is larger than those of Ce, Pr, Nd10. 37

Figure 1. The microstructure of as-casted RE(NiCoMnAl)5 alloy (serial number A-3). Special Edition No. 1, 2006

Figure 2. The XRD patterns of the RE(NiCoMnAl)5 alloys with different compositions of RE.

Electrochemical Characteristics

The discharge curves of the electrodes of four different alloys are shown in the Figure 3. The capacity of the RE(NiCoMnAl)5 electrode decreases with the decreasing La content. The effect of the composition of rare earth on the electrochemical capacity of the alloy is marked. However, the activation characteristics of AB5 type alloy is almost not affected by A-side composition, as the maximum electrochemical capacity of all alloys is reached in two to three charge-discharge cycles. So the electrochemical capacity of La(NiCoMnAl)5 is higher than that of Mm(NiCoMnAl)5. The electrochemical capacities Cmax and the activation cycle numbers are illustrated in Table 3. The size of the unit-cell volume of any RE(NiCoMnAl)5 alloy is closely related with the composition of Re in it. It is well known that the hydrogen absorbed in a hydrogen storage alloy enters and stays in atomic state in the interstitial cavities of crystal lattice of the alloy. So the crystal structure of the alloy pays an important role on the hydrogen storage performance especially the hydrogen storage capacity. Alloys with bigger cell volumes and hence bigger interstitial cavities generally have larger hydrogen storage capacity11. The lattice parameters and cell volumes calculated on the basis of the XRD data all increase with the increase in the content of La. Here the hydrogen storage capacities agree well with the size of interstitial cavities. The high-rate dischargeability (HRD) Table 2. The cell-unit parameters of the RE(NiCoMnAl)5 alloys with the curves of RE(NiCoMnAl)5 alloys with the different compositions of RE. different RE compositions are shown in Figure 4. It can be seen clearly that the A side composition of AB5 type alloy has certain effect on the HRD of the RE(NiCoMnAl)5 alloys. And it will become more obvious when the discharge current density is above 600mA/g. The partial substitution of La by Ce (serial number A2), makes HRD of alloy better than that of La(NiCoMnAl)5 (serial number A-1). But too much Ce and Pr, Nd (serial number A-4) deteriorates its HRD. The electrochemical 38

properties of the RE(NiCoMnAl)5 hydride alloy with the different composition of RE are shown in Table 3. Many investigators are upholding the opinion that dischargeability of an alloy is mainly determined by two factors12, namely the charge-transfer at the surface of the alloy electrode and hydrogen diffusion in the bulk of the alloy. The charge-transfer at the surface of the alloy electrode (the exchange current density Io), governs the dischargeablity at small discharge current densities. While hydrogen diffusion in the bulk of the alloy (the hydrogen diffusion coefficient Do) determines the dischargeablity at large discharge current densities. So it is inferred that the change composition of rare earth elements of AB5-type alloy must have a marked influence on the hydrogen diffusion coefficient Do in the alloy. The mechanism and extent of the effect of composition of rare earth on the hydrogen diffusion coefficient Do will be further investigated. It is interesting to point out that the charge-discharge cycling stability increases with Ce content in the alloy. The discharge capacity retentions after 300 chargedischarge cycling numbers (C300/Cmax) were calculated and listed in Table 3. The cycling life of La(NiCoMnAl)5 is the shortest. This result is consistent with the experiment results of Adzic et al13. Generally speaking, there are two major factors that govern the effect of the composition of rare earth on the electrochemical cycling stability. First is the effect of the unit-cell volume which decreases with the increasing content of Ce in RE(NiCoMnAl)5 alloy as evidenced from the XRD data. It is generally believed that the volume change of an alloy during hydriding is in proportion to the amount of hydrogen absorbed in cycling or the electrochemical capacity. The increase in Ce content generally leads to a smaller lattice cavity and lower hydrogen storage capacity and smaller change of unit cell on hydriding and hence lower pulverization and hence improvement in the cycling life. The second is the formation of a protective surface film. As Ce+3 ion can be oxidized to Ce+4 ion in the alkaline electrolyte and forms a dense and strong CeO2 oxide film on the alloy surface. The dense oxide film rather effectively inhibits further oxidation of the alloy, slows down the corrosion rate, and thereby improves the cyclic stability of the alloy.

Journal of Advanced Materials

the electrochemical capacity of RE(NiCoMnAl)5 decreases, its high-rate dischargeablity improves and its cycling life enhances, when La is partially substituted by others rare earth including Ce, Pr, Nd. The effect of the composition of rare earth on the electrochemical properties is related to the change of the lattice parameters and unit cell volume of the RE(NiCoMnAl)5 alloys.

References

1. T. Sakai, H. Yoshinaga, H. Miyamura, et al., J Alloys Comp, 1992,180:37 Figure 3. The activation characteristics of the RE(NiCoMnAl)5 alloys with the different compositions of RE. 2. The Properties of Application of Metal Hydride [M] Beijing Chemical Industry Press 1990,73 3. T. Sakai, K. Oguro, H. Miyamura, et al., J Less-Common Metals, 1990,161:193 4. J.J. Willems, Philips Joural of Research,1984,19 (1):1 5. H. Ye, H. Zhang, J.X. Cheng, T.S. Huang, J Alloys Comp, 2002,308:163 6. K.E. Ren, et al., Rare Metal Material and Engineering, 2000, 29 (4):235 7. Z. Ma, et al., Rare Metal Material and Engineering, 2000, 29 (4): 255 Figure 4. The high-rate dischargeabilities (HRD) of Re(NiCoMnAl)5 alloys with the different compositions of RE. 8. W. Tang, G. Sun, J Alloys Comp, 1994, 203:195 9. C. Li, W. Wang, J Alloys Comp, 1999,286:270

Conclusion

In this paper, the microstructure and thr electrochemical properties of RE(NiCoMnAl)5 hydrogen storage alloy with the different rare earth compositions have been investigated. The results show that for all the samples studied, the crystalline structure is of CaCu 5 type hexagonal structure, the metallographic structure is dendrite and the unit-cell volume decreases when La is partially substituted by Ce, Pr and Nd. The effect of the composition of rare earth elements on the electrochemical properties of the RE(NiCoMnAl)5 is marked. The maxium discharge capacity of La(NiCoMnAl)5 has the highest electrochemical capacity 329.3mAh/g but the worst highrate dischargeablity and cyling life. Generally speaking, 10. Rare Earth, Japan. New Metal Institute Press,1991 11. A. Precheron-Guegan, C. Lartigue, J.C. Achard, J LessCommon Metals, 1990, 161:193 12. H. Pan, J. Ma, C. Wang, et al., Electrochemical Acta, 1999,44:3977 13. G.D. Adzic, J. Johnson, J.J. Reily, et al., J Electrochem Society, 1995, 142:3429 14. A.J. Davenport, H.S. Isaacs, M.W. Kendig, Corrosion Science,1991,32:653

Table 3. The electrochemical properties of the RE(NiCoMnAl)5 alloys with the different compositions of RE.

Special Edition No. 1, 2006

39

Investigation and Fabrication of Nano-sized SnO2 Powder and Its Gas Sensing Properties

Wei Yinghui, Yao Minqi, Guo Hongli, Hu Lanqing, Hou Lifeng, Xu Bingshe College of Materials science and Engineering of Taiyuan University of Technology, Taiyuan, Shanxi, China Original Manuscript Received 06/10/04; Revised Manuscript Received 08/16/04

Abstract

Nano-sized SnO2 powder can be fabricated by the sol-gel method and analyzed by means of XRD, TEM,etc. The results show that the grain size of SnO2 is less than 100nm, in which the sintering temperature has a great influence. When the sintering temperature is lower than 500°C, the grain size has good heat stability, however when the temperature is higher than 500°C, the grain size of SnO2 will grow due to the increase in temperature. The sensor made of this powder has good gas sensing properties and the response and reversion time are 8s and 20s respectively. The optimum working current is 170mA, while the sensor has high sensitivity to some gases, especially CH4. The environment around us, either global or local, is becoming worse due to chemical pollutants and as a consequence of the ever increasing activity of human beings around the world. In the past, major attention has been paid to typical air pollutants which cause a green-house effect, ozone layer destruction, acid rain and so on. Now, the atmosphere contains a variety of artificially produced chemical components, as well as the apprehension of other effects by pollutants which has increased recently1. Therefore, it is necessary to analyze and control these gases. Semiconductor detecting methods are an economical and effective way. At present, SnO2, a surface-controlled sensor, is the main type of gas sensing materials used in the world. The use of SnO2 is prompted by the finer grain and the higher surface ratio, the more obvious change of conductance and the higher sensitivity. Compared with traditional materials, nano-sized materials are composed of superfine particles and more atoms in surfaces and boundaries, therefore nano-sized materials off on many excellent properties. Nano-sized materials offer effective prospects in sensing materials field2. Using nano-sized powder materials will improve the gas sensing properties of SnO2 semiconductors.

Experiments

Many methods have been employed to fabricate nanosized SnO2 powder, such as chemical vapor deposition, evaporation and sputtering. In this paper, a sol-gel method was used to produce SnO2 powder. Pure SnO2 sols were prepared from SnCl4·5H2O and NH3·H2O. 1M NH3·H2O was dropped into 0.5M SnCl4 solution to form a white precipitate Sn(OH)4, then the white precipitate was placed into a half-penetrating film bag for several minutes to obtain SnO2 sols. The sol was concentrated at 70°C and transformed into gel. After the gel was dried and sintered at various temperatures, SnO2 powder was obtained. The powder produced by this method was analyzed by means of XRD and TEM and some properties could be obtained, such as response and reversion time, working current, relationship between sensitivity and gas concentration, and sensitivity and sintering temperature of the powder.

crystalline particles. With the increase in sintering temperature, the diffraction peaks become more sharper. When the sintering temperature is higher than 500°C, the peaks with respect to plane (220), (200), (002), (310), (301) respectively, except for the four peaks above. The average grain size of powder sintered at various temperatures can be calculated by the Schemer equation, which is as follows: D = Kl/(b·cosq) D - average grain size l- wave length of X-ray b- full-width-at-half-maximum q-the diffraction angle

Results and Discussion

X-ray Diffraction Patterns and TEM Images

The X-ray diffraction patterns of SnO2 powder sintered at various temperatures are shown in Figure 1. The XRD patterns of powder sintered at 300°C is a normal pattern for nano-sized SnO2 , for the four wide diffraction peaks of Figure 1 with respect to plane (110), (101), (211), (112) respectively. The reason why the pattern has the wide peaks is that the powder is composed of some amorphous and 40

Figure 1. X-ray diffraction pattern of SnO2 powder sintered at various temperatures. Journal of Advanced Materials

Lattice Distortion Ratio

On the basis of the equation: (2w)2·cos2q = (4/p2)(l/D)2 + 32<e2>hkl1/2sin2q 2w - full-width-at-half-maximum q - diffraction angle D - average grain size of crystalline particle perpendicular to plane (hkl) <e2>hkl1/2 - lattice distortion ratio of grain perpendicular to plane (hkl) The relationship between average lattice distortion ratio and sintering temperature is shown in Figure 5. The lower the sintering temperature, the finer the particle and the larger is the lattice distortion ratio.

Figure 2. Relationship between grain size and sintering temperature.

Figure 2 shows the relationship between sintering temperature and grain size of SnO2 powder. It can be seen that when the temperature is lower than one point, the grain size has good stability; while higher than this point, the grain size grows with the increase in temperature obviously. The reason for this is that the growing rate of the grains has a close relationship with temperature. The critical temperature is about 500°C. This conclusion can be proven by TEM images and electronic diffraction patterns. It can also be shown from Figure 3 that when the sintering temperature is lower, nano-particle aggregates are so clear that no isolated particle can be distinguished. The electronic diffraction pattern indicates that those aggregates consist of nanometer amorphous and crystalline particles. With the sintering temperature increase, the decrease of particle size occurs. The reason is that the crystallization temperature of SnO2 is about 500°C, at which the transformation of amorphous-to-crystallization takes place. The particle is spherical if the temperature is lower than 600°C; however, if the temperature is higher than 600°C, it has some edges and corners, which shows that SnO2 nano-sized powder has crystallized completely at the higher temperature. This result is consistent with the result of X-ray diffraction patterns and grain size calculated by the Schemer equation. It can also be seen that from TEM images, nano-sized powder less than 100nm can be obtained by controlling temperature. Figure 4 shows the electronic diffraction patterns sintered at 300°C and 600°C for 2h. The former is composed of rings with different diameters, while the latter shows some diffraction spots except for rings, which shows SnO2 Figure 3. TEM images of nano-sized SnO2 sintered at various temnano-sized powder has crystallized completely peratures; (a) 300°C, (b) 400°C, (c) 500°C, (d) 600°, (e) 700°C, (f) 800°C. at this temperature. Special Edition No. 1, 2006 41

Figure 4. Electronic diffraction of powder sintered at 300°C and 600°C.

Gas Sensing Properties

The sensor made of SnO2 nano-sized powder was studied for its gas sensing properties by the Taiyuan Electron Group. The results showed that the optimum working current is 170mA, which is higher than a traditional sensor. The reason is that this sensor is composed of fine crystallites, which makes if resistant to a higher senser. To guarantee adsorption and non-adsorption quickly and to have higher sensitivity, a higher working current is required. The response and reversion time of the sensor are 8s and 20s respectively, which is shorter than for traditional materials.

Relationship Between Gas Sensitivity and Concentration

Figure 6 shows the working situation of sensors in C4H10, CO, CH4 in different concentrations. When the gas concentration is lower, sensitivity increase with the increase in concentration, however, it is not found when gas concentration is higher. Compared with precious documentation, sensitivity of this sensor to CH4 is higher. Figure 5. Relationship between <e2>hkl1/2 and sintering temperature.

Relationship Between Sensitivity and Sintering Temperature

The relationship between sensitivity of the sensor and the sintering temperature is shown in Figure 7. When the temperature is lower, sensitivity of the sensor is higher, with an increase of temperature, it decreases gradually. However, this trend compares favorably with correlation documents13,14.

Conclusions

The conclusions are as follows: (1) Nano-sized SnO2 powder, less than 100nm, can be obtained by the sol-gel method, and particle size can be controlled by adjusting sintering temperature. Results of XRD and TEM analyses show that when sintering temperature is lower than 500°C, the particle size increases with increasing temperature but not clearly, however it increases when temperature is higher than 500°C. The growing rate of crystalline particles shows that it is related closely to temperature. The critical temperature is about 500°C. Lattice distortion exists in the resultant powder. The lower 42 Figure 6. Relationship between sensitivity and gas concentration. sintering temperature, the smaller the particle size, and the higher rate of lattice distortion. (2) The sensor made of this powder has good gas sensing properties, while the response and reversion time are 8s and 20s respectively, and the optimum working current is 170mA. The sensor has high sensitivity to some gases, Journal of Advanced Materials

8. Masaru Yoshinaka, Ken Hirota, Masayuki Ito, Hiroshi Takano, Osamu Yamaguchi, J. Am. Ceram. Soc. [J], 1999, 82 (1): 216-218.

sensitivity

9. Shuichi Arakawa, Kenji Mogi, Ko-ichi Kikuta, Toshinobu Yogo, Shin-ichi Hirano, J. Am. Ceram. Soc. [J], 1999, 82(1): 225-228. 10. Mauro Epifani, J. Am. Ceram. Soc. [J], 2001, 84(1): 48-54. 11. Fang Guojia, Liu Zuli , Hu Yifan,Yao Kailun, Journal of Inorganic [J], 1996, 11(3): 537-541.

temperature

Figure 7. Relationship between sensitivity and sintering temperature.

12. Dong Xiangting, Liu Guixia, Zhang Wei et al., Rare Metal Materials and Engineering, 2000, 29, (3): 197-199. 13. Liu Xingqin, Tao Shanwen, Shen Yusheng,Chinese Journal of Applied Chemistry [J], 1996, 13(1): 65-67. 14. Wang Dao, Ye Qing, Long Jie, Jin Jun, Journal of the Chinese Rare Earth Society [J], 1999,17(1):16-19.

especially to CH4, and the sensitivity of the sensor increases with the increase in gas concentration. When the concentration is higher, the increasing trend is small, and the resultant curve is even. The sensitivty is also influenced by the power size; the finer the size , the lower the sensitivity. Acknowledgements This work was funded by Natural Science Foundation of Shanxi Province of China (20011044). The author would like to thank Mr. Zhao Xingguo for his aid in producing the photomicrographs.

References

1. Noboru, Yamazoe "The State of the Art of Gas Sensing for Environmental Protection," Technial Digest of the Seventh International Meeting on Chemical Sensors, Beijing International Convention Center [M]. International Academic Publishers, Beijing, China, 1998, 7:27-30. 2. Chen Chunhua, Liu Xingqin, Xu Wendong, Shen Yusheng, Bulletin of Transducer Technology [J], 1992, (4):17-21. 3. Yang Shenghong, Zhang Xiaoming, Zhang Tingjie, Wang Keguang, Zhu Yubin, Rear Metal Materials and Engineering [J], 2000, 29 (5):354-356. 4. Junjie Zhu, Miaogao Zhou, Jinzhong Xu, Xuegong Liao, Materials Letters [J], 2001, 47: 25-29. 5. Huang Shizhen, Lin Wei, Chen Wei, Chen Zhiqian, Journal of Transducer Technology [J], 2001, 20(1): 21-22. 6. Jose L. Solis, Anders Hoel, Laszlo B. Kish, Claes G. Granqvist, J. Am. Ceram. Soc. [J], 2001,84 (7):504-508. 7. Cai Qishan, Piezoelectrics & Acoustooptics [J], 1993, 15(5): 1-8.

Special Edition No. 1, 2006

43

Nanometer CeO2 Material: Sol-gel Synthesis with Different Precursors and Strong Ultraviolet Absorption

Bing Yan and Wengang Zhao Department of Chemistry, Tongji University, Shanghai, P. R. China Email: [email protected] Original Manuscript Received 03/08/04; Revised Manuscript Received 08/02/04

Abstract

Sol-gel synthesis of CeO2 with different precursors was reported and especially a novel synthesis technology was achieved by assembling polybasic polymeric hybrid precursors involving rare earth salicylate coordination polymers and organic polymeric template of polyvinyl alcohol or polyacrylamide. The XRD analysis and estimation of average crystalline and SEM measurements indicate that the CeO2 reaches nanometer size dimensions. Particularly modified sol-gel technology by hybrid polymeric precursors is a candidate synthesis technology for rare earth oxides. The strong ultraviolet absorption property of nanometer CeO2 was studied using excitation spectrum. The maximum excitation bands exhibit some shifts with changing of different synthesis sol-gel methods, different doped rare earth ions and different calcination temperature.

Introduction

The fact that optical and electronic properties of nanocrystals and nanocrystalline materials based rare earth oxides differ from those of conventional materials and can be influenced by the particle size has initiated interest in this class of materials during last few years1-5. Much research was focused on the Y2O3 doped with Eu3+ ions because it is the unsurpassed red emitting phosphor with high luminescence quantum efficiency in fluorescence lamps and projection television tubes6-10. CeO2 is an important functional host material of rare earth compounds which has a variety of practical value in such fields as special glass and fine ceramics, solid state luminescence phosphors and catalysis assistant regents of petroleum chemical engineering and environment, etc11-14. Especially the CeO2 material with nanometer size exhibits strong ultraviolet absorption property and has been applied to such fine chemical technology fields as antibiotics and antiseptics, etc because strong ultraviolet possesses high energy. Some works were mainly reported on the preparation of CeO2 by means of soft chemical synthesis at low temperature such as chemical coprecipitation, sol-gel methods and hydrothermal synthesis16-18. A novel in-situ sol-gel technology was achieved to prepare nanometer CeO2 doped with rare earth ions by different precursors (citric acid precursors, stearic acid precursors and hybrid polymeric precursors). The particle sizes and the strong ultraviolet absorption properties were studied in detail.

Synthesis of CeO2

CeO2 was synthesized with citric acid precursor by a sol-gel method according to the ref.16. CeO2 was synthesized with stearic acid precursor by a sol-gel method according to the ref.17. CeO2 by assembling polybasic hybrid precursors were synthesized as follows: some amounts of rare earth nitrate with molar rate 100:1 of Ce(NO3)3 to RE(NO3)3 (RE = Eu, Tb, Sm, Dy) were mixed accompanied with a little Hydrogen peroxide. Salicylic acid and ammonia solutions were added to adjust the pH value to about 7.0. Then polyvinyl alcohol (or polyacrylamide) and urea were added and heated to evaporate until red-brown gel appeared and further fired with a resistance-stove under 600°C for three hours, finally light yellow CeO2 powders were obtained.

Characterization of CeO2

The CeO2 particle was characterized by means of X-ray powder diffraction (XRD, 40 kV and 20 mA, CuKa, Philps PW1710) and scanning electronic microscope (SEM, Philps XL-30). Excitation spectra were determined with Perkin-Elmer LS-55 model fluorophotometer (emission wavelength = 564 nm, scan rate = 1000 nm/s, excitation slit width = 10 nm, emission slit width = 5 nm).

Results and Discussion

X-rays diffraction of the nanometer CeO2 powders was measured. Figure 1 shows the XRD of CeO2 with stearic acid sol-gel method (Figure 1a), citric acid sol-gel method (Figure 1b), and polybasic hybrid polymeric precursor solgel method. All CeO2 show single phase crystal and the same crystal structure for the thermal decomposition. CeO2 belongs to cubic systems, space group OH5 - FM3M. The mean crystallite size was estimated from the full width at half maximum of the diffraction peak by the Scherrer equation10,11: Journal of Advanced Materials

Experimental

Starting Materials

Rare earth oxides (RE2O3 (RE = Eu, Sm, Dy) and CeO2, Tb4O7, purity ³ 99.9 %) were converted to their nitrate by dissolving into concentrated nitric acid. Other reagents were all analytical pure. 44

(a)

(b)

(c)

Figure 1. XRD of nanometer CeO2 by stearic acid sol-gel method (a), citric acid sol-gel method (b), and polybasic hybrid precursors sol-gel method (c). Dhkl = kl/[b(2q)cosq)] [1]

Where b(2q ) is the width of the pure diffraction profile in radians, k is the constant, 0.89, l is the wavelength of the X-rays (0.154056 nm), q is the diffraction angle, and Dhkl is the average diameter of the crystallite. From the estimation, it can be found that CeO2 material synthesized with sol-gel technology of three different are in the range of 20 100 nm size (Table 1). We further selectively measured the scanning electric microscope (SEM) of some nanometer CeO2 samples (as shown in Figure 2). There exist some conglomeration phenomena in SEM diagram so that the large powder particles reach to micrometer dimension. It can be predicted approximately the small CeO2 powder particles without conglomeration are in the size of about 100 nm, which take agreement with the data from the estimation of mean

Figure 2. SEM of nanometer CeO2 by stearic acid solgel method (a), citric acid sol-gel method (b), and hybrid precursors sol-gel method (c). crystallite size by XRD. Table 1 gave the distribution of particle size with different precursors. Rare earth coordination polymers with salicylic acid exhibits infinite chain polymeric structure is used as the precursors to prepare the luminescent species. Organic polymers, polyvinyl alcohol or polyacrylamide were introduced to form the polybasic polymeric hybrid precursor template with nanometer size. Therefore, this preparation technology connects the assembly of hybrid material and synthesis of nanometer material, and can be expected to apply into the synthesis of other luminescent materials based with rare earth oxides.

Table 1. Average crystal particle rate of nanometer CeO2.

Special Edition No. 1, 2006

45

Figure 3. Ultraviolet absorption property of nanometer CeO2 doped by rare earth ions by stearic acid sol-gel method with excitation spectra.

Figure 4. Ultraviolet absorption property of nanometer CeO2 doped by stearic acid sol-gel method rare earth ions with excitation spectra.

Four CeO2 doped with Eu3+, Tb3+, Sm3+ and Dy3+ ions were prepared and the excitation spectra were measured. Figures 3, 4 and 5 showed the excitation spectra of nanometer CeO2: RE3+ (RE = Eu, Tb, Sm, Dy) by citric acid sol-gel method and hybrid precursors sol-gel methods. It can be seen that all these spectra exhibit strong excitation bands in long wavelength ultraviolet region of 300 - 450 nm (exactly 350 - 450 nm). The maximum excitation peak is at about 390 - 400 nm, and a shoulder peak of 410 nm, which suggests that CeO2 has very strong ultraviolet absorption property. Generally under the ideal conditions without any non-radiative energy loss, excitation spectrum may correspond to the absorption feature, and so the absorption property can be characterized with the corresponding excitation spectrum19,20. The excitation spectra are similar for all the nanometer CeO2 doped with different rare earth ions except for some red or blue shifts of the maximum excitation wavelength. We merely draw a conclusion that the doped rare earth ions may cause crystal lattice distortion to some extent and further influence the corresponding absorption property. It can be observed that no energy transfer process takes place between the host CeO2 and the luminescent doped rare earth ions for we do not observe the characteristic luminescence of corresponding doped rare earth ions. Comparing the excitation spectra of CeO2 derived from the different temperature (as shown in Figure 6). It can be obviously found that the excitation intensity of CeO2 increases with the increasing of thermolysis temperature (600 - 800 - 1000°C), suggesting that the ultraviolet absorption intensity of CeO2 increases with the temperature of firing. The excitation spectra of the CeO2 prepared at 800°C show some blue shift compared with that of nanometer CeO2 at 600°C. While the excitation spectra of CeO2 at 1000°C is more complicated than that of 800°C. This can be interpreted that the particle size of CeO2 occurs over 800°C and nanometer size effects cause the changing complexity of corresponding spectra. Blue-shift can be 46

Figure 5. Ultraviolet absorption property of nanometer CeO2 by different temperatures by hybrid precursors with excitation spectra. attributed to be the quantum size effect21. The quantum energy level splits to cause the energy gap to increase. Other hand, the red shift is affected by the second effect indirectly caused by quantum size. Surface and interface effects of nanometer particle cause the surface tension, then the surrounding environment of luminescent nanometer particles changes to cause the energy gap increases21.

Conclusions

Nanometer CeO2 was sol-gel synthesized by assembling different precursors. Especially a template comprised of a cerium coordination polymer with salicylic acid used as the precursor material of CeO2 species for its infinite chain chelated structure and organic polymers (polyvinyl alcohol and polyacrylamide) as both dispersing medium and fuel. The strong ultraviolet absorption properties of nanometer Journal of Advanced Materials

8. R. Schmechel, M. Kennedy, H.V. Seggern, H. Winkler, M. Kolbe, R.A. Fischer, X. M. Li, A. Benker, M. Winterer, and H. Hahn, Journal of Applied Physics, 89, 1679 (2001). 9. G. Wakefield, E. Holland, and P.J. Dobson, Advanced Materials, 13, 1557 (2001). 10. Q. Li, L. Gao, and D.S. Yan, Chemistry of Materials, 11, 533 (1999). 11. A.P. Bartko, L.A. Peyser, R.M. Dickson, A. Mehta, T. Thundat, R. Bhargava, and M.D. Barnes, Chemical Physics Letters, 358, 459 (2002). Figure 6. Ultraviolet absorption property of nanometer CeO2 by different temperatures by citric acid sol-gel method with excitation spectra. 12. J. Mckittrick, C.F. Bacalski, G.A. Hirata, K.M. Hubbard, S.G. Pattillo, K.V. Salazar, and M. Trkula, Journal of the American Ceramic Society, 83, 1241 (2000). 13. T. Yi, G.W. Zhao, W.P. Zhang, and S.D. Xia, Materials Research Bulletin, 32, 501 (1997). CeO2 were studied firstly with excitation spectra to compare with different precursors. The CeO2 exhibits broad excitation band at the long wavelength ultraviolet region of 300 - 450 nm. The maximum excitation wavelength changed little with the different synthesized CeO2 and appeared some shifts. It is also found that different doped rare earth ions (Eu3+, Tb3+, Sm3+, Dy3+ with doping concentration of 1 mol %) can influence the ultraviolet absorption property of nanometer CeO2 for influencing the crystalline structure of CeO2. 14. M.-H. Lee, S.-G. Oh, S.-C. Yi, Journal of Colloid Interface Sciences, 226, 65 (2000). 15. J. Dhanaraj, R. Jagannathan, T.R.N. Kutty, and C.H. Lu, Journal of Physical Chemistry B, 105, 11098 (2001). 16. X .T. Dong, G.Y. Hong, and D.C. Yu, Journal of Material Science and Technology, 13, 113 (1997). 17. X.T. Dong, J.S. Mai, W. Zhang, G.Y. Hong, and J.H. Liu, Materials Science and Engineering, 19, 99 (2001). 18. X.T. Dong, M. Li, W. Zhang, C.X. Liu, and G.Y. Hong, Journal of the Chinese Rare Earth Society, 19, 24 (2001). 19. G. Blasse and B.C. Grabmeter, Luminescent Materials, Springer Verlag, (1994). 20. D.O. Cowan and R.L. Drisko, Elements of Organic Photochemistry, Plenum Press, (1976). 21. G.C. Xu and L.D. Zhang, Nanometer Composite Materials, Chinese Chemical Industry Press, (2001).

Acknowledgements

This research was supported by the Science Foundation of Shanghai Universities for Excellent Youth Scientists.

References

1. A.J. Kenyon, C.E. Chryssou, C.W. Pitt, T.S. Iwayama, D. E. Hole, N. Sharma, and C.J. Humphreys, Journal of Applied Physics, 91, 367 (2002). 2. R.N. Bhargava, Journal of Luminescence, 70, 85 (1996). 3. H. Weller, Angew Chemistry, 105, 43 (1993).

4. A. P. Alivisatos, Science, 271, 933 (1996). 5. A. Konrad, T. Fries, A. Gahn, F. Kummer, U. Herr, R. Tidecks, and K. Samwer, Journal of Applied Physics, 86, 3129 (1999). 6. G. Tessari, M. Bettinelli, A. Speghini, D. Ajo, G. Pozza, L.E. Depero, B. Allieri, and L. Sangaletti, Applied Surface Sciences, 144-145, 686 (1999). 7. W.Y. Jia, Y.Y. Wang, F. Fernandez, X.J. Wang, S.H. Huang, and W.M. Yen, Material Science and Engineering C, 16, 55 (2001). Special Edition No. 1, 2006 47

Characteristics of Y2O3:Eu Phosphor Thin Films by Post-deposition Annealing

Xiaosong Zhang, Kaishun Zou, and Yi Tao, Institute of Material Physics, Tianjin University of Technology Lan Li, Institute of Material Physics, Tianjin University of Technology and Insititute of Information, Hebei University of Industry, Tianjin Zheng Xu, Insititute of Optpelectronics Technology, Beijing Jiaotong University, Beijing E-mail: [email protected], [email protected] Original Manuscript Received 04/13/04; Revised Manuscript Received 07/15/04

Abstract

Y2O3:Eu thin films have been grown on ITO substrates by an electron beam evaporation method with a sintered Y2O3:Eu target, and were annealed under different conditions. The construction, composition, surface morphology and luminescence properties of the Y2O3:Eu thin films were evaluated by X-ray diffraction, X-ray Photoelectron Spectroscope, Scanning Electron Microscope and Photoluminescence spectra. It was found that crystallization was improved, and the disfigurement on the crystal surface was repaired. The luminescent properties of the Y2O3:Eu thin films were enhanced with the increase in annealing temperature.

Introduction

At present, displays, a medium between people and computers, play an important role. Flat panel displays (FPD) have been the focus of the display market. Field emission displays (FED), which have integrated the predominance of CRT and LED, are one of the most promising displays with full color FPD owing to its advantages such as wide view angle, wide temperature range for driving, high picture quality, low power consumption, high response speed, and without a magnetic field and X-ray radiation. Development of new electron sources for the field emission has been diamond films and carbon nano tubes (CNT). Otherwise, FED operates at lower anode voltages and high current densities. Some obstacles that have prevented the success of FED are lack of suitable phosphors operation under high current densities and low voltage excitation with excellent luminescent character. Normally, the low-voltage cathodoluminescent phosphors had a chance to the FEDs, but they don't have some indispensable properties, which include high conductivity, hazard to cathode, chemical and thermal stability1,2. Screens require new luminescent materials and fabrication technology are one of the important segments in FED. The phosphor thin films are treated as one of the important research methods. Y2O3:Eu is still considered to be one of best red oxide phosphors and a promising candidate for FED because of its excellent luminescence efficiency, color purity, and stability3. Yttrium oxide films have been grown mainly using radio-frequency sputtering4,5, sol-gel techniques6 and pulsed laser deposition (PLD)7. At the same time, the thermal treatment not only recrystallizes the phosphor layer, but also activates the dopants. Therefore, postdeposition annealing is one of important methods to enhance the properties of films. 48

In this paper, the preparation of Y2O3:Eu thin films was carried out by electron beam evaporation, which can enhance the display quality and we obtained strong adhesion. Specifically, the quality of Y2O3:Eu thin films was improved by post-deposition annealing. Also the composition, structure, surface morphology and photoluminescence of Y2O3:Eu thin films were researched respectively.

Experiment

The thin films of Y2O3 doped with europium were deposited on an ITO glass substrate by an EVA450 electron beam evaporation system. A diameter of a 3 mm Y2O3:Eu target was employed for evaporation, and the vacuum pressure of the depositing chamber was 10-5 Torr. The substrate temperature during the deposition was 200°C. The thicknesses of the prepared Y2O3:Eu thin films were from 80 nm to 200nm with 2Å/s deposition. Post-deposition annealing of these films was carried out at temperature ranging from 400°C to 600°C in air atmosphere. The time of the annealing was 3h for the samples. The composition of the Y2O3:Eu thin films was detected by a PHI-5300 X-ray photoelectron spectroscope (XPS). The structural property was characterized by a Philips X'Pert MPD X-ray diffraction (XRD). The surface morphology of the thin films was observed by a Cambridge S200 scanning electron microscope (SEM). The luminescence properties of Y2O3:Eu thin films was examined by photoluminescence (PL) spectra on a Hitachi F4010 Fluorescence Spectrophotometer with excitation wave length of 249 nm at room temperature.

Results and Discussion

Structural Properties

The emission of Y2O3:Eu thin films is of interest with the Journal of Advanced Materials

Figure 1. The XRD curves of Y2O3:Eu thin films grow on ITO substrates: (a) as deposited; (b) annealed at 400°C; (c) annealed at 600°C.

crystallization property and surface morphology. The deposition and post-deposition annealing conditions including evaporation speed, substrate temperature, annealing temperature and annealing time are important factors for determining the crystallinity and surface morphology. Figure 1 shows the XRD curves of Y2O3:Eu thin films deposited on ITO glass substrate at the substrate temperature of 200°C, then annealed at several temperatures in air. The thickness of the thin films was 200nm. A strong peak at 2Q =35° shows the diffraction of Y2O3(411), and all films have the cubic structure in different post-deposition annealing conditions. It also can be seen from the figure that the diffraction intensity increased with increasing annealing temperature. Meanwhile, the (222) peak of Y2O3 increased and can be differentiated from noise. Nearly the same tendencies have been obtained at other film thicknesses. Figure 2 shows the full width at half maximum (FWHM) of (411) peak of Y2O3:Eu thin films deposited at 200°C and annealed in air as a function of annealing temperature. It can be seen from the figure that FWHMs reduced with an increase in annealing temperature and attains a minimum at 600°C. Figure 2 also shows that the crystallinity of the Y2O3:Eu thin films are improved by annealing. And at the same time, the crystallite size can be calculated from the formula; Crystallite where Wsize = Wb -Ws, l=0.9

Figure 2. The FWHM and the intensity of peak versus annealing temperature of Y2O3:Eu thin films.

Composition

The result of XPS measurement is shown in Figure 3. It can be seen from the Figure 3d that the Y2O3:Eu thin films are made of Y, Eu, O element. Figure 3a shows that the XPS of O1s and the dominant peak at binding energy (ev) is 532.1ev. Figure 3b shows that the XPS of Y3d and the dominant peak at binding energy (ev) is 157.7ev. Figure 3c shows that the XPS of Eu3d and the dominant peak at binding energy (ev) is 1135.0ev. It also affirms that the elements of Y, Eu, O exist in Y2O3:Eu thin films.

Surface Morphology

where: Wsize-Broadening caused by small crystallites; WbBroadened profile width; Ws-Standard profile width; K-Form factor; l-Wavelength. The crystallite size of Y2O3:Eu thin films as-deposited, annealed at 400°C and annealed at 600°C are 32.55 nm, 41.42 nm and 46.03 nm respectively. It shows that the crystallite size increased with the increase annealing temperature. The surface morphology and roughness of the Y2O3:Eu thin films had a strong effect on the PL response of the films. Figure 4 shows the surface morphology of the Y2O3:Eu thin films on an ITO glass substrate annealed at different conditions. It can be seen from the Figure 4a that the asdeposited Y2O3:Eu thin film is not even and the distance between grains is the largest in samples. Figure 4b shows that the 400°C annealed Y2O3:Eu thin film is rather even and the distance between grains decreased. Figure 4c shows that the 600°C annealed Y2O3:Eu thin film is highly even, granularity of grains is uniform, and the distance between grains decreased. 49

Special Edition No. 1, 2006

Figure 3. XPS of Y2O3:Eu thin film (a) XPS of O1s in Y2O3:Eu thin film; (b) XPS of Y3d in Y2O3:Eu thin film; (c) XPS of Eu3d in Y2O3:Eu thin film; (d) XPS of all element in Y2O3:Eu film.

Photoluminescence

Compared with semiconductor material, Y2O3:Eu is a different kind of luminescent material because of the different role of luminescence. In general, semiconductors belong to the group of luminescent materials with recombination luminescent center. It is the recombination of the electrons in a conduction band and the hole in the valence band that causes the luminescence. On the other hand, rare-earth compounds such as Y2O3:Eu mainly belong to

the group of luminescent materials with the individual luminescent center. Figure 5 illustrates the process of 4f configuration splitting. Eu3+ has six 4f electrons, S=3, L=3, J=L-S=0, 2S+1=7. So the energy level is 7F0, 7F1, ······7F6. The main factor is spin-orbit interaction. With the spinorbit interaction, the degenerate 4f configuration is split into several energy levels such as 5DJ and 7FJ. The photoluminescence spectra of Y2O3:Eu consists of the relative broad 5DJM7FJ emission transition. The magnetic dipole

(a)

(b)

(c)

Figure 4. SEM micrograph of surface 200nm Y2O3:Eu thin film. 50 Journal of Advanced Materials

Figure 5. Luminescence model of Y2O3:Eu.

Figure 6. Photoluminescence spectra of as-deposited and annealing Y2O3:Eu thin film: A as-deposited, B annealing at 400°C, C annealing at 600°C.

D0M7F1 transition has peaks at 593nm, while the electric dipole 5D0M7F2 transition has peaks at 618nm. It can be seen from Figure 6 that the electric dipole 5D0M7F2 transition occupies the prime transition in Y2O3:Eu thin films (Figure 6). It shows that photoluminescence intensity increases with increasing annealing temperature. J.DexpertGhys had discussed the effects of crystal field in8. For the 5 D0 term, J=0, so it cannot be split (only one energy level). While the 7F2 term J=2 and 2J+1=5, it can be split into five energy levels (G1, G2, G3, G4, G5). The strongest peak (about 617nm)in the emission spectrum corresponds to the transition of GMG5. From the results of XRD measurement (Figure 1), the diffraction peak intensity indicated and the peak position differentiates with the standard spectrum. It is possible that the large surface tension of nano-structured thin films leads to the decrease of the lattice parameter9. That affects the crystal field around Eu3+, and in return a small shift phenomenon of emission spectra. At the same time, surface defects, unsaturation bond and the luminescent center concentration quench the luminescence; the luminescent center concentration is the most important factor. The reflectance and transmission of Y2O3:Eu thin films leads to a decrease in the PL intensity.

5

Acknowledgements

This work was supported by Tianjin China Science Foundation award (023602111) and (013615211), Tianjin China Education Committee award (01-20114) and Key Discipline of Education Committee of Tianjin Province.

References

1. A.A. Talin, K.A. Dean, J.E. Jaskie, Solid-State Electronics, 2001, 45:963-976. 2. K. Werner, IEEE Spectrum, 1997, 34: 40-49. 3. J.H. Gwak, S.H. Park, J.E. Jang, J. Vac. Sci. Technol. B: Microelectronics and Nanometer Structures, 2000, 18: 1101-1105. 4. A.F. Jankowski, L.R. Schrawyer, and J.P. Hayes, J. Vac. Sci. Technol. A, 1993, 11:1548. 5. W.M. Crantom, D.M. Spink, R. Stevens, Thin Solid Film, 1993, 226:156. 6. R. Rao, Solid State Commun., 1996, 99:439. 7. J. Greer, J. Vac. Sci. Technol. A, 1995, 13:1175. 8. G. Dexpert, M. Faucher, Physical Review B, 1979, 20(1):10. 9. D. Li, S. Lv, B. Chen, Acta Physica Sinica, 2001, 50(5): 933.

Conclusion

Y2O3:Eu phosphor thin films on ITO substrates were synthesized by the electron beam evaporation method with a sintered Y2O3:Eu target, and were annealed under different conditions. The construction, composition, surface morphology and luminescent properties of the Y2O3:Eu thin films were evaluate X-ray diffraction, X-ray photoelectron spectroscope, scanning electron microscope and photoluminescence spectra. The effect of post-deposition annealing on photoluminescence of these films was also investigated. We concluded that the optimal temperature of post-deposition annealing can increase luminescent properties of Y2O3:Eu phosphor thin films. In our future work, we will integrate the diamond thin films cathode to study the Y2O3:Eu phosphor thin films more intensely. Special Edition No. 1, 2006

51

Low Temperature Polarized-dependent Spectrum Studies of ZnO Film

Q. Cao and X.-Y. Li Department of Physics, Nanjing University of Science & Technology, Nanjing, People's Republic of China E-mail: [email protected]

Abstract

This paper, presents a detailed study of high quality (001) ZnO film, epitaxially grown on sapphire substrates by microwave phase-enhanced molecular beam epitaxy (MBE) method. Infrared polarized optical reflectance (OP) and polarized photoluminescence (PL) were measured at low temperature, which indicated a strong optical anisotropy of ZnO film. The phonon modes and dielectric functions of ZnO film could be obtained by researching OP spectrum with spectroscopic ellipsometry method. The polarized PL of ZnO film was TE polarized and indicated that the ZnO film was almost strain free. The polarized rotation aimed at the c-axis associated with the optical anisotropy could be utilized to demonstrate an optically addressed ultra-fast, ultraviolet light modulator.

Introduction

ZnO is a direct-gap, II-VI compound semiconductor with a wurtzite-type structure. It has a large fundamental band gap of approximately 3.37eV at room temperature, and strong near- band-gap excitonic absorption even at room temperature1. Its large band gap and high excitonic binding energy make ZnO and ZnO-related compounds promising candidates for the development of short-wavelength optical devices, as well as for applications in optics and optoelectronics, such as transparent conducing electrodes for flat panel displays and solar cells2,3 like indium tin oxide (ITO). ZnO with appropriate dopants, such as aluminum, is transparaent in the visible region and electrically conductive. ZnO is available as substrate material for homoepitraxy and promising for the fabrication of wurtzite groupIII-nitride based device heterostructures due to its almost perfect lattice match to GaN4-6. ZnO-related alloys such as Mg xZn 1-xO allow band-gap engineering for extremely short wavelength regions. Eventually, ZnObased heterostructures may compete with group-III-nitridebased optoelectronics7. Fundamental parameters for the design of optoelectronic devices with multiple layers are the free-carrier properties of the individual layers, such as the free-carrier effective mass parameters, their anisotropy, and optical mobility values. Determination of the free-carrier and photo mode parameters of thin films in complex layered structures can be done by infrared (IR) spectroscopic ellipsometry (IRSE), which was demonstrated recently as an excellent technique for the precise measurement of the complex IR dielectric function (DF)8. At IR wavelength the DF is affected by polar phonon modes, free-carrier plasma excitations, and the background transition. For wurzire-type materials, the DF differs for electric field polarization parallel (e||) and perpendicular (e^) to the c-axis. A prerequisite for the determination of the free carrier parameters from the DF is the accurate knowledge of phonon mode and e¥ contributions to e|| and e^. High-quality single crystalline ZnO with a sufficiently low number of free carriers are required for the measurement of these contributions on 52

the DF. The focus of the present work was to study phonon modes, precise DF spectra, and e0 parameters for ZnO samples. Our results were obtained from IRSE measurements. A straightforward and useful method for studying the band structure of a direct band gap semiconductor is through the detection of recombination radiation from optically pumped electrons. Such radiation is referred to as photoluminescence (PL). The PL spectrum is related to the energy dependence and separation of participating bands. If the polarization states of the optical pump beam and PL are considered, additional information regarding the symmetry of valence and conduction band wave functions may also be obtained. This paper shows the polarized optical reflectance (OR) and polarized photoluminescence (PL) on the quality of ZnO films in low temperature. From the polarized OR study we were able to get the complex IR DF and from the polarized PL study we received the information on the ZnO band structure.

Experiment

High quality ZnO films were grown on (001) sapphire substrates by microwave phase- enhanced molecular beam epitaxy (MBE). A detailed description of the growth apparatus and the structural characterization is provided in9. Employing a variable-polarized, rotating- compensator type spectroscopic ellipsometer the IRSE spectra were recorded at low temperature from 300 to 1200 cm-1, with a spectral resolution of 2 cm-1, and at 55° and 70° angle of incidence F. High intensity optical pumping is carried out using the frequency-tripled output (355 nm) of a Nd:YAG laser operating with a repetition rate of 10 Hz and 6 ns pulse length The excitation laser beam is focused by a cylindrical lens to form an excitation strip of 50mm height and of a sufficient length to span the laser cavity. Significant changes in facet reflectivity and some variations in material quality lead to the nonunformity of the edge emission Journal of Advanced Materials

intensity along a typical laser bar. By moving the excitation strip along the laser bar we can locate high intensity regions which emit speckled light. This edge emission is then focused into 150 mm, or 3×0.64m focal length monochromators allowing detection by a charge coupled device (CCD camera) with a choice of high resolution or wide spectral range. The PL signal was collected by the objective, dispersed by a SPEX 1877 0.6 m triple spectrometer, and detected array. A linear polarizer was used to analyze the polarization of the luminescence, and a half-wave retarder was used to ensure that the light entering the spectrometer was always polarized in the same orientation. The substrate misorientation was taken into account in aligning the polarizer. All measurements were carried out at 8K.

broadening parameters. Instead by fitting a parametricized model dielectric function (MDF) to experimental data, simultaneously for all spectral data points, it provides an indirect connection between measured data and physical parameters of interest. Parametric models further prevent wavelength-by- wavelength measurement noise from becoming part of the extracted dielectric functions. Parametric model assumptions greatly reduce the number of free parameters. The contribution of a single polar crystal lattice mode to the infrared dielectric function ej for electric field polarization parallel (j= "||") or perpendicular ("^") to the c-axis can be described by the harmonic oscillator approximation with Lorentzian broadening12. [3]

Results and Discussion

The wurzite ZnO compound has two formula units in the primitive cell. The optical phonons at the G-point of the Brillouin zone belong to the following irreducible representation:10 Gopt = 1A1 + 2B1 + 1E1 + 2E2 [1] where wTO, wLO and g denote the TO and LO mode frequency parameters and the lattice mode broadening parameters respectively. The high-frequency limit parameter e" is related to the static dielectric constant e0 through the Lyddanc-Sachs-Teller relation13. [4]

both A1 and E1 modes are polar and split into transverse (TO) and longitudinal optical (LO) phonons with different frequencies due to the macroscopic electric fields associated with the LO photons. The short range interatomic forces cause anisotropy, and A1 and E1 modes have different frequencies. Because the electrostatic forces dominate the anisotropy in the short range forces, the TO-LO splitting is larger than the A1 - E1 splitting. For the lattice vibrations with A1 and E1 symmetry, the atoms move parallel and perpendicular to the c-axis, respectively. The ellipsometric parameters y and D are related to the complex reflectance ratio11. r = rp/rs = tany exp(iD) [2]

Equations [3] and [4], however, do not account for dispersion within the band gap due to the electric bandto-band transitions at higher photon energies. Dispersion effects must be considered appropriately by fitting a dispersion model, e.g., a Sellmeier expression,14 to the measured index-of-refraction data at different wavelengths l in order related n(l) for l®¥ with e¥. We have observed strong optical anisotropy from the ZnO film. Figure 1 shows the reflectivity R of the ZnO film measured as a function of photon energy for light polarized parallel and perpendicular to the c-axis of ZnO.

where rp and rs are the complex reflection coefficients for light polarized parallel (p) and perpendicular (s) to the plane of incident, respectively. The parameters y and D depend on the photon energy hw, the layer sequence within the sample, each layer's dielectric function e, each layer's thickness d, e of the substrate material, e of the ambient material, and the angle of incidence Fa.11 in principle, the thickness and e of each layer can be determined from a spectroscopic ellipsometry (SE) experiment by comparing calculated data with measured data. Traditionally, pointby-point fits are performed where the dielectric function (DF) values for the materials of interest are extracted from the experimental data for each wavelength, and independent for all other spectra data points. For this procedure, the thickness and DF values of all other sample constituents have to be known. The DF obtained from the point-by-point results needs further comparison with model assumptions in order to obtain values of physically relevant parameters such as photon mode frequencies and Special Edition No. 1, 2006

Figure 1. Polarized optical reflectance spectra of ZnO film. 53

Figure 2. IRSE y (top) and D (bottom) spectra for ZnO film. Figure 2 present the spectra of the ellipsometric parameters at 70° angle of the incidence for the ZnO samples in the spectral range from 300 to 1200 cm-1. The calculated data in Figure 2 was obtained using a threephase model for the ZnO film sample. Figure 3 depicts the real (Re{e}), and imaginary (Im{e}) parts of the best-fit MDF spectra for e|| and e^ together with the imaginary part of the dielectric loss functions (-e-1) for ZnO film. The TO phonons give rise to peaks in Im{e}, while the LO phonons peak in Im{-e-1}. A derivative like structure in y can be seen around 650cm-1 in Figure 2. This feature, undetected so

far, is driven by the subtle anisotropy effect. This effect was explained in12, with an example for sapphire. The derivative-like structure occurs just above the reststrahlen band of the ZnO sample. Both ZnO DF spectra e|| and e^ approach unity at different wave numbers above the reststrahlen band. Because rp (rs) reaches its loss when |e|||~1 (|e^|~1), y undergoes a minimum (maximum). It can be directly read from the y spectra in Figure 2 that ZnO is uniaxial positive. Photoluminescence (PL) provides a versatile tool to study the band structure because it can reveal many sharp lines in the bandedge region at low temperature. Figure 4 is a comparison of two PL spectra of sample for polarizations parallel and perpendicular to the c-axis, respectively. Sharp PL lines between 3.3555 and 3.372 eV are tentatively assigned to a series of bound exciton/complex recombinations. A weak peak at 3.3941 eV and a shoulder at 3.3830 eV in the bottom spectrum is assigned to the emission of upper and lower B excitonic polariton branches in the preceding paragraph. PL peaks at 3.3783, 3.3768, and 3.3755 eV in the bottom spectrum are basically polarized to E^c, and are assigned to emissions of upper and lower A excitonic polariton branches and a bound exciton, respectively, judging from the energy positions. A PL peak at 3.3764 eV in the upper spectrum might be due to the emission of a lower polariton branch of exciton. Observation of this peak is considered to be due to the finite momentum of an electromagnetic wave. The calculated polarization ratio is plotted in Figure 5 as a function of the crystal-field- splitting parameter DCF and the band-gap energy Eg. DCF is a quadratic function of the order parameter h|DCF(h=h2DCF(1))|15, and is related to the band-gap energy by16 Eg(h)=Eg(0) ­2.66DCF(h) [5]

In Figure 5, the bottom (Eg) and the top (DCF) abscissas are scaled according to Equation 4. It should be noted

Figure 3. Spectra for the real [(a), Re{e}], and imaginary part [(b), Im{e}] of the dielectric function, and for the imaginary part of the dielectric loss function [(c), Im{-e-1}] according to the best-fit MDF for e|| and e^ in the reststrahlen band region for ZnO film. 54

Figure 4. Polarized PL spectra of ZnO film. Journal of Advanced Materials

Figure 6. Polarization dependence of laser emission. Figure 5. Measured and calculated polarization ratios of ZnO film. that this scaling is only approximate for the experimental data because there is no exact one-to-one correlation between the two scales due to a slight variation of actual composition of the sample. The experimental R values are consistent with the calculated curve. This agreement is especially remarkable when one considers the fact that for the band-to-band transition without the excitonic effect, the R value would be infinity. Figure 6 shows the high polarization dependence of the emission intensity. It can be seen that the ZnO film is almost strain free. The emission above the threshold is found to be strongly TE polarized (parallel to the growth plane) as would be expected for stimulated emission from the edge of an epitaxial layer. The polarization rotation towards the c-axis associated with the optical anisotropy is utilized to demonstrate an optically addressed ultrafast, ultraviolet light modulator. According to Kanes's selection rule, polarization of the luminescence in edge geometry enables a clear distinction between the QW and QD cases. The heavy hole exciton luminescence in QWs must be completely TE polarized. As opposite to the QW case, a significant contribution of TM emission has been observed, pointing to a significant role of exciton lateral confinment. Such as verticallycoupled InGaAs-GaAs Stranski-Krastanow QD s with large number of stacks17, The most remarkable observation has been done for polarization of edge emission in the case of vertically-coupled QD states 18 . This emission is predominantly TM polarized. This indicates that the heavyhole wavefunction is more extended in the growth direction and has a rippled-cigar shape.

could be obtained by researching OP spectrum with the spectroscopic ellipsometry method. The polarized PL of ZnO film was TE polarized.

References

1. K.H. Hellwege and O. Madelong, Numerical Data and Functional Relationships in Science and Technology, Landolt-Bornstein, New Series, Group III, (Spring, Berlin, 1982), Vol. 17, Part A. 2. C.F. Klingshim, Semiconductor Optics (Springer, Berlin, 1997). 3. Z.C. Zin, I. Hamberg, and C. G.Granqvist. J. Appl. Phys., 64, (1998) 117; Y. Li, G. S. Tompa, S. Liang, C. Gorla, Y. Lu, and J. Doyle, J. Vac. Sci. Technol., 16, 994 (1997). 4. S. Nakanmura, M. Senoh, N. Iwasa, S.I. Nagahama, T. Yamada, T. Matsushita, H. Kiyoku, and Y. Sugimoto, Appl. Phys. Lett., 68, 2105 (1996). 5. T. Deteprohm, H. Amano, K. Hiramatsu, and I. Akasaki, Appl. Phys. Lett. 61, 2688 (1992). 6. T. Deteprohm, H. Amano, K. Hiramatsu, and I. Akasaki, J. Cryst. Growth. 128, 384. (1993) 7. A. Ohtomo, M. Kaswasaki, T. Koida, K. Masubuchi, H. Kainuma, Y. Sakurai, T. Yasuda, and Y. Segawa, Appl. Phys. Lett., 72, 2466 (1998). 8. G. Leibiger, V. Gottschalch, and M. Schubert, Phys. Status Solidi A., 228, 259 (2001). 9. Y.F. Chen, D.M. Bagnall, Z.Q. Zhu. T. Yao, M.Y. Shen, T. Goto, J. Cryst. Growth (to be published). 10. J. Sik, M. Schubert, G. Leibiger, and V. Gottschalch, J. Appl. Phys., 89, 294 (2001). 11. R.M.A. Azzarn and N.M. Bashara, Ellipsometry and Polarized Light, (North-Holland, Amsterdam, 1984). 12. M. Schubert, T.E. Tiwald, and C.M. Herziger, Phys. Rev. B., 61, 8187 (2000). 13. R.H. Lyddane, R.G. Sachs, and E. Teller, Phys. Rev., 59, 673 (1941). 14. M. Born and E. Wolf, Principles of Optics, (Pergarnon, Oxford, 1989). 15. D.B. Laks, S.-H. Wei, and A. Zunger, Phys. Rev. Lett., 69, 3766 (1992). 16. B. Fluegel, Y. Zhang, H.M. Cheong, A. Mascarenhas, J.F. Geisa, J.M. Olson and A. Duda, Phys. Rev. B., 55, 13647 (1997). 17. P. Yu, J. Hvam, N.N. Ledentsov, V.M. Ustinov, D. Bimnerg, Phys. Rev. B. (2000) in press. 18. N.N. Ledentsov, I.L. Krestnikov, et al., Thin Solid Films, 36, 7(2000).

Conclusion

The infrared polarized OP and polarized PL were measured at a low temperature of MBE-grown ZnO film. The phonon modes and dielectric functions of ZnO film Special Edition No. 1, 2006

55

Mode-I and Mixed-Mode I/II Fracture Behaviour of Sintered Alumina at Ambient and Elevated Temperatures

Sweety Kumari, N. Eswara Prasad, and G. Malakondaiah Defence Metallurgical Research Laboratory, PO Kanchanbagh, Hyderabad, India B. Bhaskar and B. Naga Prasad Rao Department of Mechanical Engineering, Osmaina University, Hyderabad, India Email: [email protected] and [email protected]

Abstract

The mode-I and mixed-mode I/II fracture toughness of sintered alumina has been evaluated using SENB (Single edge notched bend) specimens under three point bend loading at both ambient and elevated temperatures. The specimens were subjected to asymmetrical loading conditions to determine the mixed-mode I/II fracture toughness. The fracture resistance of the ceramic, in terms of plane-strain fracture toughness (KIc) and total fracture toughness (Kt) parameters, has been found to decrease with increasing test temperature. However, the nature of fracture is found to be similar (quasi-cleavage) at both ambient and elevated temperatures. The imposition of mode-II components on mode-I loading has resulted in only a marginal decrease in the total fracture toughness (Kt) value. The crack path profiles of the mixed-mode tested specimens reveal that the crack front rotates towards mode-I fracture direction. Thus, designing of the components and structures made of such materials can be done based on mode-I fracture toughness, KIc. The results obtained in the study are discussed in the light of microstructure, crack path profiles and fracture mechanisms.

Introduction

The fracture resistance of materials is traditionally characterized based on the mode-I (tensile) fracture toughness. However, it has been found that in many practical situations the structures are subjected to a combination of two or more of the three general modes of loading; mode-I (tensile), mode-II (in-plane shear) and mode-III (out-of-plane shear). Hence, it is essential to study the fracture behaviour of materials, especially structural materials like alumina under mixed-mode loading. Reported studies1-4 on fracture behaviour of sintered alumina are confined to ambient temperatures. Mixed-mode fracture studies on other ceramic materials too are limited and they too are confined to the ambient temperatures5-12. The present investigation therefore aims at the evaluation of fracture resistance under mode-I and mixed-mode I/II loading at both ambient and elevated temperatures. The mode-I and mixed-mode I/II fracture toughness in the present investigation was evaluated using single edge notched beam (SENB) specimens. The specimens were subjected to asymmetric loading conditions to determine the mixed-mode I/II fracture toughness. The details of the specimen geometry and the methodology adopted to determine the fracture resistance parameters have been obtained from the work of Fett and coworkers1,9,10. The material's mode-I fracture behaviour is studied at ambient and elevated temperatures of 1000°, 1200° and 1300°C. On the other hand, the mixed-mode I/II fracture resistance parameters and mode of fracture under these loading 56 conditions have been evaluated at a chosen high temperature of 1200°C. The results thus obtained have been rationalized in terms of microstructural features, fracture mode and crack path methodologies.

Experimental Details

Materials and Specimens

The material for the present investigations, the sintered alumina, was received in the form of rectangular (120mm x 120mm) blocks of thickness 8mm. A small amount, 3 ­ 4 volume % of magnesia was intentionally added to aid sintering. Cuboidal samples of each side measuring 8mm were prepared from the as- received blocks using diamond tipped cutter on a high speed cutting machine. The specimens were hot mounted and then polished using 200 and then 600 grade emery papers. Finally, they were fine polished with diamond paste. The samples were etched using 20% HF solution at 80°C for metallographic investigations. The optical micrograph in Figure 1 shows the presence of uniform distribution of MgO.Al2O3 mixed oxide phase (appearing as dark phase), which might have resulted due to presence of higher volume fraction of MgO as compared to the conventional extent of MgO used during pressureless sintering of alumina. The alumina grains are too fine to be resolved and also, because of high chemical inert nature of alumina, the grains do not get revealed by the chemical etching employed in the present study. Journal of Advanced Materials

Figure 1. Optical micrograph of sintered alumina showing the uniform distribution of clusters of magnesia particles in the alumina matrix.

The material also exhibits noticeable volume content of porosity. The extent of porosity was estimated from density measurements. The dry weight and the weight in water provide specific weight in the as-received condition and that suspended in water, respectively. On the other hand, the saturated weight was obtained after drying the specimens at high temperatures in vacuum. Such a treatment of the specimens could reveal the internal porosity of the material. All the weights are obtained using a sophisticated weighing machine to an accuracy of 100 micrograms. The study has revealed that the apparent density is 3659 kg/m3 and bulk density is 3654 kg/m3. The extent of porosity is 0.14%.

Mixed-Mode I/II Fracture Toughness The three-point bend specimens containing edge cracks loaded under asymmetric bending arrangements (Figure 2) employed by Fett et al.1 have been used in the present investigation to evaluate the mixed mode fracture toughness. Straight ­ through notches of depth, nearly half the specimen thickness were introduced in test specimens at appropriate location so that 2d/L, the ratio of twice the value of the distance from the notch to the mid-point of the specimen to the total span, is of constant values of 0.4, 0.6 and 0.7. Such notches were essential to evaluate the mixed-mode I/II fracture resistance for want of values of the geometrical factors determined by Fett et al.1. These specimens after the introduction of the notches were marked with both centre line and the end points (of total span) of three-point bend loading. The specimens were loaded in the ramp control on computer controlled Instron 8801 servohydraullic test system at a ramp rate of 0.5 mm/min. The data logging rate is chosen to be 10 data points per second. The load ­ displacement data thus obtained have been analyzed to determine the resolved mode I (KI) and mode II (KII) stress intensity factors and from these values the total stress intensity factor (Kt) has been evaluated. Fractography The fractured test specimens have been examined under stereo-optical microscope in order to record the crack front rotation. Constant magnification of 10X was used to record the crack front profiles. Then, the fracture surfaces were carefully cleaned ultrasonically and gold coated using sputtering technique. These gold-coated specimens were examined under Leo-440i scanning electron microscope to record the fracture mode and also to determine the operating mechanisms of fracture.

Fracture Toughness Evaluation

Mode-I (pure tensile) Fracture Toughness There are no standard test procedures to evaluate the mode I fracture toughness of ceramic and composite materials. Hence, ASTM standard E-399 (1997)11 for the evaluation of mode-I plane-strain fracture toughness (KIc) of metallic materials was followed in the present investigations. Specimen blanks of approximate dimensions 10 x 8 x 50 mm were cut from the as-received blocks on high speed cutter and then ground to shape by using diamond impregnated grinding wheel. Notches of approximate depth of half the width (4 - 4.5 mm) have been introduced with the help of a high speed cutting machine, Isomet ­ 4000 and then a fine notch of 2 mm depth using thin wafer blades of thickness 0.15 mm. Mode-I fracture toughness of the specimens was determined using SENB specimens loaded under threepoint bend loading. The tests were conducted both at ambient and elevated temperatures of 1000°, 1200° and 1300°C, on a computer controlled servo hydraulic Instron 8801 test system, in laboratory air atmosphere. Threepoint bend tests have been conducted using selfarticulating silicon carbide fixture in ramp control at a constant ramp rate of 0.5 mm/min. The load ­ displacement data thus obtained were analysed to evaluate the mode-I fracture toughness, KIc.

Special Edition No. 1, 2006

Figure 2. Three ­ point bend loading using asymmetric notch for the mixed-mode I/II loading conditions. 57

are also included. From the load displacement data in Figure 3, it is evident that the sintered alumina has shown mode-I fracture with little or no stable crack extension up to 1200°C and thus a valid KIc up to this temperature. The material at 1300°C has shown considerable stable crack extension and hence, yielded an invalid KIc. The data in Table 2 and Figure 4 clearly reveal that the sintered alumina exhibits a decreasing mode-I fracture toughness with increasing temperature. The material loses considerable mode-I fracture resistance as the temperature is increased. The ambient temperature Figure 3. Variation of load with the ramp displacement during mode-I fracture toughness is more than 50% fracture toughness evaluation of sintered alumina at different test higher as compared to the KIc up to temperatures. 1200°C. The steep fall in the mode I fracture resistance can be attributed to the sharp decrease in the strength of sintered alumina and for the reason that Results and Discussion the fracture is controlled at these temperatures by the Mode ­ I Fracture Toughness Figure 3 illustrates the variation of load with displacement conditions of linear elastic fracture mechanics (LEFM). for the specimens tested under mode-I loading at ambient and elevated temperatures. The material shows gradual increase in the load with the displacement before the onset Mixed-Mode I/II Fracture Toughness The mixed-mode I/II fracture toughness in the present of final failure. The fracture toughness is evaluated as per the ASTM standard E-399 using the following expression: investigation is evaluated using three point bend specimens having an eccentric notch. Asymmetric bending arrangements have been used to determine the stress KIc = (PQ.L/ B W3/2) F(a/W) [1] intensity factors which show a negligible mode-I contribution. The conditions of loading, specimen details Where, f (a/W) is a polynomial function of a/W and is and the resulting fracture parameters for the specimens obtainable from the ASTM standard E-399 (1997)13. The tested at ambient temperature are included in Table 1. specimen dimensions and the value of the mode-I fracture The resolved mode-I (KI); mode-II (KII) and the total (Kt) toughness obtained at ambient temperature are given in stress intensity factors under mixed-mode I/II loading are Table 1. Since, all the validity conditions are fulfilled, the derived as per the equations1: mode I fracture toughness is termed as the mode-I planeKI = s0 FI (IIa)1/2 [2] strain fracture toughness (KIc) and its value is found to be 4.16 MPa.m1/2 at ambient temperature. KII = s0 FII (IIa)1/2 [3] Similarly, the specimen details and determined values of fracture toughness at elevated temperatures of 1000, 1200 and 1300°C are given in Table 2. For the sake of Where, sO = 3 P.L / 2 b.W2 [4] comparison, the data pertaining to the ambient temperature Table 1.Mode-I and mixed-mode I/II fracture toughness parameters, evaluated in the present study.

58

Journal of Advanced Materials

Table 2. Mode-I fracture toughness of sintered alumina at elevated temperatures.

Finally, total stress intensity factor (Kt) can be evaluated using simple co-planar strain energy release rate criterion14 for the mixed-mode loading condition as: Kt = (KI2 + KII2)1/2 [7]

Figure 4. Variation in mode-I fracture toughness with test temperature. The mode-I and mode-II geometrical factors (FI and FII) are calculated by the weight functions. The results are expressed by the normalized geometric functions1: F¢I = FI(1-a/W)3/2 and, F¢II = FII(1-a/W)1/2 [6] [5]

The variation of load with ramp displacement for the varied 2d/L values is shown in Figure 5. It is to be noted that the asymmetrical loading provides increasing mode-II component with increasing 2d/L values. The data in Table 1 clearly show that the increase in the imposed mode-II fracture component (KII) decreases mode-I (KI) as well as total (Kt) fracture toughness of the sintered alumina, though the extent of such decrease is marginal (within 10%). This is clearly shown in Figure 6. Such marginal effects of imposed mode-II component on the mode-I fracture resistance indicate that the design could still be based on mode I fracture toughness in sintered alumina; though, it is better and technically appropriate if the design is based on Kt. In order to examine whether the fracture behaviour of sintered alumina under mixed-mode I/II conditions remain the same even at elevated temperatures, tests were conducted on one set of specimens with 2d/L values of

where, FI and FII are defined by Equations [2] and [3]. The values of the geometrical factors (F¢I and F¢II) can be derived from the work of Fett and coworkers1 for varied specimen dimensions. Using these values of the geometrical factors, one can derive the values of KI and KII, the resolved mode-I and modeII stress intensity factors for a mixed-mode I/II loading condition. Special Edition No. 1, 2006

Figure 5. Load ­ ramp displacement curves obtained from the mixed-mode I/II testing of the sintered alumina specimens with varied 2d/L ratio at ambient temperature. 59

Figure 6. Variation of resolved mode-I and mode-II fracture toughness along with the total fracture toughness (Kt) of sintered alumina at ambient temperature. 0.4, 0.6 and 0.7 at 1200°C. Figure 7 shows the variation of load with the ramp displacement for these tests at 1200°C. Load ­ displacement plots obtained at the elevated temperatures are found to be similar to those recorded at ambient temperature. The values of resolved mode-I, mode-II and the total mixed-mode fracture toughness (Kt), derived from the data in Figure 7 are given in Table 3 and are shown in Figure 8. The data in Table 3 and Figure 8 clearly show that the imposed mode-II components of fracture has a marginal effect (a slightly increased value up to 2d/L of 0.4 and then a small drop) on the total fracture resistance of sintered alumina even at this elevated temperature of 1200°C.

Figure 7. The variation of load ­ ramp displacement for the mixed-mode I/II specimen loaded at 1200°C with notch being located for 2d/L values of 0.4, 0.6 and 0.7. For the sake of comparison, the data pertaining to pure mode-I loading (2d/L = 0) are also included.

Crack Front Profiles

Crack front profiles provide valuable information regarding the nature of fracture and aid in determining the controlling fracture mechanism and also, the fracture criteria4-6,13,14. The fractured specimens tested under mode-I and the mixed-mode loading are observed under stereo-optical microscope to record the crack front profiles. These are presented in Figures 9 and 10 for the ambient and elevated (1200°C) temperatures, respectively. It can be inferred from these figures that the sintered alumina shows mode-I controlled fracture. This is because that all the specimens without exception have shown uniform crack front rotation towards mode-I fracture and the extent of such rotation increases with increased values of 2d/L. Such an observation was found valid for both the test temperatures.

Fractography

Figures 11a and b show the fracture features of sintered alumina, subjected to mode-I (tensile) loading at room temperature. Both low and high magnification pictures show that the material fails by quasi-cleavage, a low energy fracture mode. The mixed-mode I/II fracture features are also similar. Figures 12a and b show the fracture features of the mixed-mode specimen with 2d/L of 0.6, tested at ambient temperature. The sintered alumina shows fracture features under mixed-mode I/II loading conditions that are again similar to those of mode-I fracture, i.e., grossly quasicleavage. The nature of fracture does not change with the test temperature. The specimens tested at the elevated temperature of 1200°C too have revealed quasi-cleavage fracture under both tensile and mixed-mode loading conditions (Figures 13 and 14).

Figure 8. Variation of resolved mode-I and mode-II fracture toughness along with the total fracture toughness (Kt) with 2d/L of sintered alumina at an elevated temperature of 1200°C. 60

Journal of Advanced Materials

Figure 9. Crack path profiles showing the crack front rotation in the mixed-mode I/II tested specimens, tested at ambient temperature (All micrographs are obtained at 6X).

Comparison with Published Work

In the studies reported till date on the ambient temperature mode-I and mixed-mode I/II fracture behaviour of various grades of alumina, Suresh and co-workers4 and Fett and co-workers1 have shown that the additional imposed modeII components decrease the overall fracture toughness of the material, Kt (Table 4). Apart from this important finding, Suresh and co-workers have also shown that the sintered alumina has a mode-II fracture toughness that is quite comparable in magnitude with that of the mode-I fracture toughness and the four point bend asymmetrical loading (similar to the presently used three point asymmetrical loading) is very effective in reducing the frictional effects and thus, provide accurate fracture toughness values. The

ambient temperature mode-I fracture toughness values of the present material is significantly higher (nearly 25%40%) as compared to that determined for various grades of alumina reported by Suresh and co-workers4 and Fett and co-workers1. Such a difference in fracture resistance could be attributable to the nature of second phases present in the material, assuming the notch effects are negligible. The data reported by Fett and co-workers1 clearly show that the alumina material posseses similar values of modeI fracture toughness with different notch preparations such as saw cut and precracked under a variety of specimen loading conditions. Assuming that the presently used sharp machined notches do not influence the materials' mode-I fracture toughness, KIc, then the above difference

Table 3. Mode-I and Mixed-Mode I/II Fracture Toughness Parameters at an elevated temperature of 1200°C.

Special Edition No. 1, 2006

61

Figure 10. Crack path profiles showing the crack front rotation in the mixed-mode I/II tested specimens, tested at 1200° C (All micrographs are obtained at 6X).

in the fracture toughness values could be attributed entirely to the presence of a different second phase. While the present sintered alumina contained 3-4 vol. % magnesia, the other two materials have similar quantities of silica, in addition to small quantities of magnesia and traces of Ca, Na and Fe. Though both Suresh and co-workers4 and Fett and coworkers1 did not report total fracture toughness, a simple determination of the overall fracture toughness Kt using the equation 7, clearly indicate that their results too follow a trend similar to the one observed in the present study indicating that the imposed mode-II components have lowered the overall fracture toughness Kt, their effects strongly depend on the notch condition and the evaluating method (see data in Table 4). When compared to the results of present material, these two materials have exhibited a considerable fall in Kt with imposed mode-II fracture components. Hence, these results indicate that while it is more appropriate to design the structures and components based on the Kt values, the KIc values sufficiently describe the fracture resistance for design purposes in case of the present study material - sintered alumina.

increasing test temperature. This has been attributed to the loss in strength and the fact that the fracture is mainly controlled by linear elastic fracture conditions. 2. The imposition of mixed-mode I/II loading does not have a significant influence on the fracture resistance of sintered alumina at both ambient and at elevated temperature of 1200°C which is evident from the crack profiles, which shows crack front rotating towards mode-I fracture direction and also, Kt values are only marginally lower than the KIc values. 3. The mode of fracture remains the same both at ambient and at elevated temperatures, which is essentially "quasicleavage". Imposed mode-II components too have no effect on the nature of fracture mode. 4. The results from this study point to the fact that the design of structures and components made from sintered alumina could be based on KIc values and more precisely on Kt values, which are only marginally lower (within 10%) than the KIc values.

Conclusions

1. Both, mode-I and mixed-mode I/II fracture resistance parameters, K Ic and K t , respectively decrease with 62

Acknowledgements

The authors are indebted to Dr. D Banerjee (CCR&D, Journal of Advanced Materials

Table 4. Fracture Toughness Data of Various Grades of alumina.

Figure 11. (a) Low magnification and (b) High magnification fractographic pictures showing quasi-cleavage fracture of the mode-I (tensile) tested specimens at ambient temperature. Special Edition No. 1, 2006 63

Figure 12. (a) Low magnification and (b) High magnification fractographic pictures showing quasi-cleavage fracture of the mixed-mode I/II tested specimens at ambient temperature.

Figure 13. (a) Low magnification and (b) High magnification fractographic pictures showing quasi-cleavage fracture of the mode-I (tensile) tested specimens at 1200°C.

Figure 14. (a) Low magnification and (b) High magnification fractographic pictures showing quasi-cleavage fracture of the mixed-mode I/II (2d/L = 0.6) tested specimens at 1200°C. AMS, DRDO) for constant encouragement and valuable suggestions on this work. The authors thank Dr. A.M. Sriramamurty, Director, Defence Metallurgical Research Laboratory (DMRL) for the permission to publish these results. The authors are grateful to Defence Research and Development Organisation (DRDO) for the provision of facilities to conduct the present investigations. The authors would like to thank Dr. T Balakrishna Bhat, Division Head, ADDD of DMRL for kindly providing the material for the present study. 64

References

1. T. Fett, G. Gerteisen, S. Hahnenberger, G. Martin, D. Munz, "Fracture Tests for Ceramics Under Mode - I, Mode - II and Mixed ­ Mode Loading," J. Eur. Ceram. Soc., 15 307 (1995). 2. D. Hardy and D.J. Green, "Mechanical Properties of a Partially Sintered Alumina," J. Euorp. Ceram. Soc., 15 769 (1995). Journal of Advanced Materials

3. A. Muchtar and L.C. Lim, "Indentation Fracture Toughness of High Purity Submicron Alumina," Acta Materialia, 46 1683 (1998). 4. S. Suresh, C.F. Shih, A. Morrone, and N.P. O'Dowd, "Mixed ­ Mode Fracture Toughness of Ceramic Materials," J. Am. Ceram. Soc., 73 1257 (1990). 5. J.J. Petrovic, "Mixed ­ Mode Fracture of Hot-pressed Si3N4," J. Am. Ceram. Soc., 68 348 (1985). 6. J.J. Petrovic, "Mixed ­ Mode Fracture of Ceramics," In Fracture Mechanics of Ceramics, Plenum Press, New York, 8 127 (1986). 7. D.K. Shetty, A.R. Rosenfield, and W.H. Duckworth, "Mixed ­ Mode Fracture of Ceramics from Surface Flaws in Diametral ­ Compression," J. Am. Ceram. Soc., 69 437 (1986). 8. D.K. Shetty, A.R. Rosenfield, and W.H Duckworth, "Mixed ­ Mode Fracture in Biaxial Stress State: Application of the Diametral-Compression (Brazilian Disk) Test," Eng. Fract. Mech., 26 825 (1987). 9. T. Fett, "Mixed ­ Mode Stress Intensity Factors for 3 ­ Point Bending Bars," Inter. J Fracture, 48 R67 (1991). 10. T. Fett, "Mixed ­ Mode Stress Intensity Factors for the Oblique Edge Crack in Rectangular Specimens," Inter. J Fracture, 61 R3 (1993). 11. V. Tikare and S.R. Choi, "Combined Mode I and Mode II Fracture of Monolithic Ceramics," J Am. Ceram. Soc. 76 2265 (1993). 12. M. Li and M. Sakai, "Mixed-Mode Fracture of Ceramics in Asymmetric Four-Point Bending: Effect of Crack-Face Grain Interlocking/Bridging," J Am Ceram. Soc. 79 2718 (1996). 13. ASTM standard E ­ 399 (1997), Annual Book of ASTM Standards, Vol. 03.01 Philadelphia, Am Soc. For Testing and Materials, 408 ­ 427. 14. D. Broek, "Elementary Engineering Fracture Mechanics," Sijth Off and Noordhoff Publications, The Netherlands (1978).

Special Edition No. 1, 2006

65

Pressureless Sintering of a Magnesia Doped Gelcast Alumina Ceramic

C. T. Bodur Department of Mechanical Engineering, Istanbul Technical University, Istanbul, Turkey F. Cinar Sahin Department of Metallurgical and Materials Engineering, Istanbul Technical University, Istanbul, Turkey

Abstract

Small amount (a few hundreds of ppm) of magnesia doped gelcast formed alumina was pressureless sintered at 1600°C for 1 hour. The sintered alumina was examined by scanning electron microscopy (SEM) for microstructural investigation. Linear shrinkage, density and Vickers hardness of the sintered alumina were measured as 16.57%, 3.97 g/cm3 and 21.6 GPa, respectively. It was observed from the SEM investigation and density measurements that a full densification of the alumina was achieved with a fine grain structure.

Introduction

Alumina ceramics have excellent hardness, good chemical stability, high heat and electrical resistance1. They find applications in such diverse areas as cutting tools, abrasives, corrosion resistance materials, biomaterials, refractories and electrical insulators. Ceramic parts should be as dense as possible for the mechanical stability of the material. Densification of the ceramic powders is typically achieved by cold/hot-axial pressing, cold/hot-isostatic pressing (CIP/HIP), chemical reaction bonding and sintering processes2. In order to achieve dense ceramic parts, both pressure and heat should be employed either separately or simultaneously during the production of these materials. Applying axial pressure during cold forming, i.e., green body production (compaction of the powders) or during the sintering stage, restrict the shape of the ceramic parts. Shaping of axially pressed ceramic parts by machining could be difficult and costly. Applying CIP and HIP during the production of ceramic parts would also increase the production costs. Complex shape ceramic parts can be produced by reaction bonding or sintering without pressure, but then the material may become porous. Shape factor (formation) of the ceramic parts and the densification method of the material should be optimized. Gelcasting is one of the modern ceramic forming technologies3,4. In gelcasting of ceramic materials, a slurry of ceramic powders mixed with organic monomers cast into molds. Polymerization of the monomers leads to green ceramic parts3,4. If these green parts were sintered by HIP, production costs would increase. Another method of producing ceramics from these green parts could be sintering without pressure. Complex shape ceramic parts have been successfully produced by gelcasting processes3,5. In this work, gelcasting and pressureless sintering of a MgO doped Al2O 3 ceramic were investigated. Small amounts, a few hundreds of ppm of MgO additions in Al2O3 powders, prevent abnormal grain growth and this result in a homogenous distribution of equiaxed Al2O3 grains with a pore free microstructure6). 66

Densification and microstructures were analyzed by density measurements, linear shrinkage measurements and scanning electron microscopy (SEM). Microhardness measurements were performed on the sintered alumina samples for the mechanical characterization of the material.

Experimental

Materials 1. Powder: Alumina (Al2O3) Alumina powder, RC-HP (99.98 % a- Al2O3 calcined phase with 0.05 % MgO addition) was obtained from the manufacturer, Malakoff Industries, Inc., Reynolds Company, Texas, USA. Manufacturer's data of the specific surface area and the median particle diameter of the starting alumina powder were 7.5 m2/g and 0.35 mm, respectively.

2. Chemicals: 2.1. Binder: Monomers 2.1.1. Methacrylamide (MAM) 2.1.2. Crosslinker: N, N'-methylene-bis-acrylamide (MBAM) 2.2. Solvent: Deionized water 2.3. Dispersant: Darvan 821A, a 40 % aqueous solution of ammonium polyacrylate (APA) 2.4. Catalyst: N,N,N',N'-tetramethylethylenediamine (TEMED) 2.5. Initiator: Ammonium persulfate (AP) Details of the chemicals and the vendors can be found in3,4. 3. Mold material: Wax. 60x5.4x5 mm3 sized cavities were machined in wax for casting bar shape specimens.

Preparation of the Slurry A 55 vol. % (83 wt. %) Al2O3 powder content (solid content) slurry was prepared. The recommended solid content of the slurry for gelcasting of ceramics was 45 to 65 vol. % 3,4. In this work, the slurry consisted of Al2O3

Journal of Advanced Materials

powder, binder, solvent (deionized water), dispersant, catalyst and the initiator. To calculate the volume percents, the theoretical density of Al2O3 was taken as 3.98 g/cm3 and the density of chemicals was taken as of water, 1 g/ cm3. MAM/MBAM weight ratio was 6. Solvent (deionized water) content in the slurry was approximately 15 wt. %. Dispersant content in the slurry was 2 wt. %. The amounts of the catalyst and the initiator in the slurry were small, 0.1 ml/g slurry and 1 ml/g slurry, respectively. For this reason, they were not taken into account in the calculation of the solid content of the slurry. First, the monomers (MAM and MBAM) were mixed with a small amount of deionized water in a plastic beaker. This mixture was stirred by using a glass bar until all of the monomers were dissolved (solution became clear) in the deionized water, heat was applied to accelerate this process. Next, the dispersant was added into the solution. Finally, the rest of the water was added to the solution. After the solution was prepared, Al2O3 powder was added in it and they were mixed thoroughly. The powder-contained solution was ball milled for at least 24 hours.

typical binder removal schedule was to heat the alumina parts in the oven at 100°C/h from RT to 150°C and 15°C/h from 150°C to 600°C. When the temperature reached 600°C, the power was switched off automatically and the parts were cooled to RT. The color of the material was white. The de-bound alumina parts were very fragile.

Sintering Sintering of the de-bound alumina parts was done in air in a temperature controlled alumina furnace. The following furnace schedule was applied for the sintering process: heating from RT to 900°C at 20°C/min, from 900°C to 1600°C at 10°C/min, holding for 1 h at 1600°C and cooling down to RT in furnace, naturally. After sintering the color of the specimens changed from white to ivory. Linear Shrinkage Linear dimensions of the bar shape specimens in the de-bound condition and also after sintering were measured for the linear shrinkage calculations. Linear shrinkage was defined as the percent change of the linear dimensions of the cast material after the de-binding and sintering processes. Density Densities of sintered alumina samples were measured by Archimedes' principle in purified water. Scanning Electron Microscopy (SEM) SEM investigations were done on the sintered material in order to analyze the microstructure of gelcast produced Al2O3 parts. 5 mm long samples of cross section of about 5x5 mm2 were cut from the sintered alumina parts with a diamond blade. The samples were ground and then polished up to 1 mm size by diamond paste. Polished samples were thermally etched in air in an alumina furnace at 1500°C for 2 h to reveal the grain boundaries. The samples were coated with gold and examined in a scanning electron microscope (Jeol, JSM-5410, Japan). Vickers Microhardness Vickers microhardness measurements were done on the in Bakelite mounted samples by a Shimadzu Micro Hardness Tester, HMV-2, Kyoto, Japan. Indenter load on the samples was 2 kgf and the load applying time was 20 s. On each sample at least three microhardness measurements were done.

Casting

Casting was done in a chamber at room temperature (RT) in a nitrogen gas atmosphere. Before the casting operation, the mold was placed in the chamber and then the chamber was degassed for about 1 h by a rotary pump in order to avoid air entrapment during the polymerization process. The slurry was filtered by using a wire mesh and degassed in the chamber a few times (evacuation and nitrogen gas filling steps). TEMED and AP were added to the slurry and were mixed thoroughly. The slurry was degassed again. The slurry was immediately poured into the wax mold in a nitrogen atmosphere. Degassing and nitrogen filling of the chamber continued until no more air bubbles were observed in the casting process. Casting was done as quickly as possible in order to avoid the early gellation of the material (the slurry should be fluid enough to fill the cavities of the mold).

Polymerization The cast material was placed in an oven at 60°C in air for 1 h to gel (polymerization of the monomers). After 1 h, the oven was switched off and the material was left overnight for cooling. Mold Removal The wax mold was removed from the gelled material by melting it in the oven at 95°C for about 1 h. The gelled Al2O3 parts were cleaned in a bioact fluid at 100°C for about 0.5 h. Drying The gelled alumina parts were dried in air at 60°C for 24 h. Binder Removal Binder removal (burning of polymers in the gelled alumina) was done in a temperature controlled ventilated oven. A

Special Edition No. 1, 2006

Results and Discussion

In Table I, the results of the linear shrinkage, density and hardness of the sintered alumina samples are shown. The values of the linear shrinkage and the hardness of the sintered alumina samples are consistent with the typical values of the alumina ceramics. It is observed from the results of the density measurements that nearly full densifications were achieved with a sintered density of 3.97 g/cm3 (theoretical density of alumina is 3.98 g/cm3) 67

Table 1. Linear shrinkage, density and vickers hardness of the sintered alumina samples.

of this MgO-doped gelcast-shaped pressureless-sintered alumina ceramic. A scanning electron micrograph of the thermally etched sintered alumina sample is shown in Figure 1. It is seen on the micrograph that the pressureless sintering of the gelcast shaped alumina ceramic achieved pore-free microstructure. The grain size of this material is relatively small. Two different grain sizes are observed in the microstructure, smaller grains with about 1-2 mm and relatively larger grains with about 3-5 mm. Scratch lines, seen on some of the grains, were probably produced during the grinding and polishing stages of the specimen preparation for the SEM investigation.

Conclusions

After this investigation, it can be concluded that relatively easy and cost effective gelcast shaping and pressureless sintering of alumina ceramic parts can lead to a pore-free and uniformly distributed fine grain microstructure with nearly complete densification.

Figure 1. SEM of sintered and thermally etched alumina sample.

References

1. M. Miyayama, K. Koumoto, and H. Yanagida, "Engineering Properties of Single Oxides," Engineered Materials Handbook 4, Ceramics and Glasses, ASM Int., 748 (1991). 2. G. Ziegler, J Heinrich, and G. Wotting, "Review: Relationships Between Processing, Microstructure and Properties of the Dense and Reaction-Bonded Silicon Nitride," J. Mater. Sci. 22 3041 (1987). 3. O.O. Omatete, M.A. Janney, and R.A. Strehlow, "Gelcasting- A New Ceramic Forming Process," Ceram. Bul. 70 [10] 1641 (1991). 4. A.C. Young, O.O. Omatete, M.A. Janney, and P.A. Menchoffer, "Gelcasting of Alumina," J. Am. Ceram. Soc. 74 [3] 612 (1991). 5. A.G. Cooper, S. Kang, J.W. Kietzmann, F.B. Prinz, J.L. Lombardi, and E. Weiss, "Automated Fabrication of Complex Molded Parts Using Mold Shape Deposition Manufacturing," Mater. and Des. 20 83 (1999). 6. S.J. Bennison, "Grain Growth," Engineered Materials Handbook 4, Ceramics and Glasses, ASM Int., 304 (1991). 68 Journal of Advanced Materials

Study of the Damage Induced by Ar+ on the Highly Oriented Pyrolitic Graphite (HOPG ) Surface Using Extended Electron Energy Loss Fine Structure in Reflection Mode

C. González-Valenzuela and A. Duarte-Moller Centro de Investigación en Materiales Avanzados S.C., Chihuahua, México E-mail: [email protected] Original Manuscript Received 05/13/02; Revised Manuscript Received 10/08/02

Abstract

Extended Electron Energy Loss Fine Structure (EXEELFS) technique has been used to perform systematic experiments to detect the damage induced by Ar+ on a clean Highly Oriented Pyrolitic Graphite (HOPG) surface. Different sputtering times with Ar+ were used in order to perform these experiments. A commercial UHV chamber equipped with a single pass Cylindrical Mirror Analyzer (CMA) and an energetic Argon ion gun were used to sputter the surface. Results show that the electronic structure of the sputtered surface is substantially different from non-sputtered surface, likewise the atomic distribution around carbon atoms is changed with respect to clean surface showing a contraction of 0.04 nm, for first nearest neighbors and 0.08 nm, for the second nearest neighbors. Auger and EXEELFS results show that the damage induced by Ar+ ions on clean HOPG surface give rise to local amorphization on itself.

Introduction

In the last few years much attention has been paid in development of extended electron energy loss fine structure (EXEELFS) technique. Some people around the world focused their work in characterizing materials in both, reflection and transmission geometries1-8. These studies shown that EXEELFS technique is a reliable way to obtain the local structure of a wide number of materials. By other hand, is possible to have a semi quantitative idea of the sp2:sp3 states ratio in some structures of carbon by measuring the difference on energy between the maximum and minimum in a characteristic Auger peak9 or by measuring the intensities of the p* and s* transitions in the energy loss spectra10. In an Auger experiment this difference in energy is denoted by DE and is measured as shown in Figure 1. Thus, for HOPG we measured DE =23 eV, (which contain 100% of sp2 states and for diamond we have De=13 eV which contain 0% of sp2 states or 100% sp3 states. In this sense it is possible to find structures with 13eV < DE < 23eV with sp 2 states corresponding to 0< %sp2 <100.

electron energy loss spectra were collected in standard Auger spectroscopy mode, N(E)*E, using a primary electron energy of 1500 eV with a current of 1mA and a typical resolution of . The first derivative

of the collected spectra around the characteristic Auger transition energy was done in order to measure their respective line shapes and the

Experimental Details

All the experiments were taken at room temperature in a commercial UHV chamber for surface analysis, with a base pressure of 2.5 x 10-10 Torr. A Perkin-Elmer PHI-595 scanning Auger microprobe with single pass cylindrical mirror analyzer, (CMA), and coaxial electron gun, which can operate from 1 to 10 K eV. The chamber was equipped with Ar ion gun system for cleaning surfaces. Auger and Special Edition No. 1, 2006

Figure 1. C-CVV Auger transition for HOPG and diamond structures. 69

Table 1. Auger peak to peak difference on energy, DE, sowing the sp2:sp3 ratio among different sputtering times.

corresponding DE values. Experimental conditions were changed in order to perform EXEELFS experiments around the ionization C- K edge. In this, case, a primary electron energy of 1500 eV, a current of 1.5mA with a resolution of and typical time of 120 minutes were used. Raw data were collected in a range of 350 eV beyond the ionization C-K edge, located at an energy loss of 283.8 eV with respect to the elastic peak. The samples were HOPG surfaces collected after different sputtering times. These sputtering times were: 5, 10, and 15 minutes respectively. Auger, plasmon and fine structure spectra were carried out often each sputtering time. Great care was taken on fine structure acquisition in order to have a good signal-noise ratio. Radial distribution function (RDF) spectra were done using the standard extended X-ray absorption fine structure (EXAFS) procedure11,12 without applying the phase shift correction.

Results and Discussion

Figure 2 shows the characteristic C-CVV Auger peak from each experiment. It is noticeable the difference on energy, DE, among them. Similar structure (in form and shape) appears at the sputtered surface during five minutes and the respective clean surface. In these cases, the

quantify DE is about 22 eV. This quantity is slightly different to that corresponding to the sputtered surface during 10 and 15 minutes. In these spectra, apparently there are changes of sp2 states in our surface. It is well known that graphite has 100% sp2 states with DE =23eV and diamond has 100% sp3 states (0% sp2 states) with DE = 13 eV. On the other hand, there are other structures containing carbon as the only element having a sp2 ratio and 70% or less, like amorphous carbon. In this particular case, there is a range of 78.25% to 95.65% sp2 states for high and low sputtering times respectively, as shown in Table 1. Figure 3 displays the plasmon window where the most prominent plasmon peak of each case is located at the same energy loss of 26 eV, corresponding to the plasmon (p+s) in the graphite structure13. Apparently no changes have been occurred during the sputtering process. However by analyzing the non-sputtered surface we note the prescence of the known plasmon p located at an energy loss of 6.7 eV, which is characteristic to the HOPG structure. Figure 4 shows the second derivative of the fine structure spectra where is noticeable that the fine structure features are substantially different among them. This difference can be explained in terms to the different density of states for the analyzed samples. The differences are even more noticeable distinct than in the case of the Auger spectra.

Figure 2. Auger spectra from HOPG surface exposed at different sputtering times. 70

Figure 3. Low loss window for the HOPG surface at different sputter times. Journal of Advanced Materials

Figure 4. Raw data of the extended fine structure around the ionization C-Kedge taken with an incident energy of 1500 eV. In this region, the respective p* and s* is well defined in the clean surface (before erosion), indicating that a structural change has been occurred. Unfortunately, experiments in this geometry, (reflection mode), have not a good signal to noise ratio, because of the low primary electron energies were used (1 to 3 KeV). However in our opinion this is the most efficient way for studying damage or changes on surfaces at UHV environment. In order to verifying any changes on the sputtered surface, the radial distribution function, (RDF), have been done according to the EXAFS procedure without applying the phase shifts correction. Figure 5 reports the RDF obtained by applying the Fourier transform to spectra shown in Figure 4. This figure shows that effectively, strong structural changes have been occurred in the HOPG surface during the sputtering process with Ar+. On the clean surface there are values of 0.12 and 0.26 nm for the first and second nearest neighbors. That same surfaces sputtered for five minutes shows values of 0.1 and 0.21 nm respectively. Increasing the sputtering time at 10 and 15 minutes these values are shifted to 0.08 and 0.17 nm for the first and second nearest neighbors. Table 2 shows the numerical values for the first and second nearest neighbors.

Figure 5. Radial distribution function obtained by applying the Fourier transform to the spectra of Figure 4.

Conclusions

Through systematic experiments have been found that the sp2 changes from a sputtered HOPG surface can be detected in situ by Auger electron spectroscopy with a good resolution. The idea that a structural change occurred during the sputtering process has been stored through EXEELFS analysis, which indicate a marked difference in their respective atomic positions. Locally it has been found a redistribution of carbon atoms on the HOPG surface. We have been demonstrated that EXEELFS, in the reflection mode, can be used successfully to determinate in situ structural changes on surfaces. It is of importance when one need to make studies on clean or non-damaged surfaces in UHV environment. In these cases EXEELFS can give an initial information about the atomic structure of the surface under analysis. In this case the sp2 states ratio was 78.25<%sp2 <100.

References

1. M.M. Disko, C.C. Ahn, B. Fultz, "Transmission Electron Energy Loss Spectrometry in Materials Science," EMPMD Monograph series, 2. (1991). 2. R.F. Egerton, "Electron Energy Loss Spectroscopy in the Transmission Electron Microscope," Plenum Press (1996).

Table 2. Numerical values for the first and second nearest neighbors obtained by applying the EXEELFS procedure to the EELS spectra from Figure 4.

Special Edition No. 1, 2006

71

3. M. De Crescenzi, "Structural Surface Investigations with Low-energy Backscattered Electrons," Surf. Sci. Reports. 21:89-175 (1995). 4. A. Cook, AG. Fitzgerald, F. Ibrahim, J.I.B. Wilson, and P. John, "An EELS and EXELFS Study of Amorphous Hydrogenated Silicon Carbide," Mikrochim. Acta, 114/115: 255-260 (1994). 5. E. Zschech, G. Biode, and V. Klemm, "The Promise of EXAFS/EXELFS Spectroscopy for Structure Analysis in Ceramic Composites," J. Europ. Ceram. Soc. 10: 213220 (1992). 6. A. Duarte-Moller, L. Cota-Araiza, L. Morales de la Garza, G. A. Hirata, D.H. Galván, M. Avalos-Borja, "Identifficagtion of Different Forms of Carbon by Extended Energy Loss Fine Structure," Appl. Surf. Sci. 108: 59-63 (1997). 7. A. Duarte-Moller, O. Contreras, G.A. Hirata, M. AvalosBorja, D.H. Galvean, L. Morales de la Garza, L. CotaAraiza, "PEELS and EXELFS Characterization of Diamond Films Grown by the HF-CVD Technique on Non-scratched Silicon Substrates," Thin Sol. Films. 304: 45-47 (1997). 8 A. Duarte-Moller, F. Espinosa-Magana, R, Marteinez S.M. Avalos-Borja, G.A. Hirata, L. Cota-Araiza, "Study of Different Forms of Carbon by Nalytical Electrón Microscopy," J. Elect. Spect. Relat. Phen. 104: 61-66 (1999).

9. S.T. Jackson, R.G. Nuzzo, "Determinig Hybridization Differences for Amorphous Carbon from XPS C 1s Envelope," Appl. Surf. Sci. 90: 195-203 (1995). 10. S. Zhang, X.T. Zeng, H. Xie, P. Hing, "A Phenomenological Approach for the Id/Ig and sp3 Fraction of Magnetron Sputtered a-C Films," Surf. Coat. Technol. 123: 256-260 (2000). 11. D.P. Woodruff, "Fine Structure Ionization Cross Sections and Applications to Surface Science," Rep. Prog. Phys. 49: 683-723 (1986). 12. E.A. Stern, D.E. Sayers, F.W. Lytle, "Extended X-ray Absorption Fine-structure Technique. III. Determination of Physical Parameters," Phys. Rev. B. 11: 4836-4846 (1975). 13. P. Kovarik, E.B.D. Bourdon, and R.H. Prince, "Electronenergy-loss Characterization of Lase-deposited a-C, a-C:H, and Diamond Films," Phys. Rev. B. 48: 12 123-12 129 (1993).

72

Journal of Advanced Materials

Synthesis and Mechanical Properties of La2O3-WSi2/MoSi2 Composites

Houan Zhang, Ping Chen, Siwen Tang Institute of Material Surface Engineering, Hunan University of Science and Technology, Xiangtan, Hunan, China Original Manuscript Received 04/27/04; Revised Manuscript Received 09/13/04

Abstract

La2O3-WSi2/MoSi2 composites phase transformation was analyzed with X-ray during the synthesis process by mechanical alloying and self-propagating high-temperature synthesis. Effects of La2O3 on mechanical properties of WSi2/MoSi2 were discussed by SEM. Results showed that the phase transformation order is Mo, W and Si, followed by Mo (W) solid solution and Si solving in Mo (W) and WSi2/MoSi2 during the mechanical alloying process. With the increases of La2O3 content, the alloying process is delayed. The effect of strengthening and toughening MoSi2 matrix simultaneously reinforced by La2O3 and WSi2 is more obvious than only by WSi2, especially with the addition of 0.8wt% rare-earth, which increase the composite hardness value and fracture toughness value by about 2.07HRA and 49.5%, respectively. The transformation of fracture model and crack deflection and the existence of microbridging results in very high toughness of La2O3-WSi2/MoSi2 composite.

Introduction

Intermetallic molybdenum disilicide (MoSi2) has been welcome as an advanced high temperature structural material because of its high melting point, moderate density, excellent elevated oxidation resistance and erosion resistance and high conductivity, especially useful in aerospace applications 1-3. However, its practical application is limited by brittleness at room-temperature and low creep strength under high-temperature (>1000ºC). Compounding, for instance, by adding SiC, ZrO2, TiB2 etc., and alloying with W, Al etc. are effective methods to increase high-temperature strength and room-temperature toughness4-6. Schwarz et al.7 compared the properties of MoSi 2 matrix composites with the addition of 50mol.%Mo5Si3 and 50mol.%WSi2, and found that the solution strengthening effect of WSi2 on MoSi2 was more obvious than that of the fine-crystal strengthening effect of Mo5Si3. Subrahmanyam et al.8 also reported that the yield stress of WSi2/MoSi2 composite at 1500ºC was increased about 8~10 times than that of pure MoSi2. Authors et al.9,10 investigated the properties of MoSi2 matrix composite compounded with La2O3 or WSi2, and found that both of them could strengthen and toughen MoSi2. All of the above researches show that the addition of only single WSi2 or La2O3 should be an effective method to improve the strength and toughness of MoSi2. It is well worth discussing whether

the simultaneous addition of La2O3 and WSi2 will further improve the property of MoSi2. Therefore, this paper aims at discussing the synthesis and mechanical properties of La2O3 - WSi2/MoSi2 composite.

Experimental

Synthesis of La2O3 ­ WSi2/MoSi2 Composites The element powders used in this study were 99.9% pure Mo powder (2-4mm), 99.5% W powder (2-4mm), 99.3% Si powder (<43mm) and La2O3 powder (2-4mm). Three kinds of powder mixtures as described in Table 1 were mixed for 10 hours in a TM-2 type blend machine. A DSP-1P type planetary mill equipment was employed for the mechanical alloying (MA) experiments with a ratio of refractory alloy ball to powder weight of 20:1 and a rotation speed of the milling vial of 416 rpm. The whole milling process was under hydrogen atmosphere. Self-propagating hightemperature synthesis (SHS) proceeded in the home-made equipment. Suitable quantities of the above powders were taken out to observe the alloying progress by D500 type automatic X-ray diffraction using CuKa radiation. The powder size was determined by Winner 2000 type laser grain size equipment.

Mechanical Properties of La2O3 ­ MoSi2 Composites The synthesized powders were hot pressed in a graphite die at 1600~1700°C for 1h under 25MPa. The hot-press block was wire-cut to the size of a test Table 1. The component of Mo-W-Si-La2O3 system materials in the sample. The hardness of materials was measured by a HV-150A type sclerometer, experiment (atom,%). and the fracture toughness was determined by the three-point beam method. Each of the data was the average of five measure values. The microstructures of test samples were analyzed by using a S-570 or KYKY-2800 scanning electron microscope with EDS.

Special Edition No. 1, 2006 73

Figure 1. The XRD of A, B, and C powders milled for different times.

Results and Discussion

Phase Formation Process Figure 1 shows the X-ray diffraction patterns of A, B and C powder after high-energy milling at different times. It was found that Mo (W) solid-solution phase and Si phase exist after milling for 5h. Because the atomic radius of Mo and W after is 0.201nm and 0.202nm, the lattice constant is 0.3147nm and 0.3165nm, respectively, and they have the same body-centered cubic structure, it is easy to form Mo(W) solid solution . With increased milling time, the Si phase gradually vanishes, which means that Si is solidsolved into Mo and W. The same phenomenon is also found in the mechanical allying process of Mo: 2Si mixture powders11. After milling for 40h, a small amount of MoSi2 and WSi2 Bragg peaks were found. But it is too difficult to recognize them, which can be attributed to tetragonal MoSi2 and WSi2 having long-range structure. But the very proximity lattice constant (a is 0.3202nm and 0.3211, c is 0.7855nm and 0.7835nm, respectively), it can be considered that Mo in the MoSi2 is replaced by W and resulting in the formation of (Mo,W)Si2. After milling for 90h, there still exists an intense Bragg peak of Mo, therefore the mechanical alloying process is unfinished. Schwarz et al.7 got the similar result that Mo:W:4Si mixture powder was not completely alloyed after high-energy milling for 70h.

74

Effects of La2O3 on the phase formation are compared in Figure 2. It can be found that the strength of (Mo,W)Si2 phase peak decreases, and the strength of residual Mo(W) phase peak increases with the increase of La2O3 content. Meanwhile, the disappearing time of Si phase lengthens with the increase of La2O3 content as shown in Figure1. The above results indicate that the addition of La2O3 delays the formation of (Mo,W)Si2 phase by mechanical alloying.

Figure 2. The XRD of A, B and C powders milled for 40 hours. Journal of Advanced Materials

Figure 3. The XRD of powders synthesized by SHS. It is worth pointing out that La2O3 exists in the form of an oxide, but it cannot be observed an obvious diffraction peak in the Figure 1 and Figure 2, which is attributed to a small content of La2O3. Three mixture powders, A, B and C, are ignited by tungsten filament using a self-propagating high-temperature synthesis device. Because the reactive heat is enough to

cm2/cm3. However SHS powder is 16.04mm and 8607.63 cm2/cm3, respectively. Obviously, the grain size of the former powder is finer than that of the later powder; the surface activity of powder is stronger, which is useful to improve sintering characteristics. Authors12 contrasted the influence of MA and SHS on the sintering process and density of MoSi2 block. The sintering temperature of MoSi2 by MA is at least lower about 250ºC than that by SHS. The sintering apparent activation energy value of MoSi2 by MA is only about 37% of that by SHS, which displays an obvious mechanical activation sintering effect. However, we know it is difficult to synthesize La2O3 ­ WSi2/MoSi2 composite by MA. Therefore, the technology of SHS followed by milling for 2~4h is suggested in practical application, which not only can quickly synthesize La2O3 ­ WSi2/MoSi2 composite powder, but also can be an advantage to prepare compact block at very low sintering temperature. Many details have been reported in12.

Mechanical Property Table 2 shows the mechanical properties of La2O3 ­ WSi2/MoSi2 composite. As a comparison, table 2 also lists the properties of MoSi2 and WSi2/MoSi2 composite. It can be found, compared with pure MoSi2, that the

Figure 4. The granularity of powders synthesized by MA and SHS supply barrier energy of an adjacent reaction, the reaction properties of La2O3 ­ WSi2/MoSi2 composites are all largely is propagated in the form of combustion wave and the improved. The hardness values and fracture toughness whole reaction is completed within a few seconds. Figure values are increased 2.5 ~ 4.0HRA and 89.21% ~178.26%, 3 shows the XRD pattern of B powder after a combustion respectively. Compared with WSi2/MoSi2 composite, the formation. It can be found that there is only (Mo,W)Si2 phase. A and C powders have the similar XRD pattern, too. Therefore La2O3 ­ Table 2. Mechanical properties of different materials. WSi2/MoSi2 composite can be more quickly compounded by self-propagating high-temperature synthesis technology than by mechanical alloying. The grain sizes of two kinds of composite powder by SHS and MA are shown in Figure 4. The average grain diameter and surface area unit volume of MA powder is 4.28µm and 24724.69 Special Edition No. 1, 2006 75

toughen ceramic composites. These mechanisms include matrix microcracking, crack branching, crack deflection, crack bowing and fiber pullout. Carter and Hurley 15 have suggested crack deflection as an important toughening mechanism in SiC-whiskerreinforced MoSi2. Bhattacharya and Petrovic16 have provided evidence for crack-interface grain bridging (crack (a) 0.2% (b) 0.8% microbridging) in 20vol%SiC/ Figure 5. SEM micrographs of the fracture of WSi2/MoSi2 composites with different MoSi2 composite and crack content of La2O3. branching in 40vol%SiC/MoSi2 composite. The crack propagation behaviors in La2O3-WSi2/MoSi2 composite are hardness values and fracture toughness values of composite shown in Figure 6. Figure 6a-b indicate the crack deflection with complex La2O3 and WSi2 are also enhanced to a and mricrobridge. The similar phenomenon was observed different degree according to the different content of La2O3. in SiC toughening MoSi material by Henager et al17 and 2 Here, with addition of 0.2% La2O3, the hardness value is in ZrO phase-transformation toughening MoSi material 2 2 only increased 0.51HRA, but the fracture toughness value by Yi et al.18. These behaviors observed in this case, which is increased 21.6%. With the addition of 0.8% La2O3, the can be explained by the presence of a complex residual hardness value and fracture toughness value are increased stress field19 and the high resistance to crack propagation 2.07HRA and 49.56%, respectively. However with the large because of the addition of La O particles in WSi /MoSi 2 3 2 2 content of La2O3 (2.0%), the hardness value and fracture composite, will absorb more energy and thus improve the toughness value begin to decrease but they are still higher fracture toughness. than that of WSi2/MoSi2 composite. In this experiment, MoSi2 matrix composite with addition of 0.8% La2O3 and Conclusions 50mol.% WSi2 exhibited higher hardness and fracture (1) The phase transformation order is Mo, W and Si, toughness than the others. Above results exhibit excellent followed by Mo (W) solid solution and Si solving in Mo complex hardening and toughening effect of La2O3 and (W) and WSi2/MoSi2 during the mechanical alloying WSi2. process. With the increases of content of La2O3, the La2O3 exists in composite in the form of particles, which alloying process is delayed. It is more suitable to act as a grain refining and particle strengthening phase. synthesize La2O3 - WSi2/MoSi2 composite powder by SHS Meanwhile, WSi2 exists in composite in the form of solid than by MA. solution, which acts as a solid solution strengthening (2) Effect of strengthening and toughening MoSi2 matrix phase. All of these strengthening effects make La2O3 ­ simultaneously reinforced by La2O3 and WSi2 is more WSi2/MoSi2 composite have high hardness. The high obvious than only by WSi2, especially with the addition of toughness of La2O3 ­ WSi2/MoSi2 composite may be 0.8wt% rare-earth. Hardness values and fracture toughness contributed to the transition of the fracture mode. Figure 5 values of the former are higher about 2.07HRA and 49.5% shows the fracture surface of La 2 O 3 ­ WSi 2 /MoSi 2 composite. It can be seen that with the addition of 0.2% La2O3, the fracture surface is still a mixed intergranular-transgranular just as the fracture surface of pure MoSi2 and WSi2/ MoSi 2 composite 9,10 . However with the addition of 0.8% La2O3, the fracture surface is almost occupied by transcrystalline. (a) Crack deflection (b) crack microgridge Rice13,14 had reviewed the mechanisms that can Figure 6. The crack propagation behavior in La2O3-WSi2/MoSi2 composite. 76 Journal of Advanced Materials

than that of the later, respectively. The transformation of fracture mode and the existence of crack deflection and microbridging result in very high toughness of La2O3-WSi2/ MoSi2 composite.

13. R.W. Rice, "Mechanisms of Toughening in Ceramic Composites," Ceramic Engineering and Science Proceeding, (2)661(1981). 14. R.W. Rice, "Ceramic Matrix Composite Toughening Mechanisms," Ceramic Engineering and Science Proceeding, (6)589(1985). 15. D.H. Carter, G.F. Hurley, "Crack Deflection as a Toughening Mechanism in SiC-Whisker-Reinforced MoSi2," Journal of the American Ceramic Society, 70(4) C-79 (1987). 16. A.K. Bhattacharya, J.J. Petrovic, "Hardness and Fracture Toughness of SiC-Particle-Reinforced MoSi2 Composites," Journal of the American Ceramic Society, 74(10) 2700 (1991). 17. C.H. Henager, J.J.L. Brimhall, J.P. Hirth, "Synthesis of a MoSi2/SiC Composite in Situ Using a Solid State Displacement Reaction," Scripta Metallurgica et Materialia, 26(4) 585(1992). 18. D. Yi, C. Li, "MoSi2-ZrO2 Composites ­Fabrication, Microstructure and Properties," Materials Science and Engineering A, 261(1-2)89(1999), 19. K.T. Faber, T. Iwagoshi, and A. Ghosh, "Toughening by Stress-Induced Microcracking in Two-Phase Ceramics," Journal of the American Ceramic Society, 71(9) C399(1988).

Acknowledgements

The authors would likes to thank the National Natural Science Foundation of China and the Natural Science Foundation of Hunan Province of China for supporting this research under the Number 50272015 and 02JJY5003, respectively.

References

1. A.K. Vasudevan, J.J Petrovic, "A Comparative Overview of Molybdenum Disilicide Composites," Materials Science and Engineering A, 155(1-2)1(1992). 2. Y.L. Yen, E.J. Lavernia, "Review Processing of Molybdum Disilicide," Journal of Materials Science, (29)2557(1994). 3. J.J. Petrovic, "Key Development in High Temperature Structural Silicides," Materials Science and Engineering A, 261(1-2)1(1999). 4. J.J. Petrovic, "MoSi2-based High-temperature Structural Silicides," MRS Bulletin, 18(7)35(1993). 5. D.G. Morris, M. Leboeuf, M.A. Morris, "Hardness and Toughness of MoSi2 and MoSi2­SiC Composite Prepared by Reactive Sintering of Powders, " Materials Science and Engineering A, 251(1-2)262 (1998). 6. K. Niihara, Y.Suzuki, "Strong Monolithic and Composite MoSi2 Materials by Nanostructure Design," Materials Science and Engineering A, 261(1-2)6(1999). 7. R.B. Schwarz, S.R. Srinivasan, J.J. Petrovic, C.J. Maggiore, "Synthesis of Molybdenum Disilicide by Mechanical Alloying," Materials Science and Engineering A, 155(1-2)75(1992). 8. J. Subrahmanyam, R.R. Mohan, "Combustion Synthesis of MoSi2-WSi2 Alloys," Materials Science and Engineering A, 183(1-2)205(1994). 9. H. Zhang, D. Wang, S. Chen, X. Liu, "Toughening of MoSi2 Doped by La2O3 Particles," Materials Science and Engineering A, 345(1-2) 118(2003). 10. H. Zhang, P. Chen, M. Wang, X. Liu, "Room Temperature Mechanical Properties of WSi 2/MoSi 2 Composites," Rare Metals, 21(4) 304(2002). 11. H. Zhang, X. Liu, "Analysis of Milling Energy in Synthesis and Formation Mechanisms of Molybdenum Disilicide by Mechanical Alloying," International Journal of Refractory Metals & Hard Materials, 19(3)203(2001). 12. H. Zhang, P. Chen, X. Liu, "Effect of Mechanical Alloying on Sintering Densification of MoSi2," Rare Metal Materials and Engineering, 32(1)70(2003). Special Edition No. 1, 2006

77

Manufacturing and Mechanical Properties of Grids Braided from Stainless Steel/PP Functional Ply Yarn

Jia-Horng Lin and Shih-Hua Chiang Laboratory of Fiber Application and Manufacturing, Graduate Institute of Textile Engineering, Feng Chia University, Taiwan, R. O. C. Ching-Wen Lou Institute of Biomedical Engineering and Material Science/Center of General Education, Chungtai Institute of Health Sciences and Technology, Taichung, Taiwan, R. O. C. Original Manuscript Received 02/13/04; Revised Manuscript Received 06/25/04

Abstract

Fiber reinforced concrete have been researched and used for many years, which strengthen the building structures to prevent from cracks initiation and propagation. Nevertheless, fibers just belong to "one-dimension linear reinforcement", not "two-dimension plane reinforcement". In the study, polypropylene (PP, wrapping material) and stainless steel filaments (core material) were used to braid the ply yarn by a braiding machine. Then, the ply yarn was woven into grids and thermo-set in an oven. The grid will be used to reinforce the concrete in our future work. In the study, we would investigate the manufacturing and mechanical properties of grids braided from stainless steel/PP functional ply yarn. The results revealed that the braid stainless steel yarn (BSY) exhibited desirable tensile strength than others, which implied the braiding process could improve the performance of ply yarn.

Introduction

is much better than the tensile strength. Therefore, fiber With the increasing in population in the world for the reinforced concrete was created to improve the past few years, and residential areas for mankind have shortcoming. However, cracks on the concrete still been decreasing relatively, high buildings are the only propagate while subjecting to an excessive tensile force. solution nowadays. However, earthquakes occurred one In 1950s, engineering experts found the tensile properties after another around the world. It not only caused buildings of concrete could be improved by adding of stainless steel to topple, but also imposed great threats upon the lives fiber, and got satisfied results4. and properties of the human beings. Reinforced concrete In addition, the concrete improved with stainless steel has always been the major building structure. To prevent fiber is better than traditional concrete, such as deflection cracks growing on the concrete surface is the way that resistance, shear resistance, etc. The purpose of this study several works have attempted. For example, blend the aims at the manufacturing and mechanical properties of filaments or staples into the concrete. Redon Carl et al. novel ply braided yarn and grids. The tensile properties of studied reinforced concrete with iron fiber and effectively the ply yarn and grids would be determined to evaluate prevented cracks increasing1. Sydney Furlan Jr. et al. the practicability of reinforcing concrete. studied the concrete reinforced with steel and PP fibers, It is very interesting to study manufacturing process of and the result showed that the reinforced mechanism of ply yarn. Lin and Lou used nonwoven selvages to produce fibers in concrete just like with stirrups2. Jin-Kyung Lee et ply yarn5. Besides, they studied the electrical properties al. investigated carbon fiber sheet reinforced concrete, and of Stainless Steel/PP ply yarn6. the result reveal that it could remarkably decrease the Geosynthetic materials made of polymers mostly crack initiation and propagation3. After surveying the literatures, we found many researchers have paid attention to reinforcing the concrete with fiber materials which belong to "one-dimension linear reinforcement". In this work, a grid (a netlike structure) was produced to reinforce the concrete for the purpose of "two-dimension plane reinforcement". A new concrete material, stainless steel fiber reinforced concrete, has been developed successfully for several years. Concrete possesses greatly compressive strength which Figure 1. The appearance of ply yarn. 78 Journal of Advanced Materials

Experimental

Specification of the Materials (see Table 1)

Experimental Steps 1. The core material used in this experiment was a stainless steel filament with a diameter of 0.25 mm and the sheath material was a polypropylene filament. The ply yarns were braided with stainless steel filaments as core yarn and eight PP filaments, which wrapped the stainless steel filaments by a braiding machine as shown in Figure 2. 2. Three kinds of core yarn were braided in this experiment with single stainless steel filament (SSF), three stainless steel filaments (TSF) and braid stainless steel yarn (BSY) composed of three stainless filaments. Figure 1 shows the appearance of TSF ply yarn. 3. The ply yarn was then weaved into functional grids. Figure 2. The mechanism of braiding machine to produce ply yarn. 4. After fabricating the grids, it was put into the oven at 190°C for four minutes so that polypropylene filaments could be melted to surround and possess flat surface structure. Most of them are applied encircle the stainless steel filaments and to adhere to geoengineering, they are called geosynthetics. Eight junctions as shown in Figure 3 and Figure 4. types consist of geotextile, Geogrid, geonet, 5. Two testing methods about geotextile materials were geomembrane, geopipe, geoply, geofoam and conducted, which were rib tensile strength and junction geosynthetic clay liners7. strength, respectively. Development of geogrid has been the fastest technology 6. Upon completion of the tests, data were analyzed and among Geotextiles recently. According to ASTM D4439, compared. geogrid is defined as: "a geosynthetic formed by a regular network of integrally connected elements with apertures greater than 6.35 mm (1/4 in.) to allow interlocking with surrounding soil, rock, earth, and other surrounding materials to function primarily as reinforcement." According to the standards of geogrid flexibility strength test in ASTM D1388, geogrids can be divided into hard geogrids and soft geogrids. Soft geogrids are fabricated by weaving, while hard ones are formed by drilling on the plastic sheets. Resin is used to coat on the grid surface for protection. In this study, the soft grids would be fabricated and its mechanical properties would be determined. Polypropylene and stainless steel filaments were used to braid the ply yarn (as shown in Figure 1) by a braiding machine (Figure 2). Then, these ply yarn was woven into grids. Finally, the functional grids were setting in an oven at 190°C for four min. During the setting process, the polypropylene filaments were melted to surround and encircle the stainless steel filament to prevent stainless steel filament from damage and to reinforce the junction bind. Figure 3. The appearance of Stainless Steel/PP grid before heat setting.

Special Edition No. 1, 2006 79

Table 1. Specification of the materials.

*1D=1 g/9000 meter strength, width tensile strength, pull-out resistance and abrasion resistance. Tensile strength and junction strength are two major mechanical properties for evaluating geogrids performance. Tensile strength is expressed by the strength per width. In the study, rib tensile strength, and junction strength are briefly introduced as follows:

Figure 4. The appearance of Stainless Steel/PP grid after heat setting.

Rib Tensile Strength Test Geogrid rib tensile strength test is based on GRI-GG18, which is a method of testing single rib elongation and tensile strength for geogrids of various shapes and serves as a reference of the strength of a unit width for geogrids. A tension tester of constant rate extension is required. The clamp device of the testing machine has to be wide enough to clamp the specimens and prevent from slipping. The tensile testing was performed on HT-9101 computer servo control materials testing systems (Hung Ta Instrument CO., Ltd.). Three junctions are required for the specimens of hard geogrids. Upper and lower junctions are needed for the

Ply Yarn In this study, we used the braiding machine (Nan Xing Co., Ltd) to braid three types of ply yarns. The first one was fed in single stainless steel filament for core yarn. The second one was fed in three stainless steel filaments. The last one was braided three stainless steel filaments with braiding machine, and then used it for core yarn. Grids Ribs of grids8 Ribs can be divided into longitudinal and transverse ribs. 1. Longitudinal ribs: It is a major source for reinforced geogrids, parallel to mechanical direction. 2. Transverse ribs: It is a secondary source for reinforced geogrids, vertical against mechanical direction. Junction of grids8 Junction is a joint between transverse and longitudinal ribs. Shear stress could be transferred and dispersed by the junction, and the integral geogrid structure becomes stabilized. Introduction of Basic Properties of Geogrids Testing of geogrids mainly aims at mechanical properties, including rib tensile strength and junction

80

Figure 5. The experimental flow chart. Journal of Advanced Materials

Figure 6. The sample of single rib test. clamp to fasten; however, a roller type clamp is applied for soft geogrids. Samples are shown in the Figure 6. The specimens have to be maintained at a relative humidity of 65±5% with a temperature of 21±2°C for 12 hours before testing. The tension rate is 5 mm/min. Each groups are tested at least 10 samples.

Figure 7. The sample of junction test. Besides, it should be note that these three samples had greater elongation in a yarn than in the grids. The external braided PP layer had a flexible length while being stretched. It was discovered in the S-S curve that irregular variation occurred in the latter part of the curve since the external PP layer began to break. Recurrence of a shortterm smooth range until breakage was the elongation of stainless steel filament.

Geogrid Junction Strength Geogrid junction strength is based on GRI-GG2 8. Generally speaking, junction tensile strength is an important index for the integral tensile strength of a geogrid. Specimens need to be cut into T shape as shown in Figure 7. At least 10 specimens are required for each group, which are placed in the clamp of a tension tester of constant rate extension. The specimens have to be maintained at a relative humidity of 65±5% with a temperature of 21±2°C for 12 hours prior to testing. The tension rate is 50 mm/ min.

Comparison of Stainless Steel/PP Grids Tensile Properties Rib Tensile Strength of Stainless Steel/PP Grids The rib tensile strength of Stainless Steel/PP grids was conducted in Figure 9. The results showed that the BSY

Results and Discussion

The influence of different core material types on the tensile properties of Stainless Steel/PP ply yarn As shown in Figure 8, three samples of different types were designed in this experiment to investigate the influence of stainless steel filament, including SSF, TSF and BSY, on the mechanical properties of the ply yarn. It was found from the stress-strain curve (S-S curve) that the strength of SSF for core yarn was about 7 kg/cm2; however, that of the other two ply yarns were over 10 kg/ cm2 and the braided sample had the best strength, which was because BSY ply yarn have been twisted during braiding and the content of stainless steel filament was higher than SSF and TSF.

Special Edition No. 1, 2006

Figure 8. The S-S curve of ply yarn. 81

Figure 9. The S-S curve of rib tensile strength for three type core yarns' grid.

Figure 10. The S-S curve of junction strength for three type core yarns' grid. of PP provided the most strength against the tensile stress and the improvement of junction might need the reformation of weaving method.

achieved the best performance. Single rib tensile strength was around 23 kg/cm2, the tensile strength of TSF was about 19 kg/cm 2 and that of SSF was 10 kg/cm 2 approximately. The results could contribute to that melting the external PP layer, which wrapped around stainless steel filament and bond the ribs in the geogrids. When PP was melted, it dripped down along stainless steel filament and braided it, which resulted in irregular shape on the surface of the stainless steel filament and prevented PP from dripping. As a rough surface was formed between PP layer and stainless steel filament, it would not be easy to pull out stainless steel filament and damage could be prevented. Likewise, the feeding of TSF resulted in irregular arrangement, which rendered to higher strength. As PP dripped down easily while being melted and could not wrap around stainless steel filament, SSF slipped easily and resulted in worse strength.

References

1. R. Carl and Chermant, Jean-Louis, "Damage Mechanics Applied to Concrete Reinforced with Amorphous Cast Iron Fibers," Cement and Concrete Plys, 21, 197, (1999). 2. S. Furlan Jr., João Bento de Hanai, "Shear Behaviour of Fiber Reinforced Concrete Beams," Cement and Concrete Plys, 19, 359 (1997). 3. J.K. Lee and J.H. Lee, "Nondestructive Evaluation on Damage of Carbon Fiber Sheet Reinforced Concrete," Ply Structures, 58, 139 (2002). 4. P.S. Son, "Application of Fiber Concrete," Modern Construction of Civil Engineering and Architecture Monthly, 55(1991). 5. J.H. Lin, C.W. Lou, C.K. Lu, and W.H. Hsing, "Processing of Thermoplastic Plys Produced by Polypropylene Nonwoven Selvage," Journal of Advanced Materials, Vol. 36, No.1, 57 (2004). 6. J.H. Lin, C.W. Lou, C.K. Lu, and W.H. Hsing, "Functional Fabric of Hybrid Stainless Steel/Polypropylene and the Electrical Properties of Thermoplastic Plys", Journal of Advanced Materials, Vol. 36, No.1, 63 (2004). 7. C. W. Xhe, "The Introduce of Geomaterials Test," Journal of Sino-Geotechnics, No. 71., 13 (1998). 8. C.C. Wu, "Processing Technique of PP/High Tenacity PET Hybrid Core Yarn Manufactured Geogrids," Master Thesis of Textile Engineering Institute, Feng Chia University, 15(2003)

Junction Strength of Stainless Steel/PP Grids As shown in Figure10, there was no significant difference in junction strength of these three samples because PP was adhered at junctions after being melted. One note was the junction strength of the Stainless Steel/PP grid made of SSF was lower because wrapping effect was not good enough, which resulted in incomplete adhesion at junction.

Conclusion

In the study, we successfully braided three kinds of ply yarns and weaved into functional grids. We expect that the functional grids can be applied for reinforcing concrete in the future work. The results showed that the braid stainless steel yarn (BSY) exhibited desirable tensile strength than others, which implied the braiding process could effectively improve the performance of the ply yarn. Besides, the junction strength of each sample showed no significant difference. This result indicated that the bonding 82

Journal of Advanced Materials

INSTRUCTIONS TO AUTHORS (REVISED FEB 06)

The Journal of Advanced Materials is a refereed journal. All papers submitted for publication should be original. All papers will be peer reviewed, with comments returned to the author for necessary corrections or modifications. 1. A typical paper should contain the following: Abstract (100 words or less); Keywords line (see SAMPE Web Site, www.sampe.org for official list); Introduction (citing the work's significance and its relationship to the work of others); Experimental (discussion of the materials and processes and give a description of testing techniques); Discussion of Results (which should include an explanation of all assumptions); and Conclusions. 2. Papers are to be written in English using good grammar and style. (Translation and major language revision services are not available.) Papers that are not well written may be rejected. 3. Measurements should be given in metric (SI) units. 4. The manuscript is to be written in MS Word, double spaced, full page width, formatted for only one side of each 22 x 28 cm (8.5 x 11 inch) sheet in Arial 12-point font. Pages should be electronically numbered. (No other word processor text will be accepted.) 5. The title of the article, names of all authors, and abstract should be on a separate page before the body of the text. 6. References should be cited in the text by Arabic numbers as superscripts. 7. References should be on a separate page following the text pages. The references should use the Chicago Manual of Style for documentary notes. The following format is typical for a cited journal article: 1. R.M. Baucom and J.M. Marchello, "Powder Curtain Prepreg Process," Journal of Advanced Materials, 37 (July 2005): 17-22. 8. It is especially important that the artwork be of publication quality. Some specifics regarding figures, graphics and text are: · Each figure and table should be on a separate page and these should follow the reference page. · Do not embed the figures and the tables within the text. · Each figure must be submitted in a separate TIFF, EPS or JPEG file. · If the figure is scanned, the minimum resolution shall be at least 350 dots-per-inch (dpi) with 600 dpi preferred for black-and-white line art. · Figures will be reproduced in black-and-white and not color. Therefore, be sure to use contrasting patterns rather than colors in the figures, graphs, charts, etc. 9. Refer to figures as: Figure 1, Figure 2, etc. and include all captions. Refer to tables as: Table 1, Table 2, etc. and include all titles. 10. All equations should be on a separate page and should each be in a separate TIFF, EPS, or JPEG format or made using MS Word Equation Editor. 11. The paper is to be submitted by email to the SAMPE Publications Office: [email protected]

Annual Subscription Rates: SAMPE Members ­ $30 Non-Members ­ $70 Institutions ­ $150 (Add $15 for outside North America mailing or $8 for Canada/Mexico mailing.)

Payment: (U.S. Funds Only) `Check `Visa `MC `AmEx

Credit Card #___________________________________Exp_________Signature___________________________________ Name________________________________________________ ` Member No.______________________ ` Non-Member Company/Organization__________________________________________________________________________________ Division/Dept/P.O.Box___________________________________________________________________________________ Address______________________________________________________________________________________________ City__________________________________State__________Zip________________Country__________________________

SAMPE® 1161 Park View Drive, Suite 200, Covina, CA 91724-3751 · 626/331-0616 · Fax 626/332-8929 Web Site: www.sampe.org

Special Edition No. 1, 2006 83

Journal of

Advanced Materials

"Documenting Creative Advances in Materials and Processing"

General Topics of Interest

The Journal of Advanced Materials (JAM) is published by the Publications Department of the Society for the Advancement of Material and Process Engineering (SAMPE). The Journal of Advanced Materials (JAM) is published quarterly (January, April, July, October) and regularly reports on the advancement of materials technology and process engineering aspects necessary to utilize advanced materials. The JAM welcomes articles in the following areas: · Advanced materials technology · Design and analysis of composites · Fibers, reinforcements and advanced textile materials and structures · High performance polymer matrix composites · Joining and adhesives technology · Advanced coatings and sealants · Sandwich construction and core materials · Impact damage and repair to composites · Test methods for advanced materials · Properties of advanced materials · Aerospace vehicles · Non-aerospace applications such as: medical, sports, antiterrorism technology, automotive, electronic materials, infrastructure, marine, energy, UCAV and UAV, fire safety materials, fuel cell materials · Smart materials · Nanomaterials and nanocomposites · Metal matrix composites · Ceramic matrix composites · Geopolymers and inorganic polymers · Innovative processing of advanced materials · Materials and processing (M&P) applications · Tooling methodology for processing · Process modeling and analysis · Globalization of manufacturing

The JAM classifies submissions as:

· Regular Papers: Full-length manuscripts that deal with new and original research work. · Reviews: Lengthy papers (often invited) which cover a specific area of research in great depth. · Technical Notes: Short manuscripts of 1-2 pages that represent a research highlight or brief summary of progress. · Selected papers from conferences and symposia sponsored by SAMPE.

Special Edition No. 1, 2006

84

Journal of Advanced Materials

Information

April 2005 JAM

84 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

20525


You might also be interested in

BETA
e news_August.p65
Brochure-Pamav-Substation-2010.cdr
INFLUENCE OF WATER/CEMENT RATIO IN MORTARS CONSTITUTED BY KAOLIN WASTES