Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Chemically bonded composite SOL-Gel ceramics : a study of alumina-phosphate reaction products Moorlag, Carolyn 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2000-0496.pdf [ 12.3MB ]
Metadata
JSON: 831-1.0078780.json
JSON-LD: 831-1.0078780-ld.json
RDF/XML (Pretty): 831-1.0078780-rdf.xml
RDF/JSON: 831-1.0078780-rdf.json
Turtle: 831-1.0078780-turtle.txt
N-Triples: 831-1.0078780-rdf-ntriples.txt
Original Record: 831-1.0078780-source.json
Full Text
831-1.0078780-fulltext.txt
Citation
831-1.0078780.ris

Full Text

CHEMICALLY BONDED COMPOSITE SOL-GEL CERAMICS: A STUDY OF ALUMINA-PHOSPHATE REACTION PRODUCTS By Carolyn Moorlag B.Sc. (Honors) Department of Chemistry University of Victoria, Canada, 1997 Thesis Submitted in Partial Fulfillment of the Requirements for the Degree of M A S T E R O F SCIENCE in the Department of Metals and Materials Engineering We accept this thesis as conforming to the required standard T H E UNIVERSITY O F BRITISH C O L U M B I A August, 2000 © Carolyn Moorlag, 2000 UBC Special Collections - Thesis Authorisation Form Page 1 of 1 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I further agree that permission for extensive copying of t h i s t h e s i s for s c h o l a r l y purposes may be granted by the head of my department or by hi s or her representatives. It i s understood that copying or p u b l i c a t i o n of t h i s t hesis f o r f i n a n c i a l gain s h a l l not be allowed without my written permission. The U n i v e r s i t y of B r i t i s h Columbia Vancouver, Canada Date http://www.library.ubc.ca/spcoll/mesaum.htail 9/8/00 11 ABSTRACT Chemical bonding is an alternative bonding process to sintering and fusion and can usually be carried out at relatively low temperatures. This study focuses on the analysis of the aluminum phosphate reaction products formed by reaction of alumina and phosphates (phosphoric acid or monoaluminum phosphate) in low-temperature (100-600°C) systems to form chemically-bonded ceramics. The main objectives are the identification of products formed, determination of the influence of alumina sol phases, and characterization of some chemically-bonded coatings. Characterization studies are carried out by X-ray diffraction, scanning electron microscopy, electron dispersive spectroscopy, infrared spectroscopy, and magic angle spinning nuclear magnetic resonance spectroscopy. It is found that two product types, aluminum orthophosphates (AIPO4) and aluminum metaphosphates (A1(P03)3), are formed by separate reaction sequences. Aluminum metaphosphates (meta-A and meta-B) are formed from monoaluminum phosphate via aluminum triphosphate hydrate (ATH), and are favored by high phosphate-loading conditions (A1/P<1) and phosphate reaction with heat-treated (500°C) alumina sol. Aluminum orthophosphates form directly and are favored by low phosphate loading conditions (A1/P>1) and phosphate reaction with hydrated alumina sol phase. Analysis of morphology of mixed aluminum phosphate products generally shows 50-100 urn particles composed of alumina and AIPO4 phase with clusters of fine particles of phosphate rich phases (ATH, meta-B, meta-A) surrounding. When monoaluminum phosphate is used as a phosphate source rather than phosphoric acid, the reaction products are affected less by the Al/P ratio or the hydration of alumina sol. Heat-treated alumina sol phases, composed of either microcrystalline boehmite or y-alumina, are shown to affect the type and crystallinity of the aluminum phosphate reaction products. M A S N M R experiments are used to identify and quantify aluminum phosphate products formed. Microhardness of composite sol-gel ceramics Ul impregnated with phosphoric acid is -2.5-3.0 GPa after heat treatment at 400-500°C. Decrease in microhardness with heat treatment at temperatures >400°C is correlated with crystallization of aluminum metaphosphate phases. Analysis of some ceramic coatings on metals analyzed for product content indicate that phosphate content in the coatings is low (Al/P>2). The information obtained from the study may be used to further develop energy-efficient alternatives to high-temperature firing of certain industrial ceramics. iv TABLE OF CONTENTS Abstract ii List of Figures viii List of Tables xi Nomenclature xii Acknowledgments xiii CHAPTER 1 INTRODUCTION 1 1.1 Chemically-Bonded Ceramics 1 1.1.1 Background 1 1.1.2 New Applications 2 1.2 Sol-Gel Ceramics 3 1.2.1 Sol-Gel Processing 3 1.2.2 Applications 5 1.3 Focus of Present Study 6 CHAPTER 2 LITERATURE REVIEW 7 2.1 Phosphate Ceramics and Glasses 7 2.1.1 Bonding of Ceramics and Glasses 7 2.1.2 Chemical Bonding Processes 7 2.1.3 Phosphate Glasses 8 2.1.4 Aluminum Phosphate Bond Formation 9 2.1.5 Aluminum Phosphate Reaction Products 10 2.1.6 Reactions and Variables Affecting Aluminum Phosphate Phase Formation 12 V 2.2 Sol-Gel Ceramics and Composite Systems 19 2.2.1 Sol-Gel Processing 19 2.2.2 Sol-Gel Ceramics , 21 2.2.3 Composite Sol-Gel Ceramics 23 2.3 Phosphate Bonding in Sol-Gel Systems 26 2.3.1 Sol-Gel Incorporating Phosphorus 26 2.3.2 Sol-Gel Ceramics Chemically Bonded with Phosphate 28 CHAPTER 3 SCOPE AND OBJECTIVES 31 3.1 Scope of Investigation 31 3.2 Objectives 32 CHAPTER 4 EXPERIMENTAL METHODOLOGY 33 4.1 General Procedures 34 4.1.1 Materials 34 4.1.2 Analysis 34 4.2 Preparation and Characterization of Sol-Gel Ceramics 36 4.2.1 Alumina Sol-Gel 36 4.2.2 Composite Sol-Gel 36 4.3 Processing and Characterization of Phosphated Materials 37 4.3.1 oc-Alumina and Phosphate 37 4.3.2 Monoaluminum Phosphate 38 4.3.3 Alumina Sol-Gel and Phosphate 38 4.3.4 Composite Sol-Gel and Phosphate 39 4.3.5 Phosphate Impregnation 40 4.3.6 Coatings 41 vi CHAPTER 5 EXPERIMENTAL RESULTS AND DISCUSSION 42 5.1 a-Alumina Reacted with Phosphate 43 5.1.1 a-Alumina and Phosphoric Acid 44 5.1.2 Gel-Casting versus Slip-Casting 45 5.2 Heat Treatment of Monoaluminum Phosphate 47 5.3 Hydrated Alumina Sol Phases Reacted with Phosphate 50 5.3.1 Alumina Sol and Phosphoric Acid 50 5.3.3 Composite Sol-Gel and Phosphoric Acid 53 5.4 Heat Treated Alumina Sol Phases Reacted with Phosphate 55 5.4.1 Alumina Sol-Gel 55 5.4.1.1 Heat Treatment 55 5.4.1.2 Reaction with Phosphoric Acid 56 5.4.1.3 Reaction with Monoaluminum Phosphate 61 5.4.2 Composite Sol-Gel 62 5.4.2.1 Heat Treatment 62 5.4.2.2 Reaction with Phosphoric Acid 63 5.4.2.3 Reaction with Monoaluminum Phosphate 70 5.5 Sol-Gel Phases Impregnated with Phosphoric Acid 71 5.5.1 Heat Treated Composite Sol-Gel 71 5.5.2 Aluminum Phosphate Coatings 75 CHAPTER 6 SUMMARY AND CONCLUSIONS 77 6.1 Summary 77 6.2 Conclusions 80 vii CHAPTER 7 RECOMMENDATIONS FOR FUTURE WORK 84 APPENDIX I CALCULATION OF CRYSTALLITE SIZE 85 APPENDIX II SUPPLEMENTARY INFRARED SPECTRA 86 APPENDIX III SUPPLEMENTARY X-RAY DIFFRACTION SPECTRA 88 APPENDIX IV SUPPLEMENTARY MAS NMR SPECTRA 89 APPENDIX V MICROHARDNESS VALUES AND ESTIMATED ERROR 90 APPENDIX VI OTHER DATA 93 REFERENCES 94 viii LIST OF FIGURES Figure 1.2-1 Schematic of sol-gel processes and ceramic products 4 Figure 2.1-1 Tetrahedral phosphate units of phosphate glasses 9 Figure 2.1-2 Polymorphs and high-low inversions of aluminum orthophosphate 11 Figure 2.1 -3 Summary of some alumina-phosphate reactions cited in literature 18 Figure 2.2-1 Hydrolysis and peptisation sequence of aluminum isopropoxide in water ... 20 Figure 2.2-2 Schematic of how sintering occurs between ceramic particles in a CSG system 26 Figure 2.3-1 Some alumina sol-gel phosphate reactions cited in literature 28 Figure 2.3-2 Chemically-bonded CSG coating process 29 Figure 2.3-3 Schematic of how bonding occurs between ceramic particles in a CB-CSG system 31 Figure 5.1-1 (a) IR spectra of a-alumina and 1:1 oc-alumina and phosphoric acid heat-treated between 200-350°C, and (b) X R D spectra of 1:1 a-alumina and phosphoric acid heat-treated between 100-500°C 42 Figure 5.1-2 S E M images of 1:1 a-alumina and phosphoric acid fired at (a) 200°C (b) 300°C, and (c) 500°C 45 Figure 5.1-3 Microhardness results for 11:1 phosphated a-alumina at room temperature and fired at 100, 300, and 500°C after (a) gel-casting, and (b) slip-casting ... 46 Figure 5.1-4 S E M micrographs of (a) slip-cast and (b) gel-cast 11:1 a-alumina (0.4 lum) and phosphoric acid fired at 500°C 46 Figure 5.1-5 EDS analysis of phosphate content in (a,c) gel-cast, and (b,d) slip-cast 11:1 a-alumina (0.4lum) and phosphoric acid fired at 500°C by elemental peaks and phosphate distribution map 47 Figure 5.2-1 X R D spectra of monoaluminum phosphate (a) heated for 1 hour between 200-600°C and (b) fired at 300°C from 15 minutes to 4 hours 48 Figure 5.2-2 S E M images of monoaluminum phosphate fired for 1 hour at (a), (b) 300°C and(c)500°C 49 Figure 5.3-1 TR. spectra of (a) monoaluminum phosphate, and (b) 1:1 alumina sol and phosphoric acid fired at temperatures between 200-350°C 51 ix Figure 5.3-2 X R D spectra of 1:1 wet-mixed alumina sol and phosphoric acid fired at 80°C, 200°C, and 350°C for 1 hour 52 Figure 5.3-3 S E M micrographs of (a) 1:1 alumina sol and phosphoric acid, mixed wet and fired at 200°C, and (b) gel-cast, dried alumina sol 52 Figure 5.3-4 Hardness measurements for CSG and 11:1 CSG to phosphoric acid heat treated at temperatures between 85-600°C 54 Figure 5.3-5 (a) S E M and (b) EDS (white dots are elemental P) measurements of wet-mixed 11:1 CSG and phosphoric acid, heat-treated between 85-600°C 54 Figure 5.4-1 (a) Air-dried alumina sol with peaks corresponding to boehmite, and (b) alumina sol at different firing temperatures 55 Figure 5.4-2 (a) X R D spectrum of 1:2 alumina sol (500°C) and phosphoric acid fired for 1 hour at 100-500°C and (b) SEM image after firing at 400°C 57 Figure 5.4-3 X R D spectra of 1:1 alumina sol (500°C) and phosphoric acid (a) fired for 30 minutes at 200-500°C, and (b) fired at 300°C from 5 minutes to 2 hours 59 Figure 5.4-4 X R D spectra of (a) 1:1 alumina sol (500°C) and monoaluminum phosphate fired for 30 minutes at 200-500°C and (b) fired at 300°C from 5 minutes to 2 hours 61 Figure 5.4-5 S E M images of 1:1 alumina sol (500°C) and monoaluminum phosphate fired for 5 minutes at 300°C 62 Figure 5.4-6 X R D spectra of 1:1 oc-alumina, alumina sol (500°C), and CSG (500°C) reacted with phosphoric acid and heat-treated at 500°C 63 Figure 5.4-7 3 1 P and 2 7 A l M A S N M R spectra of 1:1 a-alumina, alumina sol (500°C), and CSG (500°C) reacted with phosphoric acid and heat-treated at 500°C 66 Figure 5.4-8 S E M images of 1:1 (a) a-alumina, (b) alumina sol (500°C), and (c) CSG (500°C) reacted with phosphoric acid and heat-treated at 500°C 67 Figure 5.4-9 X R D spectra of 2:1 CSG (200°C) reacted with (a) phosphoric acid, and (b) monoaluminum phosphate and fired for 1 hour at 200-500°C 68 Figure 5.4-10 X R D spectra of (a) 2:1 CSG (500°C) and phosphoric acid and (c) CSG (500°C) and monoaluminum phosphate, heated for 1 hour between 200-500°C; (b) and (d) heat-treated at 300°C from 15 minutes to 4 hours with the same compositions respectively 69 X Figure 5.5-1 Analysis of the penetration layer of CSG (200°C) impregnated with phosphoric acid by (a) S E M and (b) elemental phosphorus (light dots) EDS map 72 Figure 5.5-2 Hardness values versus depth of penetration for 200°C and 500°C fired CSG samples impregnated with phosphoric acid and fired at 500°C 72 Figure 5.5-3 Hardness values for 200°C and 500°C fired CSG samples impregnated with phosphoric acid and fired at temperature between 80-600°C 73 Figure 5.5-4 Analysis of the penetration layer of CSG (500°C) impregnated with phosphoric acid by (a) S E M and (b) elemental phosphorus EDS map 74 Figure 5.5-5 X R D spectra of (a) CSG on steel, fired at 300°C, phosphated with phosphoric acid and fired at 300°C, and (b) CSG on steel, fired at 300°C, phosphated with monoaluminum phosphate and fired at 300°C 76 LIST OF TABLES Table 4.3-1 Alumina-Phosphate Experiments 41 Table 5.4-1 Peak assignments for 3*P MAS NMR spectra 64 Table 5.4-2 Peak assignments for 2 7 A l MAS NMR spectra 64 Table 5.4-3 Relative amounts of alumina and aluminum phosphate reaction products of alumina (500°C heat-treated) and phosphoric acid reacted in an 1:1 ratio at 500°C. Results are calculated from 2 7 A l MAS spectra 65 NOMENCLATURE Latin Symbols b inherent instrumental broadening in radians (Eqn AI) B breadth at half intensity maximum of peak in radians (Eqn AI) D median particle diametre K shape factor (Eqn AI) Greek Symbols B x crystallite size broadening (Eqn AI) X wavelength of incident X-rays (Eqn AI) x mean crystallite dimension Abbreviations A P H aluminum pyrophosphate hydrate (A1(HP207)-2.5H20) AIPO4 -B berlinite aluminum orthophosphate (A1P04) AIPO4-T tridymite aluminum orthophosphate (AIPO4) AIPO4-C cristobalite aluminum orthophosphate (AIPO4) A T H aluminum triphosphate hydrate (A lH 2 P 3 OioH 2 0) ATH2 aluminum triphosphate hydrate-2 (AlH 2P 3Oio-2.5H 20) CBC chemically-bonded ceramic CB-CSG chemically-bonded composite sol-gel CSG composite sol-gel M A P monoaluminum phosphate (A1(H2P04)3) M A S N M R magic angle spinning nuclear magnetic resonance N M R nuclear magnetic resonance S E M scanning electron microscope X R D x-ray diffraction xin ACKNOWLEDGMENTS I would like to thank Tom Troczynski and Quanzu Yang for providing an interesting project, and for guidance, and George Oprea for his knowledge of phosphates. I also thank Mehrdad Keshmiri for sharing knowledge of ceramics with me. People who carried out experiments which contributed to this work, and whom I would like to thank, are Scott Wilson at NRC, and J e r r y Bretherton and Colin Fyfe in the Chemistry Department. I also appreciated Alcan International providing me with the Alcan Fellowship, and financial assistance from the Cy and Emerald Keys Scholarship. 1 CHAPTER 1 INTRODUCTION Chemically-bonded ceramics develop ceramic bonds through chemical reaction rather than by "classic" sintering or fusion processes. Chemical bonding is advantageous compared with the traditional process of sintering because it usually can be carried out at lower temperatures. Sol-gel ceramics have special properties due to the ability to tailor the ceramic microstructure during processing, and high reactivity of sol precursors. The use of ceramic sol with a ceramic powder matrix allows decrease in processing temperatures. Low-temperature chemically bonded ceramics which incorporate sol-gel processing have the potential to provide an energy-efficient alternative to high-temperature firing in some industrial processes. Characterization and investigation of novel areas of chemically-bonded ceramics is carried out in this work with the aim of better understanding the advantages of chemical bonding in sol-gel systems. 1.1 Chemically-Bonded Ceramics 1.1.1 Background In ceramics, high-strength bonding is traditionally obtained via atomic diffusion during high temperature (>1000°C) sintering of powders. The high strength and inert characteristics of ceramics can be attributed to these strong bonds. Bonding can also be achieved during reaction with a chemical species at significantly lower temperatures. This type of binding action is referred to as "chemical bonding" and ceramics which are processed in this way are called chemically bonded ceramics (CBC's). A type of chemical bonding has traditionally been used for low-strength ceramics such as cements, mortars, and concrete. Materials which are cementitiously bonded are characterized by porosity and susceptibility to wear. However, due to the simplicity with which these materials may be mixed, formed, and set, their use has been widespread and on a large scale. 2 Chemical bonding is used in refractories in order to improve material properties and lower the firing temperature. Acidic phosphates mixed with various metal oxides react to produce chemically-bonded refractory material. The reaction of aluminum oxide and phosphoric acid to produce aluminum phosphate phases is especially found to be useful in producing high strength refractory products. These materials must be fired in order to achieve ceramic bonds, but heat treatment temperatures can be decreased when aluminum phosphates are formed. Chemically-bonded refractory products can be formed by the reaction of aluminum oxide (alumina) and an activated form of phosphate to yield aluminum phosphate products. These ceramics, formed at relatively low temperatures, are relatively dense, strong, and insoluble. The products of reaction of alumina and phosphoric acid act as a binder for refractories, and the resulting material has improved strength, elasticity, abrasion resistance, and thermal and dimensional stability [1]. 1.1.2 New Applications In the last 20-30 years, there has been development in new areas of chemically-bonded ceramics. Specialized applications for CBC's have been developed as a result of new requirements in the high-technology industry, improvements in the processing methods, and better understanding of these materials. For example, the berlinite polymorph of aluminum orthophosphate (AIPO4), which is an analog to a-quartz, has piezoelectric characteristics similar to quartz. Anhydrous A IPO4 could also be formed into a zeolite-type structure. The material is therefore being studied for sensor applications, and for use as catalytic supports or for gas separation. Other applications include use as porous membranes and thermal barrier coatings [2]. An area of chemically-bonded ceramics which has been gaining attention is that of ceramic binders, especially phosphate binders. Phosphate binders may be used to bind fine ceramic 3 particles to form bulk materials. Phosphate reinforcement binders have been developed for metal-matrix composites [3]. To prepare the composite, a binder solution is first mixed with ceramic reinforcement material, then formed into a preform and heat-treated at temperatures between 400-900°C. Subsequent metal infiltration results in a material of improved mechanical properties. Other similar aluminum phosphates have been developed as ceramic binders and/or hardeners. Processing generally consists of reacting an aluminum compound with a reactive phosphorous reagent [4,5]. Primary characteristics of these materials are low-temperature processing, rigidity, and water-resistance, inciting additional applications as sealing materials [6]. 1.2 Sol-Gel Ceramics 1.2.1 Sol-Gel Processing Solution-gelation, or sol-gel, refers to a process where a colloid is formed typically by hydrolysis of organometallic compounds, then an interlocking gel phase is formed. The colloidal solution is referred to as the sol, while the interlocking phase is called a gel. A gel may be defined a partially coagulating mass of intertwining "filaments" which may enclose the whole of the dispersed medium to produce an easily deformable pseudo-solid. Hydrolysis in solution which does not result in colloid formation but does result in gelation may also be referred to as sol-gel processing. Sol-gel processing of ceramics has advantages of high purity, homogeneity, good control of properties, and easy shaping. Since metal oxides are generally formed from organometallic reagents in an activated state, the process is ideally suited to the production of oxide ceramics [7]. Metal oxides which are most commonly produced from organometallic precursors via the sol-gel process are Si02, AI2O3, and Zr02. 4 Sol-gel processing is adaptable due to the ability to control the process at the earliest stages of processing and throughout. The properties of the precursor are important, affecting both the method of processing, and the economic viability of the product, since the organometallic precursors used in sol-gel processing are relatively expensive. Chemical control can be maintained during each step of conversion to oxide products. Solubility of the colloid as well as steps such as hydrolysis and peptisation are pH-dependent. T h e i r P r o d u c t s Figure 1.2-1 Schematic of sol-gel processes and ceramic products. Adapted from [8]. Sols and gels may be processed to produce a wide variety of products. Sol-gel processing has the advantageous ability to form ceramics ranging from very dense films and bulk ceramics, to aerogel-derived foams. Sol-gel processing begins with formation of the sol which is then 5 either deposited as a coating, spun as fibres, set in a mold to undergo gelation, or uniformly precipitated as sol-gel particles. Gel contains a substantial amount of solvent which can be evaporated to form a xerogel. Significant shrinkage takes place during this step. Heat treatment of xerogels results in dense ceramic products. If the solvent is extracted without shrinkage, a porous aerogel is produced, which may be used to form a variety of foam products. A summary of sol-gel processing and the various products formed is shown in Figure 1.2-1 [8]. 1.2.2 Applications The gelled framework formed in the sol-gel process is useful for forming into a variety of shapes and allows easy application. However, the production of bulk ceramic materials directly from gel is difficult since solvent evaporation from wet gel to xerogel involves substantial shrinkage, which is usually accompanied by cracking and warping unless evaporation is slow. The slow shrinkage required for retention of dimension leads to long processing times. Sol-gel applications have therefore primarily involved development of (1) thin films and (2) membranes, since in these cases processing involves little shrinkage. Sol-gel coatings are produced for use as films for electronic and sensor applications, and as protective coatings. Protective sol-gel coatings are used to improve wear and scratch resistance of frequently abraded surfaces, and to impart heat resistance and corrosion resistance to metals [1, 9]. The production of membranes from sol-gel is advantageous because fine microstructure such as porosity may be carefully controlled. Composite ceramic systems where sol-gel is utilized as an additive have also recently been developed. Composite sol-gel ceramic systems, termed composite sol-gel, incorporate fine ceramic particles as a majority filler phase, and sol-gel as a minor phase [10, 11]. In these composite systems, sol phase disperses between ceramic particles and there is enhanced sintering at lower firing temperatures. Ceramic coatings combining both composite sol-gel 6 technology and chemical bonding have also been developed which are aimed at application in the automotive and aerospace industries. Novel protective coatings have been developed which are deposited onto metals and chemically-bonded at the surface by the application of phosphate under low-temperature firing conditions. These coatings display low porosity and improved hardness. [12]. 1.3 Focus of Present Study While studies have been carried out on the processing parameters, material properties, and applications of phosphate-bonded CSG material [13,14, 15], studies have not been carried out to date analyzing the aluminum phosphate reaction products formed in these systems and the mechanisms of formation. It is the focus of this study to characterize the formation of aluminum-phosphate phases of alumina-alumina CSG systems which have been chemically bonded by phosphate. Products of alumina ceramics and phosphating agents which are formed at temperatures below 600°C are analyzed. Studies are carried out to (1) identify the products formed in fine alumina and alumina sol systems reacted with phosphate at low temperature, and (2) investigate the effect of different processing variables on phase formation. Generally, the low-temperature processing capabilities of these materials are characterized by analysis of composition and microstructure. Further understanding of the advantages of chemical bonding combined with sol-gel technology is desired so that the information may be used to further develop energy efficient alternatives to high-temperature firing in certain industrial processes. 7 CHAPTER 2 LITERATURE REVIEW 2.1 Phosphate Ceramics and Glasses 2.1.1 Bonding of Ceramics and Glasses Ceramics and glasses may be classified and distinguished by the way in which they are bonded. Crystalline and polycrystalline ceramics are characterized by mixed ionic and covalent bonds. The strength and inert characteristics of these materials can be attributed to strong bonding of the matrix. Polycrystalline materials contain weaker bonds at crystal interfaces (grain boundaries) and are therefore structurally weaker. The strength, measured by modulus of rupture (MOR) values, of aluminum oxide crystals is in the range of 550-1000 MPa, while 95% dense sintered alumina has a M O R range of 275-550 MPa. Glasses are composed of inorganic polymers bonded together in a disordered arrangement and have typically lower strength (MOR -70-100 MPa). In the final stages of traditional powder processing, sintering is carried out to form pressed powder particles into a cohesive mass. The material is heated to approximately two-thirds of the melting point, whereby mobility of the atoms or molecules causes diffusion between lattices and subsequent joining of separate particles [16]. Chemical bonding is a process where bonds are produced by chemical reaction rather than by diffusion. Chemical reaction can often proceed at relatively low temperatures, and chemical bonding is therefore a less energy-consumptive process compared with sintering. Ceramics which are processed by reaction are called chemically bonded ceramics (CBC's). 2.1.2 Chemical Bonding Processes Chemical bonds in CBC's are most commonly formed by one of three methods: (1) hydration, (2) precipitation, or (3) acid-base reactions. Portland cements are formed by 8 hydration reaction. Cements are composed of calcium silicates and aluminates which react with water to produce calcium silicate (or aluminate) hydrate. These species form crystallite fibres on the outside of grains, and the fibers interact to form an adhesion layer between particles [17]. In precipitation reactions, an insoluble product is formed from solid or liquid solution, generally by interaction of two salts. Phosphate bonding in ceramics primarily involves acid-base reaction. A ceramic material, such as aluminum oxide, acts as the proton acceptor (base) and the added reactive component is the proton donator (acid). The reactive component is a moderately to strongly acidic form of phosphate, such as phosphoric acid. Reaction occurs with the elimination of water, and the reaction product is formed. The final material is usually in the form a siliceous or aluminous phosphate. 2.1.3 Phosphate Glasses The formation of phosphate glasses occurs as a polymerization process. Glasses are composed of long-chained inorganic bonded together to form a disordered solid. Glass formation generally occurs by condensation of molecular precursors to form long-chained macro-molecules. Phosphate glass formation occurs by condensation of phosphate (PO3) groups. Phosphate glasses may be composed of varying phosphate chains, but can be divided into classes of (1) polyphosphate, (2) metaphosphate, or (3) ultraphosphate glasses [18]. A l l phosphate glasses consist of phosphate tetrahedra linked though bridging oxygens _and also attached to non-bridging oxygens. Classifications are based on phosphorus to oxygen ratios; examples of phosphate glass classes are depicted in Figure 2.1-1. Polyphosphates (0/P>3) are short-chained phosphate glasses containing end, terminal oxygens. Metaphosphates (0/P=3) are very long-chain or ringed structures. Finally, ultraphosphates (0/P<3) contain branching 9 chains, and doubly bonded terminal oxygens. In the chained glass structures, there is a delocalized negative charge per two non-bridging oxygens, which must be counterbalanced by the presence of a stabilizing cation. These are active sites which may be easily reacted. The addition of alumina coordinates A l 3 + to oxygen sites to either cross-link the chains and form tetrahedral AIPO4, or form coordinated alumina-phosphate compounds such as aluminum metaphosphates (A1(P03)3). terminal (non-bridgins) oxygens phosphorous . i f f / • • • oxygens [0]/[P]=2.5 [O]/[P]=3.0 [0]/[P]=3.5 [O]/[P]=4.0 ultraphosphate metaphosphate polyphosphates Figure 2.1-1 Tetrahedral phosphate units of phosphate glasses. Adapted from Ref [18]. 2.1.4 Aluminum Phosphate Bond Formation Aluminum phosphate materials can be directly formed by the reaction of aluminum oxide (alumina) and an activated form of phosphate. Cold-setting bonds have been reported to form in this system [1]. When excess phosphoric acid is present with alumina, acid phosphates form. Acid phosphates are responsible for binding action in phosphates. Extensive hydrogen bonding of aluminum phosphate polyhedra occurs as the surfaces of aluminum oxide crystallites become phosphated and are capable of chemical bonding with other crystallite surfaces. Direct calorimetric studies have been carried out on the reaction of alumina with phosphoric acid [19]. Reaction was found to begin at 127°C and the majority of aluminum phosphate formation was found to occur between 127 and 427°C. Separate reaction stages were observed 10 between 510-732°C and 732-1327°C. During the final stage, crystallization of phosphates takes place, and the higher strength of moderate temperature (371-815°C) versus higher temperature (1093°C) products is attributed to devitrification. The reaction of alumina with phosphate is an exothermic acid-base reaction [1]. Mixing of phosphoric acid with an acidic oxide such as SiCb yields no reaction, while reaction with a basic oxide such as MgO is violent and gives a porous, friable product. Since alumina is an amphoteric oxide, reaction with phosphate occurs at a controllable rate even at elevated temperatures. The aluminum cation has a small atomic radii, which facilitates the formation of structures with low oxygen coordination numbers and a variable bonding structure. Amorphous products can result which are more flexible than ordered materials [1]. Aluminum phosphate bonding resulting from reaction of phosphoric acid and alumina at low temperature has been described as a dissolution process which proceeds in three steps [20]: (1) transport of reactant to the interface, (2) reaction at the interface, and (3) transport of product away from the interface. After saturation in solution, and with further attack of alumina grains, there is precipitation of aluminum phosphate compounds. Acid aluminum phosphate solutions may undergo bonding by 1) acidic metaphosphate (hydrogen) bonding, or 2) chemical (reaction) bonding [21]. At median to low temperatures (20-200°C) and under highly acidic conditions, hydrogen-bonded polymers are formed, which condense to amorphous macromolecules which crystallize to metaphosphates with an increase in temperature. During chemical (reaction) bonding, acid phosphates react with weakly basic or amphoteric oxides to yield crystalline aluminum orthophosphates. 2.1.5 Aluminum Phosphate Reaction Products Various crystalline aluminum phosphates and intermediate phosphate products have been identified in the literature [1, 20, 22, 23]. The primary phases of aluminum phosphates which 11 are formed are (1) aluminum orthophosphates and (2) aluminum metaphosphates. Additional phases which are usually formed as intermediate phases are (4) acidic aluminum phosphates and (5) hydrated aluminum phosphates. The formation of (6) amorphous aluminum phosphates also occurs. Aluminum orthophosphate, A1P0 4 , occurs in three polymorphs corresponding to quartz, tridymite, and cristobalite silica [24]. The corresponding A1P0 4 phases formed are berlinite ( A I P O 4 - B , the quartz form), tridymite A1P0 4 (AIPO4 -T) , and cristobalite A1P0 4 (AIPO4 -C) . The polymorphs may convert between structures; the temperature-stability relationships are shown in Figure 2.1-2. Conversion between high and low crystalline forms also occurs. Aluminum orthophosphates have been reported as reaction products of phosphoric acid reacted with various aluminum oxides [20, 25, 26, 27]. 707°C 1047°C A I P O 4 berlinite —» A1P0 4 tridymite —» A I P O 4 cristobalite 586°C 93°C 130°C 210°C a <-> P a <-> Pi <-> p 2 a <-> p Figure 2.1-2 Polymorph and high-low inversions of aluminum orthophosphate [24]. Aluminum metaphosphate, A1(P03)3, exists as long-chain, cross-linked polymers in two forms. Aluminum metaphosphate type A is the high temperature form, and has a ring structure of P 0 4 tetrahedra linked in (P03_)n (n=3, 4) rings [28]. The low-temperature form, type B, is composed of (PC>3~)n chains. The chains are cross-linked by aluminum atoms. Aluminum metaphosphate phases are effective bonding phases due to their ability to cross-link and polymerize. Aluminum metaphosphate has been identified as a product of phosphoric acid reacted with various aluminas [20]. As a binder, metaphosphate material provides an 12 improvement in tensile strength and temperature resistance compared with silica binders [26]. It has been suggested as the major product of phosphate-bonded high alumina refractories [1]. Acidic aluminum phosphates such as monoaluminum phosphate, A1(H2P04)3, and aluminum triphosphate hydrate, A I H 2 P 3 O 1 0 , have been identified usually as intermediate products. Acid phosphates have the ability to undergo additional reaction with alumina. It has been suggested that monoaluminum phosphate forms as an intermediate product in the binding of high alumina refractories by phosphoric acid. A cross-linked form of aluminum metaphosphate forms with subsequent heating [1]. Hydrated aluminum phosphates are usually formed as low temperature products, or intermediate phases. Hydrated aluminum orthophosphate primarily exists as AIPO4JCH2O (1<x <2). Variscite, AIPO42H2O, and other hydrated A IPO4 phases have been shown to form upon reaction of alumina and phosphoric acid at low temperatures [29, 25 ,30]. Subsequent dehydration yields anhydrous phases of aluminum orthophosphate ( A I P O 4 ) . Amorphous aluminum phosphate phases have also been reported. Thermal reaction of variscite was found to result in a polymeric product [29]. Amorphous phases have also been identified as reaction products of phosphoric acid reacted with alumina [20, 23, 29]. The structures of amorphous materials are not easily identified. 2.1.6 Reactions and Variables Affecting Aluminum Phosphate Product Formation Reactions to produce aluminum phosphates have mainly consisted of phosphoric acid treatment of various aluminum oxides. These studies have indicated that the formation of aluminum phosphate products is dependent on variables such as (1) aluminum oxide type, (2) concentration, (3) temperature, and (4) time. It was first suggested by Kingery [1] that phosphoric acid and alumina form monoaluminum phosphate, and then amorphous and crystalline phases with subsequent heating. 13 The reaction of hydrated aluminum oxide, A I 2 O 3 . X H 2 O , with 68.7% phosphoric acid (-1:1 A1:P ratio) was thought to increase the bonding ability of phosphoric acid by formation of monoaluminum phosphate, A1(H 2 P0 4 )3. With dehydration, there is initial loss of 3 moles of combined water of monoaluminum phosphate to give a composition of AhCvSP^Os-SI-bO, present as an amorphous material. At approximately 500°C there is partial crystallization followed by formation of aluminum metaphosphate [1]. These experiments were substantiated by differential thermal analysis which revealed endothermic peaks at 255°C, corresponding to loss of water, and 450°C, indicating formation of metaphosphate. Exothermic reaction during formation of amorphous phase occurs between these regions. Formation of aluminum metaphosphate was complete at 800°C. Bonding ability of phosphates was measured by transverse strength (modulus of rupture) experiments [1]. Studies relating reactivity of alumina and phosphoric acid systems to aluminum phosphate phase formation at low temperatures (150-500°C) and long reaction times (10 days) were completed by Gonzalez and Halloran [20]. Reactivity was primarily defined by surface area and phosphate concentration. A high-reactivity system corresponds to high surface area of alumina and/or low phosphate loading, since excess alumina is available in the process, which is viewed as a surface reaction. Reactivity is also dependent on the material; for example, aluminum hydroxide reacts quickly, while fused alumina reacts slowly. Crystallinity is also a factor since amorphous material is more reactive than crystalline material. It was found that at low temperature, there is initial formation of monoaluminum phosphate, A1(H2P04)3. Monoaluminum phosphate can react with excess alumina present to form aluminum orthophosphates under dry conditions (100°C). Aluminum metaphosphates, A1(PC>3)3, present at 300°C to 500°C, are likely to form due to condensation reactions of A1(H2P04)3 through intermediate phases such as aluminum triphosphate, A I H 2 P 3 O 1 0 . However, if excess alumina is 14 present, acid alumina phosphate product will react with alumina particles to form A1P0 4 . This solid-state reaction has been indicated by an exothermic peak at ~680°C. Monoaluminum phosphate formation is generally observed under 1) higher (1:1) phosphate to alumina loading conditions, and 2) lower surface area alumina conditions. Under low-phosphate loading or high alumina surface area conditions, any A1(H2P04)3 or A1(P0 3)3 formed will react with excess alumina available and form A IPO4 phases via solid-state reactions. It was observed that for very small particle size (50 nm), very high surface area conditions and using an ammonium alum derived alumina, an amorphous product was observed following the formation of A1(H 2P0 4)3 [20]. It was claimed by Bothe and Brown [30] that at low temperatures (<140°C), formation of hydrated A IPO4 phases occurs from various aluminas and phosphoric acid, followed by dehydration to A IPO4 berlinite and cristobalite phases. The starting aluminas were both anhydrous and hydrated, and reacted in a dilute 33 wt% phosphoric acid solution in an Al/P ratio of 1. Bayer process alumina of small particle size (0.2 urn) but low surface area did not react at low temperatures under these conditions. However, boehmite (0.012 (lm) reacted with H3PO4 formed four hydrated reaction products: metavariscite and variscite phases of A1P0 4 -2H 2 0, A i P 0 4 1.67H20, and A 1 P 0 4 H 2 0 . The A 1 P 0 4 H 2 0 phase becomes more prominent at the higher temperature (133°C) within this range. The structure of A1P04-1.67H20 was not well defined, while the remaining hydrates showed well-defined morphology by SEM. A trend was indicated that anhydrous aluminas (flash-calcined and alum-derived) favor the formation of A1P04-H 20 rather than the more hydrated phases. Slow dehydration (12 hr) indicated that morphology of the anhydrous products is affected by the formation of hydrates. Final phase formation is also affected; berlinite A IPO4 is favored by a hydrated alumina starting 15 material, while anhydrous alumina starting materials, which react mainly through a A I P O 4 H 2 O phase, favor cristobalite formation. In another paper by Bothe and Brown [25], reactivity of aluminas toward phosphoric acid was found to be affected by surface area, phosphate concentration, and temperature. Very low temperatures (25-90°C) were used, and the phosphate concentration varied between 0-50 wt% phosphoric acid. Aluminas with high surface area and low degree of crystallinity were found to react rapidly, as measured by monitoring heat release. Faster reaction was also observed for anhydrous aluminas, which agrees with previous research [30] indicating that product formation occurs through fewer hydrated phases. There is an increase in heat evolution with temperature. Among anhydrous aluminas, Bayer process alumina, with high crystallinity and low surface area, was the slowest reacting. Hydrated aluminas displayed an increase in reactivity with temperature and phosphoric acid concentration. With boehmite, however, above 40 wt% phosphoric acid addition, liquid reactant is unable to wet the surface due to interfacial tension and the reaction is inhibited. Also, higher temperature reactions are limited by the available surface area of the alumina. Lukasiewicz and Reed [23], upon reacting different aluminas with phosphoric acid, also found that bond phase was affected by type of alumina and firing temperature. Bayer process aluminas, with particle sizes ranging from 1 to 15 |im under dilute 10 wt% H3PO4 conditions (Al/P =20.3) were fired at temperatures between 450 to 800°C for 5 hours. Bond phase could not be predicted by surface area of aluminas under these conditions. For A16 Bayer alumina (particle size <1 urn) fired with phosphoric acid at 450°C and 525°C, a mechanically-strong amorphous phosphate bonding phase formed. At 650°C and 800°C, crystalline berlinite and tridymite AIPO4 phases are formed. This amorphous phase temperature region was not observed for low-reactivity tabular alumina. 16 In another experiment incorporating a low loading of phosphate, by Craig and Francis [27], 0.3 ujn alumina was dispersed in water with a small amount of phosphoric acid (1-9.6 wt%). An amorphous bonding phase was reported to form during heat treatment between 200-300°C. The AIPO4 phase produced partially crystallized at 500°C to what was identified as the cristobalite phase by X R D . Though flexural strength measurements were carried out, no clear correlation was made between strength and crystallization. Porosity in the ceramic was shown to decrease as the phosphate loading increased. Chiou and Chung [26] carried out an extensive study of acid-phosphate binders for the purpose of improving the elevated temperature resistance of aluminum-matrix composites. Binders were prepared from aluminum trihydroxide, Al(OH)3 and H3PO4 in P/Al ratios varying from 1 to 23 and fired at temperatures <1200°C. Starting reactant ratios where found to affect product formation, which was characterized by SEM, DTA, atomic absorption, and X R D . Binder prepared in a 1:1 ratio was found to form primarily berlinite and some A1(P0 3) 3-B at 500°C, with the A1(P0 3) 3-B being converted to A1(P0 3) 3-A at 800°C. Binder prepared in a 1:3 A1:P ratio could be dried at 200°C for one week to yield type C (hexagonal) A1(H 2P0 4)3, formed in a condensation reaction. Condensation of A1(H2P04)3 resulted in the formation of A1(PC>3)3-B. At 800°C, the material was converted to A1(P03)3-A. Release of P2O5 is substantiated by weight loss at around 1000°C. Firing of commercial monoaluminum phosphate shows a similar trend, though an intermediate aluminum triphosphate hydrate phase, AlH 2P30io(2-3)H 20, forms at 200°C before conversion to metaphosphate phases, which are shown to exhibit needle or flake-like structures. Binders prepared in much higher P /Al ratios (6-23) showed the formation of only A1(P0 3) 3-A when fired at 500°C and 800°C. At these high phosphate loading levels, as the level of phosphate is increased, the main effect is increased weight loss due to the 17 release of phosphoric acid between 500-800°C. Phases formed by H3PO4 reaction with A1 2 0 3 and with Al(OH)3 have been reported as being very similar [31]. The pertinent points from the aluminum phosphate literature consist of the following: 1. Reactivity of the AI2O3 and H3PO4 system is controlled by alumina reactivity and phosphate concentration. Alumina reactivity is determined by its surface area and alumina type. 2. At 200°C, the phosphate-rich intermediate A1(H 2P04) 3 is favored by (a) low-reactivity alumina, such as a-alumina, or (b) a high loading of phosphoric acid. 3. Aluminum metaphosphate, A1(P03)3 j phases are preferred under excess phosphate conditions. Formation is most likely occurring by condensation reactions from A1(H 2 P0 4 ) 3 . A1(P0 3) 3-B formation occurs at 500°C , which converts to A1(P0 3) 3-A at 800°C. 4. Very high phosphate conditions (P/A1=6-12) favor only A1(P03)3-A formation at 500°C -800°C. 5. Orthophosphate, AIPO4, phases are preferred under conditions of (a) high reactivity, high surface area alumina, or (b) low phosphate loading. A IPO4 phases can be produced by A1(H 2P0 4)3 or A1(P0 3) 3 phases reacting with A1 2 0 3 . 6. Very low temperatures (<140°C) and long reaction times produce hydrated A IPO4 phases, which dehydrate to form crystalline or polymeric A1P0 4 . Some significant reactions between alumina or aluminum trihydroxide and phosphoric acid are summarized in Figure 2.1-3, along with thermal decomposition of monoaluminum phosphate. 18 2^°C 500°C A1 2 0 3 JCH 2 0 + H3PO4 • A 1 ( H 2 P 0 4 ) 3 ^ > amorphous phase. A1(P0 3) 3 [1] 230°_C 500°C 800°C 120Q_C A1(H 2 P0 4 ) 3 * A l H 2 P 3 O 1 0 " ' * A1(P0 3) 3-B r \ l (P0 3 ) 3 -A > A1(P0 3) 3 glass [20] >100°,C A1(H 2 P0 4 ) 3 + A1 2 0 3 • 3A1P0 4 + 3 H 2 0 [20] A1(P0 3) 3 + AI2O3 ^ £ 3A1P0 4 [20] <140°C >200°C Al 20 3(fines) + H 3 P 0 4 A l P 0 4vtH 2 0 (l<x<2) > ^ 1 > A1P0 4 -B + A1P0 4 -C [25] 10 Al 2 0 3 (<lum) + H 3 P 0 4 4 _ _ > ° C amorphous phase 6 5 0 ' 8 | > ° ° C A1P0 4 -B + A1P0 4-C [23] Al(OH) 3 + H 3 P 0 4 5 i W £ A1P0 4-B + A1(P0 3) 3-B E D D ^ AIPO4-A + A1(P0 3) 3-A [26] 200°C 500°C 800°C Al(OH) 3 + 3 H 3 P 0 4 • A1(H 2 P0 4 ) 3 • Al(PO) 3-B • A1(P0 3) 3-A -3H 20 -3H20 1200°C p. A 1 P 0 4 - C + P 2 0 5 [26] 200°C 500°C 800°C A1(H 2 P0 4 ) 3 • AlH 2 P 3 Oio * A1(P0 3) 3-B > A1(P0 3) 3-B + A1(P0 3) 3-A commercial -2H 20 -H2O izuu^, AIPO4-C + P 2 0 5 [26] 500°C 800°C Al(OH) 3 + 6-12H 3P0 4 • A1(P0 3) 3-A • A 1 ( P 0 3 ) 3 - A -3H 20 [26] Figure 2.1-3 Summary of some alumina-phosphate reactions cited in literature. 19 2.2 Sol-Gel Ceramics and Composite Systems 2.2.1 S o l - G e l P r o c e s s i n g S o l - g e l is a m u l t i - s t e p p r o c e s s w h e r e b y a f i n e l y - m i c r o s t r u c t u r e d m a t e r i a l is f o r m e d v i a s o l u t i o n , g e l a t i o n , a n d heat t r ea tmen t s teps . A " s o l " is a c o l l o i d a l d i s p e r s i o n , w h e r e the s p e c i e s are p a r t i a l l y bu t no t c o m p l e t e l y i n s o l u t i o n . S u b s e q u e n t d e h y d r a t i o n f o r m s a " g e l " , c o m p r i s e d o f a p a r t i a l l y c o a g u l a t i n g m a s s o f i n t e r t w i n i n g s o l i d f i l a m e n t s . T h i s i n t e r l o c k i n g g e l p h a s e m a y i n c l o s e the w h o l e o f the d i s p e r s e d m e d i u m to p r o d u c e an e a s i l y d e f o r m a b l e p s e u d o - s o l i d . H y d r o l y s i s i n s o l u t i o n w i t h o u t c o l l o i d f o r m a t i o n , f o l l o w e d b y g e l a t i o n m a y a l s o b e r e f e r r e d to as s o l - g e l p r o c e s s i n g . T h e s o l - g e l t e c h n i q u e f o r c e r a m i c s h a s a d v a n t a g e s o f h i g h p u r i t y , h o m o g e n e i t y , a n d g o o d c o n t r o l o f p r o p e r t i e s , a n d is i d e a l l y s u i t e d to the p r o d u c t i o n o f h i g h p e r f o r m a n c e o x i d e c e r a m i c s [7] . S o l - g e l p r o c e s s i n g b e g i n s w i t h the f o r m a t i o n o f a c o l l o i d a l d i s p e r s i o n f r o m i n o r g a n i c o r o r g a n o m e t a l l i c p r e c u r s o r s , t y p i c a l l y i n the f o r m o f M ( O R ) x , t h e n h y d r o l y s i s o c c u r s b y c l e a v a g e o f the M - O R b o n d s to f o r m M ( O H ) x h y d r o x i d e g r o u p s . P e p t i s a t i o n u s u a l l y t akes p l a c e w i t h a d d i t i o n o f a c i d to t he s l u r r y a n d resu l t s i n the f o r m a t i o n o f c h a r g e d p o l y n u c l e a r i o n s w h i c h are s tab le i n s o l u t i o n . F o r e x a m p l e , the " A l i 3 " u n i t i s f o r m e d b y a c i d p e p t i s a t i o n o f h y d r o l y z e d a l u m i n u m i s o p r o p o x i d e i n w a t e r [ 32 ] , a c c o r d i n g to E q u a t i o n 2 . 2 - 1 : A l ( i - O C H ( C H 3 ) 2 ) 3 h Y d r o l v s i s • A l ( O H ) 3 a c i d • [ A l , 3 0 4 ( O H ) 2 4 ( H 2 0 ) 1 2 ] 7 + E q n 2.2-1 p e p t i s a t i o n T h e c l e a r , i o n i c s o l u t i o n f o r m e d is the s o l p h a s e . T h e g e l s u b s e q u e n t l y f o r m s by p o l y m e r i z a t i o n o f the s o l p a r t i c l e s , w h i c h o c c u r s t h r o u g h a c o m b i n a t i o n o f c h a r g e d e s t a b i l i z a t i o n , s o l v e n t e v a p o r a t i o n , a n d c r o s s l i n k i n g o f c h e m i c a l b o n d s to u l t i m a t e l y f o r m M -O - M l i n k e d p o l y m e r s . T h e m e t h o d o f a l u m i n a s o l p r e p a r a t i o n f r o m a l k o x i d e s is g i v e n i n a c l a s s i c p a p e r b y Y o l d a s [33 ] . D u e to the p o l y m e r i z e d n a t u r e o f g e l s , the m a t e r i a l a n d its d e r i v a t i v e s are i n h e r e n t l y a m o r p h o u s , b u t m a y b e c r y s t a l l i z e d , d e p e n d i n g o n the m e t h o d o f 20 preparation and firing temperature. During drying of gels, cross-linking increases with rising solids content of the matrix and the viscoelastic or elastic gel forms an elastic solid [34]. Adjust pH=4 Constant Stirring Figure 2.2-1 Hydrolysis and peptisation sequence of aluminum isopropoxide in water. The general process parameters for the formation of alumina sol are shown in the diagram [13]. The drying of gel to form the solid phase is an important process in the production of a sol-gel derived solid. During drying, there is development of physical or chemical linkages between gel polymers or particles [35]. These linkages first develop an elastic or viscoelastic gel state, and then an elastic solid. Mass transport occurs by two processes: convection away from the gel surface, and transport within gel pores. Drying is accompanied by substantial shrinkage in volume. In the case of a film, as the film dries it usually adheres to the surface, and volume reduction is through reduction in thickness. Shrinkage can be reduced by either reducing the volume fraction of solvent or reducing size or thickness (for films) of the wet sample. A novel approach, composite sol-gel processing (CSG), is investigated in this work. It can substantially reduce gel shrinkage; refer to Section 2.2.3. 21 2.2.2 Sol-Gel Ceramics The sol to gel transition and subsequent solids formation allows the formation of solid phases such as films, fibers, microspheres and monoliths. Thin films produced through sol-gel have the advantages of requiring little raw material, and exhibiting fewer problems with cracking due to shrinkage. Deposition as a film onto a substrate has also been of interest due to homogeneity of the sol in the liquid state, and since the starting compound is highly reactive [36]. The preparation of amorphous or partially crystalline thin films were amongst the earliest applications of sol-gel technology, and most research in sol-gel applications is still devoted to thin to moderate thickness (0.1-1 urn) sol-gel films [35]. Thin sol-gel films developed for protective coatings can be divided into the following catagories [35]: 1) optical films, 2) moisture barriers, 3) scratch and wear resistant, 4) thermal and electrical insulation, and 5) corrosion resistant. Some examples include clear silica based sol-gel films of a specific refractive index which can be applied to plastics for wear and scratch-resistance [37] and aerogel films developed as thermal insulation [38]. Due to its improved sealing properties, there has been interest in the use of sol-gel coatings as low curing-temperature, non-porous sealing films [9]. Sol gel coatings are often used to impart heat resistance and corrosion resistance to metals [1, 9, 39], and have also been suggested as alternative heat-resistant coatings for automotive components [40]. Thin film deposition involves a wetting process at the substrate, displacing air. The main methods of carrying out deposition are 1) dip-coating, and 2) spin-coating [35]. The advantage of these processes is simpler equipment requirements than for chemical vapor deposition or sputtering techniques. Control of microstructure is dependent on evaporation and capillary pressure, condensation reactions, and aggregation processes, along with the structure of the inorganic species formed. Properties obtained using sol-gel thin films for protective 22 applications are low thermal conductivity, chemical inertness, high-temperature stability, adjustable porosity, and mechanical hardness. A more recent study by Guizard et al [41] has investigated developing sol-gel materials for catalyst and inorganic membrane applications. Porous oxide solids can be prepared for which there is precise control of composition, grain structure and pore structure, now possible due to advances in the control of precursors and sol-gel processing. Supported microporous (<0.2 nm) layers have been synthesized for ceramic nanofilter applications. Improvements in sol-gel processing and control of sol to gel transition is leading to developments in oxide catalysts and surface-active agents [41]. A composite sol-gel (CSG) ceramic system has been studied by Yang [13, 15]. This composite system is termed composite sol-gel, or CSG. It incorporates fine ceramic particles as a majority filler phase, and sol-gel as a minor phase. Alumina sol was primarily combined with alumina, but also with silicon carbide. In these composite systems, it was found that sintering temperatures could be dramatically reduced. The sol phase disperses between ceramic particles and undergoes enhanced sintering at lower firing temperature. Multicomponent ceramics which incorporate zirconia and alumina sols have been developed by Livage et al. [42]. Due to problems with aggregation, condensation processes were slowed by complexation with a ligand to allow other bonds to form in the mixed-sol system. It was shown that sol-gel processing allows homogeneous mixing of precursors. Nanocrystalline particles, and homogeneous microstructures may then be produced. Ceramic coatings combining both composite sol-gel technology and chemical bonding have also been developed by Yang et al [12, 13, 15] These coatings are aimed at application in the automotive and aerospace industries. These protective coatings are applied onto metals and chemically-bonded at the surface by the application of phosphate. Chemical bonding at the ceramic surface facilitates low-temperature firing, eliminates porosity and seals the surface, and 23 improves hardness. This chemically-bonded system, which is novel in its incorporation of both phosphate bonding and sol-gel ceramics in a composite system, is of primary interest in this work. Advantages of sol-gel processing can be summarized as follows [37, 43, 44,45]: 1. High purity of starting materials 2. Homogeneity and excellent control of microstructure (with proper processing steps). 3. Flexibility of shape in processing (for both near net-shape components and films). 4. Low-temperature formation and crystallization of the material, leading to lower cost and higher purity. Disadvantages of sol-gel processing are mainly [37, 43, 44, 45]: 1. High relative cost of the precursors as raw materials. 2. Shrinkage and/or deformation during solvent removal or thermal processing 3. Multiple processing steps 2.2.3 Composite Sol-Gel Ceramics Due to the flexible nature of sol-gel processing, it may be used to produce a variety of materials containing composite phases. Types of ceramic sol-gel composite materials found in literature consist of 1) infiltration composites, and 2) oxide powder composites. Infiltration composites are formed by sol-gel material penetrating into a porous second phase. Oxide-powder composites are composites of sol-gel which has been mixed with oxides by dispersion or seeding in order to improve processing and material properties. Sol-gel routes have been cited as an alternative, novel approach to forming interpenetrating phase composites [46]. Sol-gel routes are ideal for making fine-scale, interconnected porous material. A sol-gel silica and alumina layer composite has been formed by Stoch et al. [36] using deposition by sol-gel technique and infiltration. The sol-gel silica was deposited on a 1.2 24 urn thick chemical oxidation layer of anodized aluminum. The alumina layer was found to be modified by the highly dispersed silica xerogel overlayer, which was subsequently heat treated at 500°C. The porous anodic films were found to be composed of gelatinous boehmite, AlOOH. It was shown by JR analysis that silica gel addition influences the stretching vibration of the OH-group, indicating structural change in the system. There was also some reaction to form aluminosilicate (Si-O-Al) bonds, though the silica-alumina system remains amorphous even after heating. It is concluded that silica gel penetrates the porous alumina and reacts when heated to 500°C. Ceramic-sol-gel composites may be formed by the mixing of oxide powders with a ceramic-forming sol. In a paper by Kwon and Messing [47], the shrinkage of boehmite sol is addressed by the incorporation of fine a-alumina particles. Boehmite sol prepared from 10 nm commericial (pseudo) boehmite was first seeded with a-Al203 [48], then dispersed with 70% 0.2 |im a-Al203 particles. Samples sintered at 1300°C yielded densities >95%. It was found that larger sized a-Al 2 03 particles required higher sintering temperatures and a bimodal distribution of particles decreases densification efficiency. Boehmite, AlOOH, undergoes transitional alumina states during heating (boehmite —•> Y-AI2O3 — > 8-AI2O3 —> 8-Al 203 —> a-AI2O3), with the first transformation to a crystalline y-alumina structure taking place at ~500°C and the final reconstructive phase transformation to a-Al203 occurring between 1000-1200°C. During these transitions, substantial shrinkage takes place due to density and molecular weight changes. It is indicated that by incorporation of a-Al203 particles, shrinkage is reduced by 1) less specific volume change by replacing lower density boehmite with a-A^Os (3.986 g/cm3), and 2) improved particle packing to yield increased green density. It is found that the use of this system results in better dimensional control of dense alumina ceramics, allowing preparation of materials other than fibers, films, and abrasive grains. 25 Alumina sol has been used as a deagglomerating dispersant in A I 2 O 3 and SiC powder systems [49]. It was found that alumina sol affects the oxide particle agglomerates, decreasing their size by 50% compared with dispersion in pure water. It is postulated that sol clusters of A104Al i 2 (OH )24 (H 2 0 ) i 2 7 + (1-2 nm) are absorbed onto the surface of ceramic particles. In the A I 2 O 3 and alumina sol system, particles disperse due to alumina sol-clusters and A I 2 O 3 being highly electrostatically repulsive. In the case of the silicon carbide system, SiC particles absorb alumina sol, resulting in steric repulsive forces between particles. Alumina sol in composite ceramic systems is used as both a dispersion additive and as a sintering agent to prepare high performance engineering ceramics. Barrow, Petroff, and Sayer [50, 51] developed a sol-gel based coating technology and prepared high-quality, dense, crack-free, thick (5-200 |Ltm) ceramic coatings. Ceramic powders were dispersed in a sol-gel solution and then fired above 400°C. The system has the advantages of sol-gel processing, and the coatings are not limited to very thin films due to the addition of ceramic powders. Coatings incorporating lead zirconate titanate (PZT) were investigated as potential new piezoelectric devices, and zirconia and alumina films were cited as having corrosion and wear-resistant properties, and are possible thermal barrier coatings. An alumina-alumina composite sol-gel (CSG) system has been studied in detail by Yang [13]. Hydrated alumina sols of high reactivity and small particle size (<10 nm) were used to produce ceramic bodies and coatings. The basic processing steps consisted of 1) dispersion of ceramic filler into alumina sol, 2) gelcasting, 3) drying and 4) sintering. Studies of alumina sol mixed with A I 2 O 3 , Zr02, and SiC were conducted. Advantages of both filler and sol content are observed [13]. Dispersion of ceramic filler into the sol acts to decrease shrinkage, though defect formation can still occur. Addition of alumina sol decreases the time and temperature of sintering by providing a fast diffusion path 26 for mass transport (Figure 2.2.2). Calcined alumina CSG containing 14 vol% sol could be sintered at the relatively low temperature of 1400°C to yield a material with a microhardness of 17.5 GPa. With addition of the MgO sintering additive, microhardness increases to 20 GPa. At sintering temperature, pore content was found to decrease with sintering time, with grain size remaining constant until grain growth occurred in the final stage of sintering. Sintering of 50 vol% SiC CSG at 1850°C resulted in 98% density and a 22.9 GPa microhardness. Figure 2.2-2 Schematic of how sintering occurs between ceramic particles in a CSG system. 2.3 Phosphate Bonding in Sol-Gel Systems 2.3.1 Sol-Gel Incorporating Phosphorus A few literature sources report processes where a phosphorus precursor is incorporated into a sol-gel during processing. Analysis of phase formation, and in some cases, coordination, as determined by magic angle spinning nuclear magnetic resonance (MAS NMR), were carried out. Aluminum isopropoxide hydrolyzed directly with triethylphosphate in water and organic solvent has been shown at 250°C to transform from an amorphous phase to a cristobalite-type A I P O 4 , then to berlinite [52]. An increase of water in the reaction mixture was shown to favor Particle A ~ 0.1 - 1 0 [im ~ 0.001 - 0.01 ;am Particle B ~ 0.1 - 1 0 L tm 27 immediate formation of berlinite due to the high solubility of A1P0 4-B in water, while the metastable cristobalite phase prefers organic solvent conditions. With the use of solid-state M A S NMR, the chemical states and environments of 3 1 P and 2 7 A l nuclei may be elucidated. Solid state N M R is especially useful for characterizing amorphous or semi-amorphous structures with some structural defects. Grimblot et al. [53, 54] have used both conventional and two-dimensional multiple quantum angle M A S N M R experiments to analyze hydrotreated, modified alumina sol-gel catalysts. Phosphate was incorporated by addition of a P precursor (H3PO4 or P2O5) during hydrolysis of aluminum sec-butylate (ASB). Initial studies on the poorly crystallized boehmite which forms upon drying alumina sol revealed a single, distorted Al o cta (octahedral aluminum) site, which changed to several A l o c t a and Altetra (tetrahedral aluminum) sites plus some Al p e nta (pentacoordinate aluminum) upon calcination at 500°C. The boehmite structure is unchanged up to a heating temperature of 200°C, then tranforms to y-alumina above 350°C. Catalysts containing phosphorus showed the presence of both Al tetra-0-P and Al 0 C ta-O-P sites after drying, but only Altetra-O-P, corresponding to AIPO4 formation, after calcination. When studying the effect of phosphate loading, generally it was found that under low-phosphate conditions, well-dispersed mono- or di-phosphates formed. Under higher phosphate conditions, polymeric P oxocompounds and bridged AIPO4 structures with varying degrees of hydration were formed. Differences in products due to the P precursor used were also observed; P2O5 strongly prevents hydrolysis and condensation of aluminum sec-butalate. Dried H3PO4 catalysts formed monomeric and polymeric P oxospecies which converted to A1P0 4 phases (Al t e t r a -0-P) when calcined. Catalysts prepared with P2O5 formed organic monomeric P oxospecies when dried and polymeric P oxospecies when calcined. Some of these reaction are summarized in Figure 2.3-1. 28 250°C > A I P O 4 - B [52] Al(/-OCH(CH 3 ) 2 ) 3 + P(CH 2 CH 3 ) 3 >-AlP0 4 -C Al(OCH 2 CH(CH 3 ) 2 )3 + H 3 P 0 4 (P/A1>0.2) ,< 200°C + AlOOH + A l o c t a - 0 - P + A l t e t r a - 0 - P [54] Figure 2.3-1 Some alumina sol-gel and phosphate reactions cited in literature 2.3.2 Sol-Gel Ceramics Chemically-Bonded with Phosphate Composite sol-gel (CSG) ceramics which have been chemically treated to yield non-permeable, adherent coatings on metal substrates have been prepared by Yang [13, 15] for high temperature corrosion and wear-protection applications. Alumina CSG slurry containing 14 volume % alumina sol (1.5 M) is applied on stainless steel, titanium and aluminum alloy substrates by dip-coating. The resulting layer is 20-60 urn thick. After a low-temperature firing sequence (300-500°C), subsequent sol-gel "sealing" treatments were carried out to improve the microstructure and protective characteristics of the coatings in corrosive and high-temperature environments. The coatings were chemically bonded both at the surface and the metal-ceramic interface with the use of two phosphate bonding materials, phosphoric acid (H 3 P0 4 ) and monoaluminum phosphate (A1(H2P04)3). Bonding strength between the substrate and coating has been reported to be of the order of 42 MPa, and surface hardness was around 6.5 GPa [15]. A schematic of the coating process is depicted in Figure 2.3-2. The coating is first made to adhere to the metal surface by sandblasting the metal substrate so that a thin coating of dispersed a-alumina and alumina sol-gel is able to mechanically interlock to the surface upon drying. The surface may also have been treated with phosphate. After firing, typically between 300-500°C, phosphate is applied to the surface and the firing is repeated, to the effect of the 29 phosphate hardening the surface of the alumina layer. Phosphate addition has the added effect of reacting at the substrate interface, to create a thin adhesion layer. Initially, phosphoric acid, H3PO4, was used as the binding phosphate. However, it was found that with the layer of thickness of these coatings, penetration and reactivity of concentrated phosphoric acid was so strong that it could result in corrosion of the metal substrate at the metal interface. Therefore, the less reactive monoaluminum phosphate, A1(H2P04)3, was used, which resulted in only mild reaction at the interface, but not degradative corrosion. Multiple infiltration was found to decrease gas permeability of the coatings [12, 15]. Figure 2.3-2 Chemically-bonded CSG coating process. It has been shown that low temperature chemically reacted CSG systems show evidence of densification, sealing, abrasion resistant, and corrosion resistant properties [15]. Sub-micron sol particles are incorporated between ceramic powder particles to allow faster reaction with the chemical constituent, which also binds with the powder particles. The result is extensive bonding to and between the ceramic particles which binds the system together at low temperature and allows less porosity and therefore less likelihood of attack and degradation of the substrate material. The general process is depicted in Figure 2.3-3. The phosphate/alumina Deposited C S G layer Chemically Reactive Reagent Sand-blasted, treated surface Metal Substrate 30 and phosphate/alumina-sol reaction products in these coating systems are the subject of the present work. Particle A / Particle B Surface-Reacted Sol Fuly-Reacted Figure 2.3-3 Schematic of how bonding occurs between ceramic particles in a CB-CSG system. Particle sizes: sol 1-10 nm, powder 0.1-10 pirn. 31 CHAPTER 3 SCOPE AND OBJECTIVES 3.1 Scope of Investigation Chemically-bonded ceramics (CBC) is an expanding area of research. Reactive bonding within a ceramic matrix is a lower temperature alternative to high temperature sintering in achieving cohesion of ceramics. Sol-gel technology is being incorporated into the processing of many new ceramic materials due to the control of microstructure, homogeneity, and improvements in densification that it can offer. However, few processes have combined chemical bonding with sol-gel material in a ceramic composite. Through this type of novel process, development of low curing temperature, high strength chemically bonded sol-gel ceramics, in particular ceramic coatings, is desired. The coatings should adequately seal and protect a metal substrate while adhering well to the substrate surface. The possibility of very low temperature processing of these coatings is the driving force for advancements in this area. Characterization of the aluminum phosphate products formed and the dependence of product formation on starting materials, and temperature and time of firing in these systems has not been studied. The formation of chemical bonds in these systems is also little understood. The characterization and understanding of the coating system studied in this work will aid in the advancement of chemically bonded ceramic coatings. The knowledge could also be applied to other developing areas of CBC's . The main emphasis of this study is to analyze the products formed by the reaction of phosphate (i.e. phosphoric acid or monoaluminum phosphate) with alumina particles, alumina sol, and composite sol-gel (CSG) alumina ceramics systems. A variety of characterization techniques are used. Some investigation is also carried out on the "impregnation layer" of phosphated CSG ceramics. The difference in reaction products for hydrated versus dehydrated (fired) alumina reacted with phosphate is also studied. 32 3.2 Objectives The general objective is to study the low temperature reactions between alumina and phosphate for the chemically-bonded CSG system. The study objectives can be broken down as follows: 1. Identify the products formed by reaction of fine a-alumina particles with phosphate and alumina sol with phosphate in the temperature range of 100-600°C for varying compositions, concentrations and reaction conditions. 2. Identify differences between a-alumina reaction with phosphate and alumina sol reaction with phosphate, based on products formed, reaction time and temperature. 3. Characterize the material composition of chemically-bonded CSG coatings based on the results found. The objectives are addressed by the following techniques: 1. Identification of phases are carried out primarily by qualitative analysis by X-ray diffraction (XRD). Results are compared with those found in literature. 2. Scanning electron microscopy (SEM) is used as a complimentary technique to observe the morphology of particles which have been phosphated and of model coating systems prepared on substrates. Phosphate distribution is also carried out by electron dispersive spectrometry (EDS) analysis. 3. Solid state infrared (IR) analysis is carried out to observe differences in reaction products due to change in reaction temperature. 4. Magic-angle spinning nuclear magnetic resonance (MAS NMR) spectroscopy is used to provide information on atomic composition and product ratios of the reaction products. 33 CHAPTER 4 EXPERIMENTAL METHODOLOGY The reactants consisted of an alumina, either a-alumina, heat-treated alumina sol (forming boehmite or y-alumina), or composite sol-gel (CSG), and concentrated forms of phosphate. Reaction of alumina with a reactive phosphate was termed "phosphating". Materials preparation was followed by processing treatments, and characterization. The experimental methodology chapter is divided into three sections: (1) general procedures, (2) preparation and characterization of sol-gel materials, and (3) processing and characterization of phosphated materials. The general procedures section identifies the raw materials used in alumina sol-gel preparation and lists the concentrated phosphate phases used and any other materials not prepared in this work. Instrumental information and procedures are also provided. The second section describes the preparation and firing of alumina sol-gel and composite sol-gel material. Some characterization was carried out before phosphating. The third section details the phosphating procedures for a-alumina, alumina sol, and CSG, and subsequent firing. The majority of experiments followed the sequence of (a) phosphating of fired alumina material for a specific alumina to phosphate ratio, and then (b) "post-phosphate" firing. Thermal treatment of monoaluminum phosphate was also carried out. Wet alumina sol and CSG slurry were directly mixed with phosphate and an alternate phosphate impregnation procedure studied. Samples were characterized by X-ray diffractometry, scanning electron microscopy, microhardness testing, and N M R spectroscopy. 34 4.1 General Procedures 4.1.1 Materials Unless otherwise stated, all materials were used as received. a-Alumina used was A16 S G calcined alumina from Alcoa Industrial Chemicals. Median particle diameter, D, was measured to be 0.41 um (Appendix VI. 1) and the material was shown to be mainly crystalline by X R D (Appendix JH.1). Aluminum isopropoxide (98%), Al( /OCH(CH 3 ) 2 ) 3 , was obtained from Aldrich. Reagent grade H3PO4 (85% wt) was purchased from Fisher Scientific and monoaluminum phosphate (A1(H 2P0 4)3, 6 1 % wt), grade M A IP, was obtained as a sample from Albright and Wilson Company. Chemically-bonded composite sol-gel coatings on aluminum and steel were prepared by Q . Yang, Metals and Materials Engineering Department, University of British Columbia. 4.1.2 Analysis Particle size measurements were carried out with a CAPA-700 Horiba Particle Size Distribution Analyzer and pH of solutions and slurries was determined using a 130 Corning pH meter. Infrared spectroscopy was carried out using a Perkin Elmer System 2000 FTIR. Anhydrous samples were ground to a fine powder and 3 mg of this powder were mixed with 300 mg of dry crystals of potassium bromide. The material was pressed into transparent disks, and solid-state analysis was carried out using air as the background. Powder X-ray diffraction (XRD) experiments were carried out using monochromatic CuKa(1.54nm) radiation on a D5000 Siemens Diffractometer. The material to be analyzed was applied as a thin layer on 2 cm x 4.5 cm glass plates by grinding the material with a small amount of ethanol and allowing the slurry to dry. Some low-temperature or reactive samples were prepared directly on glass diffractometer plates before analysis. 35 To observe particle-size distributions, microstructure of phosphated alumina and alumina sol materials, and characterize phosphate distribution, scanning electron spectroscopy (SEM) and energy dispersive spectrography (EDS) analysis were carried out. Slip-cast and gel-cast a-alumina/phosphoric acid samples and phosphate impregnated samples were analyzed using a Hitachi S3500N environmental scanning electron microscope with high and variable vacuum modes. Analysis was carried out at -20 Pa pressure under backscatter conditions. Energy dispersive X-ray analysis was carried out using an Oxford ISIS Series 300 with quantitative analysis and X-ray mapping. A l l other analyses of powder materials were carried out using a Philips XL30 S E M with a Princeton Gamma-Tech Prism Digital Spectrometer, EDS analysis probe, and back-scattered electron (BSE) mode. Magic angle spinning nuclear magnetic resonance (MAS NMR) experiments were conducted by J.L. Bretherton from the research group of C A . Fyfe, Department of Chemistry, University of British Columbia. Tests were performed in the Environmental Sciences Laboratory (a national scientific user facility sponsored by the U.S. DoE Office of Biological and Environmental Research) located at the Pacific Northwest National Labortory, operated by Battelle for the DoE. For 2 7 A l , the 90° solid pulse width was 1.7 p.s and the spectra were referenced to 1M aqueous A1(N0 3 ) 3 solution. 2 7 A l spectra were obtained at 208.43 MHz (18.8 T, 800 MHz for protons). The 18.8 T spectra were recorded on Varian Inova systems at the Pacific Northwest National Laboratory. A l l 2 7A1 spectra were obtained using home-built M A S probes incorporating a Doty Scientific 5 mm "supersonic" stator assembly. The (solid) 90° pulse length was 1.7 (is (208.43 MHz) and the spectra were referenced as above. M A S experiments were conventional and echo experiments used a rotor synchronized "90-T-180-x-acquire" sequence. 36 Hardness testing was done using a Buehler Micromet®3 Hardness Tester. Samples were polished first by sanding, and then to a 1 u\m diamond finish. Testing was done at loads of 0.5 and 1 kg for 10 seconds. Little cracking was visible even for the more brittle samples, and the penetration layer of phosphate impregnated samples was found to be well below the depth of hardness indentation. Twenty measurements were taken per sample and two or more polished samples were used per measurement point. 4.2 Preparation and Characterization of Sol-Gel Ceramics 4.2.1 Alumina Sol-Gel Alumina sol was prepared as by Yoldas [33] by adding 306 g (1.5 mol) Al(/OCH(CH 3 ) 2 ) to 3L 75°C distilled water. The pH was adjusted to 4 by addition of I M HNO3 and the mixture was stirred vigorously for 16 hours between 75-85°C, until the bluish-white suspension turned to the clear sol solution. Excess solvent was slowly evaporated from the sol until the molarity of the sol was I M . Alumina sol was allowed to dry on glass plates or gel-cast in small molds. Gel-cast samples were ground up before further study. Ceramic films were prepared by applying a layer of gel (1-2 mm thick) onto glass plates and drying over two days at room temperature. The dried gel was determined by X R D to be composed of primarily amorphous boehmite phase. Morphology study was carried out by S E M and a thermal decomposition study was carried out by X R D between 100-500°C. 4.2.2 Composite Sol-Gel Composite sol-gel (CSG) was formed by slurrying 100 mL I M alumina sol with 45 g of powdered calcined a-alumina and sonicating for 5 minutes. To the mixture, I M HNO3 was added, approximately 20 mL, until the pH was about 4 and the suspension was thinned but not 37 runny. A dispersion which is too thin will separate during gel-casting while a dispersion which is very thick will not mold evenly. The mole ratio of A1203 to starting Al(/OCH(CH3)2) is 4.4:1, which corresponds to 11% of Al derived from the sol. Suspensions were gelcast to yield a xerogel. Suspensions were poured into 3 cm diameter plastic trays and air-dried over several weeks to form the CSG composite. After shrinkage, samples.were approximately 1.8 cm in diameter and 3-4 mm deep. Ceramic films were prepared by applying a layer of suspension (1-2 mm thick) onto glass plates. The film was dry within two days at room temperature. Air-dried samples were placed at 85°C in the drying oven for 3 hours before subsequent treatment. X-ray diffraction (Appendix in.2, 85°C) and IR analysis (Appendix II.3) showed the material to be composed of primarily crystalline a-alumina and a small amount of boehmite microcrystals as observed for alumina sol. Thermal decomposition at 100-500°C was studied by XRD (AIJI.4). Decomposition of boehmite microcrystals was observed with formation of y - A l 2 0 3 . Gel-cast CSG samples were fired at 200°C or 500°C for one hour. CSG fired at 200°C was shown to be composed of a-alumina and boehmite microcrystals, while CSG fired at 500°C contained a-alumina and a small amount of y-alumina. 4.3 Processing and Characterization of Aluminum Phosphate Ceramics 4.3.1 a-Alumina and Phosphate A16 a-alumina powder (4.0 g, 0.039 mol) was mixed with 9.05 g H 3 P0 4 (0.078 mol) directly in an Al [derived from alumina source] : P [derived from phosphate source] 1:1 ratio. The slurry was fired between 200-350°C and at 500°C. Unless otherwise stated, all post-phosphate firings were carried out for 1 hour. IR analysis was carried out for materials fired between 200-350°C. Small amounts of alumina (0.3 g) were also spread onto glass plates and 38 0.67 g H3PO4 was deposited to give a 1:1 A1:P ratio. The mixture was fired for 1 hour at temperatures between 100-600°C and analyzed at each temperature for aluminum phosphate phase formation by X R D . SEM imaging was conducted on materials fired at 200, 300, and 500°C and the sample fired at 500°C was analyzed by M A S NMR. A lower phosphate ratio with a-alumina was also prepared by mixing 45 g (0.44 mol) of a-alumina with 8.9 g H3PO4 (0.077 mol) to yield a ratio of 11:1. The mixture was slurried in 46 mL of water before gel-casting or slip-casting using 7.5 cm diameter by 2 cm deep molds. After drying at 85°C , firing was carried out at 100, 300, and 500°C for 1 hour. Samples were analyzed by microhardness testing, S E M , and X-ray phase mapping and elemental peak analysis by SEM. 4.3.2 Monoaluminum Phosphate The thermal behavior of monoaluminum phosphate (A1(H2P04)3), fired at temperatures between 200-600°C for 1 hour, was studied by IR, X R D and S E M . At 300°C, samples were fired for 0.25, 0.5, 1, 2, and 4 hours and the products studied by X R D . 4.3.3 Alumina Sol-Gel and Phosphate Wet-mixing experiments were carried out by combining alumina sol directly with phosphate. Alumina sol was mixed with phosphoric acid in a 1:1 ratio, dried at 80°C, and fired at temperatures between 200-350°C for 1 hour. IR experiments were carried out for all fired samples and material dried at 80°C, and X R D measurements were carried out for samples fired at 200 and 350°C, and dried at 80°C. S E M morphology was studied for the sample fired at 200°C. Ground alumina sol, pre-fired at 500°C, was phosphated with either H3PO4 and A1(H 2P0 4)3 in a ratio of 1:1 A1:P. Mole ratios of sol to phosphate given are based on the fired sol weight, 39 where conversion from a boehmite phase to an A1 2 0 3 phase has occurred. Ground fired alumina sol (0.1 g, 9.81xl0~4 mol), pre-fired at 500°C, was mixed with 0.22 g H 3 P 0 4 (1.96 xlO"3 mol) or 0.29 g A1(H 2 P0 4 ) 3 (6.54 xlO" 4 mol) on glass plates to give in each case an A l : P ratio of 1:1. The mixtures reacted and form a plastic-like solid immediately. Subsequent firings were at temperatures of 200, 300, 350, 400, 450, and 500°C for 30 min. Reaction products were analyzed by X R D . Time was also varied at 300°C; samples were fired for 5 min, 0.25, 0.5, 1, 1.5, and 2 hours. S E M analysis was done of alumina sol reacted with H 3 P 0 4 at 300°C for 5 minutes. X R D analysis was carried out 1:1 Fired sol and H 3 P 0 4 prepared on a larger scale was also fired at 500°C for one hour and analyzed by X R D , SEM, and M A S NMR. . Samples of alumina sol pre-fired at 500°C (0.16 g, 0.0015 mol) were mixed with H 3 P 0 4 (0.72 g, 0.003 mol) to give an A1:P ratio of 1:2. The reacted samples were fired for 1 hour at 100, 200, 300, 400, 500, and 600°C and the products analyzed by X R D . Morphology of the sample fired at 400°C was studied by SEM. 4.3.4 Composite Sol-Gel and Phosphate Wet CSG slurries were reacted directly with phosphate. oc-Alumina (45 g) was mixed with 10 g phosphoric acid to form a dry paste. Alumina sol (100 mL, 1 M) was added slowly to form a thick emulsion which was thinned with the addition of 30 ml 1M H N 0 3 . The phosphoric acid loading corresponded to an A1:P ratio of 11:1. The mixture was subsequently gelcast in 7.5 cm molds, then dried at 85°C. The gel-cast pellets were fired at 300, 400, 500, and 600°C and tested for microhardness. S E M and EDS measurements were carried out on samples fired to 500°C. 40 Ground CSG (2.0 g, 0.0196 mol), pre-fired to 500°C, was mixed with 4.5 g H 3 P 0 4 (0.0392 mol) to yield a 1:1 A1:P ratio. The mixture was fired at 500°C for one hour to yield a white powder. The product was analyzed by X R D , SEM, and M A S NMR. CSG was prepared as a thin layer on glass plates, pre-fired to 200°C or 500°C to give 0.2 g of material, and covered with a surface layer of 0.22 g H3PO4 or 0.29 g A1(H 2P0 4)3- The total A1:P ratio of CSG and phosphate was 2:1. The products were fired for one hour at 200, 300, 400, and 500°C and analyzed by X R D . Time was also varied at 300°C for 2:1 CSG layer and H3PO4 samples; firing was for 0.25, 0.5, 1, 2, and 4 hours, and the products were analyzed by X R D . 4.3.5 Phosphate Impregnation The impregnation procedure involved first polishing 1.8 cm diameter by 3-4 mm thick dried CSG samples smooth on one surface. Samples were then either immediately phosphated, or fired and then phosphated. The impregnation procedure consisted of samples being evacuated under low-vacuum pressure for 20 minutes then submersed in phosphoric acid for 10 minutes while still under vacuum. The vacuum was then released and the samples removed. Polished CSG samples were fired at 200°C before phosphate impregnation. Samples were fired after phosphating at 300, 400, 500, and 600°C for 1 hour, and microhardness measurements were carried out on all fired samples. Cross-sectional S E M and EDS (elemental P) analysis was carried out for the sample fired at 500°C. Polished CSG samples were fired at 500°C before phosphate impregnation. Samples were fired after phosphating at 300, 400, 500, and 600°C for 1 hour, and microhardness measurements were carried out on all fired samples. Cross-sectional S E M and EDS (elemental P) analysis was carried out for the sample fired at 500°C. Cross-sectional microhardness measurements were also carried out. 4.3.6 Coatings Phosphate bonded CSG coatings, deposited on metal substrates were analyzed on the coating surface by X R D . H 3 P O 4 11:1 H 3 P O 4 1:1 H 3 P O 4 1:2 A1(H2P04)3 1:1 A1(H2P04)3 1:2 H 3 P O 4 Impregnation a-alumina 100-500°C 1 hr 100-600°C 1 hr alumina sol (wet) 200-500°C 1 hr alumina sol (500°C) 200-500°C 30 min 200-500°C 30 min 300°C 5 min-2 hr 300°C 5 min-2 hr CSG (wet) 300-600°C 1 hr CSG (200°C) 200-500°C 1 hr 300°C 15 min-4 hr 200-500°C 1 hr 300-600°C 1 hr CSG (500°C) 500°C 1 hr 200-500°C 1 hr 300°C 15 min-4 hr 200-500°C 1 hr 300-600°C 1 hr Table 4.3-1 Alumina-Phosphate Experiments 42 CHAPTER 5 EXPERIMENTAL RESULTS AND DISCUSSION 5.1 a-Alumina Reacted with Phosphate 5.1.1 a-Alumina and Phosphoric Acid When A16 a-Al203 (D=0.41uxn) was reacted in a 1:1 A1:P ratio in the temperature range of 200-350°C, exclusive formation of A I P O 4 phases was indicated by a strong peak around 1100 cm"' observed by infrared (IR) analysis (Figure 5.1-la). « corundum a Al(H2PO„), -rho - berlinite + intermediate .1 * A1(HP207) 2.5H 20 * meta-B o cristobalite * AIH 2P 3O l 0H 2O Wavenumber /cm"1 29 / degree Figure 5.1-1 a-b (a) IR spectra of a-alumina and 1:1 a-alumina and phosphoric acid heat-treated between 200-350°C and (b) X R D spectra of 1:1 a-alumina and phosphoric acid heat-treated at 100-500°C. Additional products of 1:1 a-Al203 and H3PO4 fired on a glass plate are identified by X R D analysis (Figure 5.1-lb). Newly formed phases are labeled as they appear in the spectra series, from lowest to highest firing temperature. Little or no reaction is observed at 100°C; corundum is the only crystalline material identified and a wide hump indicates some amorphous content. A material which is referred to as Intermediate 1 (II) is observed at 200°C. This unidentified 43 intermediate is observed at low temperature for alumina or alumina sol reacted with H3PO4 (20=11.5, 23, 60.3). II phase disappears at 300°C. While the peaks are not identified, they are matching the structure of NH 4AlH2(PO4)20 .5H 2O. II is likely a similar, hydrated M A P structure which is formed from rhombohedral M A P . The rhombohedral phase of monoaluminum phosphate (MAP-rho, Al(H2P04)3-rho) was also formed at 200°C and some corundum remained. At 300°C, MAP-rho and II disappear, and berlinite and cristobalite A I P O 4 phases form with a small amount of corundum remaining. It is unlikely that all hydrates have dehydrated at this temperature and may be in amorphous phase. At 400°C, aluminum triphosphate hydrate (ATH, A l H 2 P 3 O i 0 H 2 O ) and aluminum pyrophosphate hydrate (APH, A1HP20"7-2.5H20) are observed along with the appearance of aluminum metaphosphate B (meta-B, A1(P03)3-B). Hydrate products disappear at 500°C. Berlinite and cristobalite phases remain present at 400 and 500°C. Meta-B formation does not coincide with a decrease in A I P O 4 phases while meta-B formation follows the formation of M A P , ATP, and A P H intermediate phases. 1:1 a-Al203 and H3PO4 was also prepared on a larger scale, fired at 500°C, and analyzed by X R D and M A S NMR. 3 1 P M A S N M R shows phosphorus sites corresponding to two aluminum phosphate compounds (see Fig. 5.4-7a and Table 5.4-1). A tetrahedral phosphorus peak corresponds to A1P0 4 , which was identified as the berlinite polymorph by X R D analysis (Fig. 5.4-6). A set of three peaks, present in 1:1:1 ratios by area, correspond to the linear chain structure of aluminum metaphosphate-B (A1(P03)3-B) identified by van der Meer [55] in a paper describing the crystal structure of A1(P03)3. The monoclinic A1(P03)3 structure was shown to be composed of chains of three P O 4 tetrahedra linked with A10"6 octahedra by shared oxygen atoms to form infinite chains. In this system, the three P 0 4 tetrahedra linked in a chain may be in chemically inequivalent positions. In the 2 7 A l M A S N M R spectrum, a tetrahedral aluminum 44 site is observed due to the presence of berlinite A1P0 4 (see Fig. 5.4-7b and Table 5.4-2). An octahedral aluminum site is also observed due to unreacted CC-AI2O3 and a separate octahedral site is due to A1(P03)3-B. By analysis of peak areas of the 2 7 A l spectrum, which is considered to contain quantitatively accurate peak information, the percent of aluminum contained in A I 2 O 3 . A I P O 4 - B , and A1(P0 3) 3-B is found to be 31:63:6, respectively. Therefore, 69% of the starting alumina has reacted, and the products are calculated to correspond to approximately a 1:0.8 A1:P ratio. Product ratios are compared with those of alumina sol and CSG reacted with H3PO4 in Table 5.4-3. S E M images of phosphated a-alumina fired at 200, 300, and 500°C are shown in Figures 5.1-2 a-c. Grains 20-100 urn in size are observed for mixtures reacted at 200-500°C. Figure 5.1-2a shows a large particle of mixture reacted at 200°C which is broken apart to reveal that the material is composed of clusters of small particles, deagglomerated by grinding pressure. Energy dispersive spectrometry (EDS) indicates that, at each reaction temperature, the outer surface of the particles, which show up lighter in colour in the S E M , contain approximately 50% more phosphorous content compared with the interior of particles. The larger particles are thought to be composed of a-alumina, berlinite and cristobalite while the clusters of smaller particles surrounding are likely high-phosphate products (MAP, A T H or meta-B as identified by X R D , depending on the firing temperature). It has been shown that a-alumina undergoes substantial reaction (69%) at a 1:1 ratio of the reactants. Some high-acid phases are formed, as predicted by a relatively high phosphate loading ratio, with aluminum metaphosphate-B forming at 400°C. However, since the a-alumina has a high surface area, so primarily A1P0 4 phases (i.e. berlinite and cristobalite) are formed. N M R analysis confirms X R D findings (Fig.5.4-7a-b). 45 Figure 5.1-2 a-c S E M images of 1:1 a-alumina and phosphoric acid fired at (a) 200°C, (b) 300°C, and (c) 500°C. 5.1.2 Gel-Casting and Slip-Casting Gel-casting and slip-casting of a-Ala03 + H3PO4 slurries were carried out and the products compared for hardness and phosphate content. Slip-casting is a much faster process than gel-casting, but fast removal of the liquid phase may not allow cold-setting of phosphate bonds to occur. Slurries of low phosphate-content, with a-Al203 and H 3P04 mixed in a 11:1 A1:P ratio, were prepared and then gel-cast or slip-cast. When gel-cast samples are heated to temperatures of 85, 100, 300, and 500°C, microhardness is found to be at the maximum value of 2.5 GPa at 500°C (Figure 5.1-3). Microhardness values for slip-cast material are lower by a 5-fold decrease. For both sets of samples, an increase in hardness is observed with increase in firing temperature. Microhardness values and estimated error are given in Appendix V. 1. The hardness results indicate that during slip-casting, phosphoric acid solution is pulled through the pores of the gypsum before substantial chemical bonding can occur. Difference in morphology of slip-cast and gel-cast material fired at 500°C is observed in the SEM micrographs of Figure 5.1-4. The slip-cast sample is powdery while the particles in the gel-cast sample are observably cohesive. 46 Figure 5.1-3 Microhardness results for 11:1 phosphated a-alumina at room temperature and fired at 100, 300, and 500°C after (a) gel-casting, and (b) slip-casting. Figure 5.1-4 a-b S E M micrographs of (a) slip-cast and (b) gel-cast and phosphoric acid fired at 500°C. a-alumina (0.41 urn) Comparison of the polished surfaces of the slip-cast and gelcast samples by electron dispersion spectroscopy (EDS) revealed a larger proportion of elemental phosphate in the gel-cast sample (Figures 5.1-5 a-b). The effect is also obvious by elemental P EDS mapping 47 (Figures 5.1-5 c-d). It is evident that gel-casting is necessary for substantial pre-fire phosphate bonding in this alumina ceramic system. Figure 5.1-5 a-b EDS analysis of phosphate content in (a,c) gel-cast, and (b,d) slip-cast 11:1 a-alumina (0.41 |jm) and phosphoric acid fired at 500°C by elemental peaks and phosphate distribution map. 5.2 Heat Treatment of Monoaluminum Phosphate Monoaluminum phosphate (A1(H2P04)3) has been cited as a possible low-temperature product of the reaction between alumina and phosphoric acid. It is therefore studied separately and characterized for comparison in this work. Monoaluminum phosphate was obtained from Albright (MAIP) and contained 39 wt% water. The IR spectra of monoaluminum phosphate fired at temperatures between 200 and 350°C (Figure 5.3-la) points to the formation of 48 metaphosphate peaks when compared to reference peaks [56]. Specifically, at 350°C the peak at -800 cm"1 and the series of peaks between 1000 and 1300 cm"1 is indicative of aluminum metaphosphate formation. Peaks observed at 300°C are shifted, indicating a slightly different product, possibly an intermediate. A berlinite ' A1(HP207) 2.5H 20 * meta-B a A1(H2P04)3 -rho « corundum o cristobalite x AlH 2 P 3 O 1 0 H 2 O • meta-A a AICHJPO^., -hex + intermediate 1 10 20 30 40 50 10 20 30 40 50 26 / degree 20 / degree Figure 5.2-1 a-b XRD spectra of monoaluminum phosphate (a) heated for 1 hour between 200-600°C and (b) fired at 300°C from 15 minutes to 4 hours. Aluminum metaphosphate phases were observed to form after firing to temperatures >300°C for 1 hour (Figure 5.2-la). When fired to 200°C, hexagonal MAP (A1(H2P04)3) is the only phase observed. At 300°C, MAP disappears and ATH (H2AlP3Oi0-H2O) and APH (A1(HP20"7)-2.5H20) are formed. SEM imaging of material fired at this temperature shows clusters of a platelet structure (Figure 5.2-2 a-b). At 400°C, meta B (A1(P03)3-B) formation occurs, with substantial ATH remaining. From 400 to 600°C, the presence of ATH decreases dramatically and meta A (A1(P03)3-A)) formation occurs. The morphology of the particles 49 changes dramatically from 300 to 500°C. Clusters of small, dense particles are observed at 500°C by S E M in Figure 5.2-2c. The transformation sequence which can be derived from this information is that the formation of crystalline MAP-hex phase mainly leads to ATP and the formation of meta B followed by meta A. F i g u r e 5.2-2 a-c S E M images of monoaluminum phosphate fired for 1 hour at (a), (b) 300°C and (c) 500°C. Monoaluminum phosphate was also fired for various times at 300°C, as is shown in Figure 5.2-lb. At 15 min, A P H (A1(HP 20 7)-2.5H 20) is prevalent, with A T H ( H 2 A l P 3 O 1 0 H 2 O ) as a minor product, and some amorphous material. After 30 min, A T H is the major phase with little or no amorphous content present, and this phase is stable up to 4 hours. Little change is observed until 2hrs of firing, at which point there is formation of meta B. A much larger presence of meta B is observed after 4 hours. These results demonstrate that the same products are obtained with lower temperature and longer time as for higher temperature and shorter time conditions. Hence, crystalline aluminum phosphate product formation is faster at higher temperatures but still takes place at lower temperatures. From the experimental work carried out, Equation 5.2-1 is proposed for the crystalline phase transformation observed during the thermal treatment of monoaluminum phosphate 50 (MA1P), based on 1 hour heat-treatment time. The mechanism is consistent with findings in literature [20, 26]. 200°C 300°C 400°C 500°C MA1P —> Al(H 2 P0 4 ) 3 -hex —> H 2 A l P 3 O 1 0 H 2 O —» A1(P0 3) 3-B —> A1(P0 3) 3-A Eqn 5.2-1 It was established that monoaluminum phosphate is not crystalline below 200°C, at which point the phase is likely hydrated or amorphous. M A P (A1(H 2P0 4) 3) and A T H (H 2AlP3Oi 0 .H 2O) differ by one molecule of water, and transformation is the result of the partial dehydration of M A P . A P H (A1(HP 20 7)-2.5H 20), which is formed immediately, may also be a minor intermediate in the reaction sequence but this is not obviously seen in the results. Since meta-B formation always precedes the formation of meta-A, stepwise phase formation as in Equation 5.2-1 is suggested. That several products are often present simultaneously in the spectra indicates that reactions are moderately slow and occur over a broad temperature range. It is noted that monoaluminum phosphate forms partially amorphous products after short firing times (<15 min at 300°C). 5.3 Hydrated Alumina Sol Phases Reacted with Phosphate 5.3.1 Alumina Sol and Phosphoric Acid Alumina sol (IM) and H 3 P 0 4 were wet-mixed in a 1:1 A1:P ratio and the aluminum phosphate reaction products formed by this processing technique were studied. The slurries prepared were fired at low temperatures, between 200-350°C. By IR analysis, it was observed that A1P0 4 was observed initially, indicated by the broad peaks at 1105, 700, and 500 cm"1. The spectra is expanded in the region of 400-1400 cm"1 (Fig. 5.3-lb, full spectra in Appendix II.4), to observe a weak peak at 809 cm"1 forming between 250 and 350°C. The peak at -800 cm"1 is indicative of aluminum metaphosphate formation, as observed for the thermal treatment of monoaluminum phosphate (Fig. 5.3-la). 51 o c 200°C 1679 350°C (a) •''806 -:-J\: 4000 3000 2000 1500 Wavenumber /cm"' 1000 500 Wavenumber /cm (b) 200°C /""X — _250°C l l ' lO / " \ 300°C 1108 809 V —25fTC 1100 . 1400 1200 1000 800 600 400 Figures 5.3-1 a-b IR spectra (a) monoaluminum phosphate, and (b) 1:1 alumina sol and phosphoric acid fired at temperatures between 200-350°C. X R D measurements were taken for 1:1 alumina sol and H3PO4 at 80°C, 200°C, and 350°C (Fig. 5.3-2). Results matched those of the IR spectra, indicating orthophosphate formation and a small amount of metaphosphate formation at 350°C. At 80°C, only amorphous humps are observed, indicating that crystalline phases have not yet formed. The IR spectrum at 80°C also indicates the possible presence of M A P . At 200°C, X R D analysis shows the presence of two forms of aluminum orthophosphates: cristobalite A I P O 4 (orthorhombic phase) as the major phase; and berlinite A I P O 4 (hexagonal phase) as the minor phase. At 350°C, there was observed to be a small amount of aluminum metaphosphate formation in the low-temperature " B " form (meta-B). Amorphous intermediate phases preceding meta-B formation are indicated. S E M analysis of 1:1 alumina sol and H3PO4 fired at 200°C and gel-cast alumina sol show two materials which have quite different morphologies (Figures 5.3-3a-b). The S E M of alumina sol shows a homogeneous mass which has been ground into hard, dense chunks. The material is not easily ground, and the particles cleave with force. The S E M of alumina sol and H3PO4 show a morphology which is continuous yet porous. A substantial amount of cracking is observed, as 52 well as some tendril-like strands across the surface of particles. The formation of what are possibly crystallites across the surface indicate the formation and growth of crystalline products. Cracking and porosity indicate fast reaction and possible release of water vapor or P2O5, or density change with the crystallization of amorphous product. o cristobalite • berlinite * meta-B 29 / degree Figure 5.3-2 X R D spectra of 1:1 wet-mixed alumina sol and phosphoric acid fired at 80°C, 200°C, and 350°C for 1 hour. Figure 5.3-3 S E M micrographs of (a) 1:1 alumina sol and phosphoric acid, mixed wet and fired at 200°C, and (b) gel-cast, dried alumina sol. 53 The results are consistent with the few citations in literature. Alumina sol phase has been previously shown to react with phosphate to form berlinite and cristobalite AIPO4 phases in solution when fired at temperatures up to 350°C [52, 54]. AIPO4 phases were also cited as being favored by phosphate reaction with hydrated alumina sol [30]. The formation of aluminum metaphosphate by wet-mixed alumina sol and phosphate reaction has not previously been reported. In comparing reaction of hydrated alumina sol with an anhydrous alumina (oc-alumina), more amorphous materiaUs present at low temperature (200°C) after wet-sol reaction with phosphate. 5.3.2 Composite Sol-Gel and Phosphoric Acid Composite sol-gel (CSG) was reacted as a slurry with H3PO4. Samples were gel-cast and the effect of phosphate addition on the ceramic observed. Phosphoric acid was loaded into the CSG slurry in a A1:P ratio of 11:1. The gel-cast pellets were dried at 85°C, fired at 300, 400, 500 and 600°C and measured for microhardness (Fig. 5.3-4). When the phosphated CSG material is compared to CSG, it is observed that even after drying at 85°C, there is a significant increase in hardness, from 0.019 to 0.74 GPa which is attributed to cold-setting phosphate bonding. The CSG material steadily increases in hardness with firing temperature, however, though it is observed that hardness is further increased when the phosphated material is fired to 300°C, hardness then decreases as the firing temperature is increased up to 600°C. Decrease in hardness may be attributed to a change in aluminum phosphate phase or crystallization of some amorphous content with higher temperature, which may exert stress on the system and introduce defects. X R D analysis of alumina sol reacted with H3PO4 showed that crystallization of amorphous phases occurs at 350°C (Fig. 5.3-2). It is indicated that crystallization accompanies a decrease in microhardness. 54 2.5 •o 0.5 0 100 200 300 400 500 600 Firing Temperature /C 0 F i g u r e 5.3-4 Hardness measurements for CSG and 11:1 CSG and phosphoric acid, heat-treated between 85-600°C. F i g u r e 5.3-5 a -b (a) S E M and (b) EDS (white dots are elemental P) measurements of wet-mixed 11:1 CSG and H 3 P 0 4 , fired to 500°C. Variation in dispersion preparation affects hardness and morphology of the final materials. Viscosity can be adjusted by pH (lowering the pH thinned the dispersion), and it is found that more viscous dispersions yielded materials which tended to crack, and hardness values were slightly lower. Gel-casting a thick dispersion does not yield a homogeneous material. SEM 55 measurement (Fig. 5.3-5a) was carried of the surface of CSG material which has been phosphated and thinned by pH adjustment, then fired to 500°C. Considerable porosity is observed amongst cohesively bound particles. EDS measurement shows an even distribution of elemental phosphorus (Figure 5.3-5b) within the aluminum oxide material. 5.4.1 Alumina Sol-Gel 5.4.1.1 Heat Treatment Alumina sol was characterized after drying at room temperature. The prepared sol was dried as a thin layer onto a glass plate and analyzed by X-ray diffraction (XRD). Wide peaks corresponding to those of a semi-amorphous boehmite ( A I 2 O 3 H 2 O ) phase are observed (Figure 5.4-la). The wide peaks are indicative of very fine crystallites <100 nm in size, calculated by the Scherrer equation [34]. The crystallite size, x, was calculated to be -22 nm. (Appendix I). Figure 5.4-1 a-b (a) Air-dried alumina sol with peaks corresponding to boehmite and (b) alumina sol at different firing temperatures. Heat treatment of alumina sol was carried out at temperatures between 100-500°C. The XRD spectra of dried sol fired at 100, 200, 300, 400, and 500°C is shown in Figure 5.4-lb. Between 300 and 400°C, amorphous peaks at -32, 41 and 66 29 indicative of an A1203 phase 5.4 Heat Treated Alumina Sol-Gel Phases Reacted with Phosphate 29 / degree 29 / degree 56 begin to form. At 500°C, boehmite has disappeared, leaving a semi-amorphous alumina phase. The conversion of microcrystalline boehmite to partially-crystalline Y-AI2O3 when heat-treated above 350°C has been reported in literature, and octahedral, pentacoordinate and tetrahedral alumina sites have been observed by N M R analysis to form [54]. Analysis by " A l M A S N M R performed in this work showed the presence of primarily octahedral sites, with some pentacoordinate and tetrahedral alumina sites (Appendix IV.2). By gravimetric analysis of dried sol fired at 500°C, boehmite accounts for 83% of the dried sol if the fired product is assumed to be entirely dehydrated Y-AI2O3. Most likely, the dried sol contains additional hydrated alumina, and the presence of an unidentified peak at 20 =7.2 is observed. The peak disappears between 200 and 300°C. Subsequent phosphating was carried out on alumina sol which had been heat-treated at 500°C. This sol is considered to be in a dehydrated, fine amorphous state of Y-AI2O3 when phosphated. 5.4.1.2 Reaction with Phosphoric Acid Alumina sol (500°C) was phosphated with H3PO4 in a 1:2 A1:P ratio and heat-treated at 100-500°C. Results of X R D analysis are shown in Figure 5.4-2a; as with previous X R D spectra, different phases are labeled as they form at progressively higher temperatures or times of firing. It is shown that there is formation of A1(H2P04)3 in the rhombohedral phase (MAP-rho) at 100°C plus some amorphous material present. At 200°C, along with M A P there is the formation of the intermediate phase, intermediate 1 (11.5, 23, 60.3 26), which disappears at 300°C. This intermediate is the same as that observed at 200°C for CX-AI2O3 reacted with H3PO4 (1:1). MAP-hex forms at 300°C along with A T H . At 400°C, all hydrated phases have disappeared and exclusively meta-B is formed. By S E M , this phase is observed as clusters of 57 small, dense particles in Figure 5.4-2b. At 500°C and 600°C, meta-A also forms. It is demonstrated that alumina sol reacted with H3PO4 in a 1:2 ratio directly forms M A P . Since the formation of II has again preceded the formation of A T H , it is likely a precursor to this phase. A T H subsequently dehydrates to meta-B, followed the formation of meta-A. It is noted that the berlinite and cristobalite forms of AIPO4 did not form at all for the 1:2 A1:P ratio experiments. This is consistent with literature reports that under excess phosphate conditions, the formation of M A P phases and subsequent transformation to metaphosphate phases will take place. If excess alumina is not available to react with M A P or meta phases which are formed, then A1P0 4 phases will not be produced. a A1(H2P04)3 -rho * AlH2P3O1 0H2O * meta-B • A1(H2P04)3 -hex f intermediate 1 • meta-A ~\V 20 30 ~40 29 / degree Figure 5.4-2 a-b (a) X R D spectrum of 1:2 alumina sol (500°C) and phosphoric acid fired for 1 hour at 100-500°C and (b) S E M image after firing at 400°C. The results from the analysis of M A P , a-Al 2 03 and H 3 P 0 4 , and this study are used to postulate a mechanism for the reaction between alumina sol and PA, in a 1:2 A1:P ratio. The hexagonal phase of M A P is observed at a higher temperature after the formation of 58 rhombohedral M A P , indicating a transformation from the rhombohedral to hexagonal phase. Intermediate 1 (II) has been shown to accompany the formation of MAP-rho and may be an intermediate between MAP-rho and ATP (H2AIP3O10H2O). Following ATP formation is conversion to meta-B and meta-A at higher temperatures. Thermal transformation temperatures are in the same range as the thermal tranformations of monoaluminum phosphate, as shown in Equation 5.4-1. 100°C 200°C 300°C AI2O3-S0I + H3PO4 -> Al(H 2P0 4)3-rho -» II -» H2AIP3CM0.H2O Eqn 5.4-1 300°C 1 V 300°C Al(H 2P0 4) 3-hex 1 400°C 500°C H2AIP3O10.H2O -> A1(P0 3) 3-B -» A1(P0 3) 3-A Fired alumina sol (500°C) was combined in a 1:1 ratio with H3PO4 and fired 200-500°C for a relatively short period of time (30 minutes) in a low-temperature range. X R D spectra after firing temperatures of 200-500°C are plotted in Figure 5.4-3a. Berlinite AIPO4 and cristobalite AIPO4 are observed at all temperatures between 200 and 500°C, and do not appear to increase in relative amount. ATP (AIH2P3O10H2O) and minor amounts of A P H (A1(HP 20 7) -2.5H20) form at 300°C. These hydrated phases are present at higher temperatures, then decrease in amount at 500°C, corresponding with the formation of meta-B (A1(PC«3)3). A small amount of meta-A is also formed. Amorphous phase is present at all reaction temperatures. However, amorphous phases is not present under higher phosphate loading conditions where M A P and its subsequent transformation products are exclusively formed. Reaction of aluminum phosphates at 300°C is of interest due to the possibility of phosphating of ceramics on metals at low temperatures. Therefore, reaction time was varied at 300°C for 1:1 59 alumina sol and H3PO4. After 5 minutes, several phases of M A P form (Figure 5.4-3b). Orthorhombic and hexagonal M A P phases are identified. These phases disappear after 15 minutes of firing. Berlinite and cristobalite AIPO4 phases are formed immediately, increasing in intensity at 15 minutes, and then the amount is unchanged for the remaining firing time. A T H forms after 30 minutes and remains present for the duration of the firing time, though the peaks decrease in intensity after 2 hours. Meta-B forms between 1-2 hours. A b e r l i n i t e o e r i s t o b a l i t e x A l H 2 P 3 O , 0 H 2 O A 1 ( H P 2 0 7 ) 2 . 5 H 2 0 * m e t a - B • m e t a - A a A l C H z P O ^ - r h o m A K H j P C v ) . , - h e x c (a) I T \ L i iff. tl ft 50Q°C 20 30 40 29 / degree (b) o Jf iff Jl j j 15 min UJL/U 3 0 m i n i \H h 1 hr^J 50 T 10 20 r 30 29 / degree 40 50 F i g u r e 5.4-3 a-b X R D spectra of 1:1 alumina sol (500°C) and phosphoric acid (a) fired for 30 minutes at 200-500°C and (b) fired at 300°C from 5 minutes to 2 hours. Alumina sol, heat-treated at 500°C, reacted with H3PO4 on a large scale in a 1:1 ratio and fired for 1 hour at 500°C was analyzed by X R D and NMR. Berlinite A1P0 4 , A1(P0 3) 3-B and A1(P0 3) 3-A were identified by X R D (Fig. 5.4-6). Analysis by 3 1 P M A S N M R showed a tetrahedral peak due to berlinite, three peaks due to the chain structure of A1(P03)3-B, and a single peak due to the cyclic structure of A1(P03)3-A, which may be present in either a 3 or 4 60 P 0 3 " unit ring (see Fig. 5.4-7a and Table 5.4-1). Aluminum metaphosphate-A has been identified as the cubic system of A1 ( P 0 3 )3 [57]. Amorphous phosphate material is also observed. The 2 7 A l M A S N M R spectrum shows unreacted amorphous Y-AI2O3, and single tetrahedral and octahedral peaks due to berlinite, and A1(P03)3-B and A1(P0 3) 3-A respectively (see Fig. 5.4-7b and Table 5.4-2). The percent ratio of aluminum in A1 2 03:A1P0 4:A1(P03)3-B:A1(P0 3) 3-A is calculated to be 28:33:16:23, which corresponds to 72% reaction of alumina with phosphate. The calculated ratio of A1:P in the reaction products was 1:1.5. The non-accordance of the calculated product ratio with the reactant ratio may be due to two factors. 27 First, the broad, overlapping peaks in the A l spectrum are difficult to measure for area and amorphous material present may be composed of alumina which is not accounted for in calculations. Also, heat treated alumina sol is measured by weight during preparation and is assumed to be entirely composed of anhydrous AI2O3, but may contain some hydrate phases which would lower the real atomic ratio of Al /P to <1. By comparison of alumina sol reacted with H3PO4 in 1:2 and 1:1 ratios, it is evident that different products are formed when there is a lower proportion of phosphoric acid reacted with alumina sol. Two reactions are taking place in the 1:1 system. The first reaction produces berlinite and cristobalite A1P0 4 phases favored under high alumina-reactivity conditions, as shown in Equation 5.4-2. In the second reaction, M A P is formed after a short period of time, transforms to an amorphous intermediate phase, and is then converted to A T H and metaphosphate phases in the same sequence as Equation 5.4-1. The reaction sequence of Equation 5.4-1 is considerably less favorable under 1:1 A1:P ratio conditions. <200°C Al 2 0 3 - so l + 2H3PO4 — • AIPO4-B (major) + AIPO4-C (minor) Eqn 5.4-2 When the results are compared to that of the 1:1 ratio a-Al 203/H 3P04 system (Figure 5.1-lb), the main differences are the appearance of more acid phosphate, metaphosphate, and amorphous products. This result indicated that the amorphous material observed for the aluminum phosphate reaction products can be attributed to reaction with y-alumina (derived from alumina sol) as a starting material. 5.4.1.3 Reaction with Monoaluminum Phosphate Alumina sol (heat-treated at 500°C) was also reacted with monoaluminum phosphate in a 1:1 A1:P ratio (Figures 5.4-4a-b). Less amorphous content and phosphate-rich products (ATH, meta-B) are observed compared with alumina sol reacted with H3PO4 under the same conditions. Otherwise, product formation is similar. At 300°C after 5 minutes of firing, monoaluminum phosphate phases are lacking, indicating that monoaluminum phosphate reacts quickly with alumina sol or is present as an amorphous phase. A berlinite ' Al(HP2Ov) 2.5H20 * meta-B o cristobalite x AlH 2P 3O 1 0H 2O • meta-A a AlCHJPOOj -rho a Al(HjP04), -hex A A 200°C A 500°C 20 30 40 20 / degree o Uteri."* S.^«^\ >W"' * J * J A A 5 min i L " H f t ' \' OX ol L ' 1 5 A m i n I \ Jl I * \h r I \ W W w v L , ' W * A ' - w I, 50 10 1 20 1 30 40 50 29 / degree F i g u r e 5.4-4 a-b XRD spectra of (a) 1:1 alumina sol (500°C) and monoaluminum phosphate fired for 30 minutes at 200-500°C and (b) fired at 300°C from 5 minutes to 2 hours. 62 S E M images of 1:1 alumina sol and monoaluminum phosphate heat treated at 300°C for 5 minutes are shown in Figures 5.4-5 a-c. X R D results indicate that AlP0 4-berlinite is the main crystalline phase to form, with minor amounts of A T H formation. The S E M images show large (>100p:m), cohesive particles covered with a platelet-like material. From EDS analysis and the observed morphology, it is thought that the bulk is composed of amorphous alumina and A1P0 4-berlinite, while the outer, lighter-toned particles are aluminum triphosphate hydrate and a phosphate-rich amorphous phase. The morphology of the platelet material is the same in appearance to that of monoaluminum phosphate fired at 300°C (Fig. 5.3-3a-b), identified as ATH. It is probable that monoaluminum phosphate phase primarily reacts with the sol to form berlinite. On the outer surface of the particles, some monoaluminum phosphate remains and is converted to A T H and meta-B. F i g u r e s 5.4-5 a-c S E M images of 1:1 alumina sol (500°C) and monoaluminum phosphate fired for 5 minutes at 300°C. 5.4.2 Composite Sol-Gel 5.4.2.1 Heat Treatment CSG material which has been formed from a-alumina and alumina sol is shown by IR (Appendix U.3) and X R D (Appendix IJJ.2) to be composed of mainly corundum and a small amount of microcrystalline boehmite after drying. The microcrystalline boehmite phase is 63 observed up to a firing temperature of 300°C, a trend which corresponds to the firing of alumina sol. At 400 and 500°C, peaks due to boehmite disappear, and the formation of a transitional alumina phase is observed by high magnification of the baseline. Analysis of the CSG material fired at 500°C by 2 7 A l MAS NMR shows the presence only of octahedral sites of alumina, as was observed for OC-AI2O3, and the tetrahedral and pentacoordinate sites seen for the fired alumina sol spectra, corresponding to y-alumina, were not observed (Appendix IV.3). Samples used for subsequent phosphating had been fired at 200°C or 500°C for one hour. 5.4.2.2 Reaction with Phosphoric Acid <s corundum A berlinite * meta-B • meta-A ' GO 4 ! 1 11 » 1, .',! ** ml ?* . KJ * *f * * i • a - A l 2 0 3 L.. • ta, . A 1 2 0 3 s o l A A . C S G 10 2 0 3 0 2 0 / d e g r e e 4 0 5 0 Figure 5.4-6 XRD spectra of 1:1 a-alumina, alumina sol (500°C), and CSG (500°C) reacted with phosphoric acid and heat-treated at 500°C. Gel-cast CSG samples were ground and fired at 500°C, then reacted with H3PO4 on a large-scale in a 1:1 A1:P ratio. The resulting XRD spectrum is shown in Figure 5.4-6, and compared with a-A^Os and alumina sol reacted under the same conditions. There is observation of 64 primarily AIPO4 berlinite, with meta-B as a minor product, and a small amount of remaining corundum. XRD peaks are similar to those observed for 1:1 a-Al 20 3 and H 3 P O 4 , but with more meta-B phase present. Since it has been shown that alumina sol combined with H 3 P O 4 produces a higher proportion of metaphosphate material than O C - A I 2 O 3 and H 3 P O 4 , it is postulated that the alumina sol which is present between a-Al 20 3 particles reacts with H 3 P O 4 during firing to create high-phosphate phases (MAP, ATH, then meta-B) between particles. Peak Number Chemical Shift /ppm Assignment 1 5.1 AIPO4, berlinite phase 2 -6.1 position-1 P0 4 groups in A1(PC>3)3-B chains 3 -7.3 position-3 P O 4 groups in A1(P03)3-B chains 4 -13.2 position-2 P0 4 groups in A1(P03)3-B chains 5 -20.8 A1(P03)3-A, 3 or 4-membered ring structure 5(s) 19 spinning side-band of (5) Table 5.4-1 Peak assignments for 3 I P MAS NMR spectra. Peak Number Chemical Shift /ppm Coordination Assignment i 70.1 tetrahedral y-Al 203, formed from boehmite ii 39.8 tetrahedral AIPO4, berlinite phase ii(s) -37.8 - spinning side-band of (ii) iii 34.8 pentacoordinate y-Al 203, formed from boehmite iv 15.0 octahedral a-Al203 (corundum) V 9.8 octahedral y-Al 20 3, formed from boehmite vi -14.3 octahedral A106 group of A1(P03)3-B chains vi(s) 54.5 - spinning side-band of (vi) vii -19.7 octahedral Al cations of A1(P03)3-A rings vii(s) 47.2 - spinning side-band of (vii) Table 5.4-2 Peak assignments for Al MAS NMR spectra. By 3 1P and 2 7 A l MAS NMR experiments, peaks assigned to berlinite A1P04, A1(P03)3-B, A1(P03)3-A, and remaining a-Al 20 3 are observed (Fig. 5.4-7a-b). The A1203:A1P04:A1(P03)3-B:A1(P03)3-A product ratios as determined from the 2 7 A l spectrum are 30:56:12:2, respectively, 65 and correspond well to phosphate ratios calculated from the 3 1 P spectrum. Therefore, 70% of alumina has reacted with phosphate, and A1:P ratio of the final products was calculated to be 1:1. Product ratios and A1:P ratios for different aluminas reacted with H3PO4 in a 1:1 ratio are shown in Table 5.4-3. Reagen ts A 1 2 0 3 AIPO4 A1 ( P 0 3 ) 3 -B A1 (P0 3 ) 3 -A C a l c u l a t e d A 1 : P % A 1 %A1 % A I % A 1 P r o d u c t R a t i o a - A l 2 0 3 + H 3 P 0 4 31 63 6 - 1 : 0.8 AI2O3 s o l + H 3 P 0 4 28 33 23 16 1: 1.5 CSG+H3PO4 30 56 12 2 1: 1 T a b l e 5.4-3 Relative amounts of alumina and aluminum phosphate reaction products of alumina (500°C heat-treated) and phosphoric acid reacted in a 1:1 ratio at 500°C. Results are as calculated from 2 7 A1 M A S N M R spectra. S E M analysis was carried out on all three phosphated alumina materials reacted in a 1:1 A1:P ratio (Figure 5.4-8). The materials were seen to be all very similar in appearance. A l l are composed of mainly >10 urn particles which have clusters of smaller particles surrounding. Point analysis during S E M measurement revealed that smaller particles around the outer surface of larger particles have a higher phosphate content and would more likely contain metaphosphate, while the larger, darker (by S E M backscatter) particles are likely composed of berlinite phase. From comparison of these three aluminas reacted with phosphoric acid, it is observed that the use of alumina sol affects the type of aluminum phosphate products formed compared with a-alumina reacted with phosphoric acid. Since alumina sol has a smaller particle size and a larger surface area, it would be expected that AIPO4 phase would be favored [20]. However, phosphate-rich and amorphous products form, which must therefore be attributed to the amorphous structure of the fired alumina sol. 66 a) 3 1P Spectra 1 a -Al 2 0 3 A1 20 3 sol 5(s) 5 A 1 CSG 23 4 —^,^/\——/v^—^  5 -I 1 1 1 1 1 1 1 1 1 1 20 15 10 5 0 -5 -10 -15 -20 -25ppm b) Z/A1 Spectra a-AUO, 3 ii iv vi I 1 1 1 1 1 1 1 1 1 1 1 80 70 60 50 40 30 20 10 0 -10 -20 ppm F i g u r e 5.4-7 a -b 3 1 P and 2 7 A l MAS NMR spectra of 1:1 a-alumina, alumina sol (500°C), and CSG (500°C) and reacted with phosphoric acid and heat-treated at 500°C. 67 F i g u r e 5.4-8 a-c S E M images of 1:1 (a) a-alumina, (b) alumina sol, and (c) CSG reacted with phosphoric acid and heat-treated at 500°C. CSG material which had been applied as a xerogel film to glass plates and fired at 200°C was covered with a layer of H3PO4 corresponding to a 2:1 A1:P ratio. Samples were fired for one hour at temperatures between 200-500°C. The X R D results are shown in Figure 5.4-9a. At 200°C, there is substantial formation of cristobalite and berlinite AIPO4 phases and some corundum remains. These same peaks are present up to a firing temperature of 500°C, indicating that reaction takes place at a very low temperature and there is no further change within the temperature range studied. The ratio studied is under higher alumina conditions, and this conforms with literature indicating that at low temperatures (100°C) and under excess alumina conditions, M A P which is formed from alumina reaction with phosphate will undergo further reaction with alumina and produce AIPO4 phases [20]. It has also been cited in literature [30] that reaction of phosphate with hydrated alumina favors A I P O 4 berlinite phase formation. The results are consistent with this statement since boehmite has not been fully dehydrated after heat-treating only to 200°C. 68 A berlinite " A1(H2P04)3 phases * meta-B p cristobalite x A1I1;P30.„H:() < corundum 10 20 30 40 50 10 20 30 40 50 26 / degree 20 / degree F i g u r e 5.4-9 a -b X R D spectra of 2:1 CSG (200°C) reacted with (a) phosphoric acid and (b) monoaluminum phosphate and fired for 1 hour at 200-500°C. CSG, also prepared as a xerogel layer on glass plates and fired at 500°C, was reacted with a layer of H3PO4 in a A1:P ratio of 2:1. X R D results of samples fired for 1 hour at 200-500°C and at 300°C for 15 min-4 hr are shown in Figures 5.4-10a-b. At 200°C, there is the presence of corundum, A T H (H 2 AlP 3 OioH 2 0) , as well as another aluminum triphosphate hydrate, ATH2 (H 2AlP 3Oio-2.5H 20), and some amount of amorphous material. At 300°C, there is the formation of A P H (A1(HP 20 7)-2.5H 20), berlinite and cristobalite, and at 400°C, meta B is formed. There is a minor amount of meta B at 500°C and decreased presence of hydrate. Results differ from the results of CSG which has only been fired to 200°C and then reacted with phosphate. Metaphosphate phases are formed in this case, which is likely because all the fine, semi-amorphous boehmite had been converted to an alumina phase which would have less 69 surface area and be less reactive. Hence, some M A P formed is stable and forms A T H and then meta-B. A b e r l i n i t e v A l H 2 P 3 O 1 0 2 . 5 H O .* m e t a - B « c o r u n d u m o c r i s t o b a l i t e x A l B f i Q o H P « A I C H . P O ^ p h a s e s A l ' ( H P 2 O v ) 2 . 5 H 2 0 c (a) b u t X f V x'v' , , . .;y x.tx,x «200°C ^\S3tkJM& A LA' 300°C 400°C 30 26 / degree c 0) (b) • *~ 1 1 15 min X « A A < A A A ' I A A . JV....A." Ajvjfe. i! j j * * J / '! || i 30 min J _ * A '..^^JA^J ' w . .A. i I i I'J V O L T 10 20 30 28 / degree 40 50 (c) i « M A i r . w i * C O C t x U-4f'i A A A A 300°C v u " \ A — A ; — 200°C 10 20 30 40 26 / degree (d) Aft x x j ^ \ W l V LJb ~r~ 10 60 min 1 hr 20 30 40 26 / degree 50 Figures 5.4-10 a-d X R D spectra of (a) 2:1 CSG (500°C) and phosphoric acid and (c) CSG (500°C) and monoaluminum phosphate heated for 1 hour between 200-500°C; (b) and (d) heat-treated at 300°C from 15 minutes to 4 hours with the same compositions respectively. 70 During firing at 300°C, after 15 min, corundum, cristobalite A1P04 and berlinite A1P04 are observed (Fig. 5.4-10c-d). After 30 min, APH and ATH form, and remain present at 1 and 2 hr. After 4 hours, there is the observed formation of meta B, but the continued strong presence of APH and ATH. It is indicated that MAP is formed very early when firing at 300°C, reacts with AI2O3 to form A1P04 phases or converts to ATH (and APH) and then meta-B, the reaction sequence observed for previous samples. 5.4.2.3 Reaction with Monoaluminum Phosphate CSG material which had been applied as a xerogel film to glass plates and fired at 200°C was covered with a layer of A1(H2P04)3 corresponding to a 2:1 A1:P ratio. Samples were fired for 1 hour at 200-500°C. XRD spectra (Fig. 5.4-9b) indicate MAP phases, ATH, remaining corundum, and berlinite and cristobalite AIPO4 phases at 200°C. At 300°C, MAP has disappeared, and with increase in firing temperature, there is a decrease in ATH phase, which disappears at 500°C and substantial meta-B is formed. The stronger presence of high-phosphate material in this system is compared with reaction with H3PO4 (Figure 5.4-7a), which does not produce these phases. Under low-phosphate conditions (Al/P=2), monoaluminum phosphate may be more stable and less reactive with alumina if it is added as a reagent and not produced as an intermediate of the particle surface, and hence converts to other phosphate phases. Xerogel CSG layers, fired at 500°C, were covered with a film of A1(H2PC>4)3 in a corresponding 2:1 A1:P ratio. XRD results are shown in Figures 5.4-10 c-d. At 200°C, there is observation of MAP phases, converting at higher temperatures to ATH and meta-B. Corundum, cristobalite A I P O 4 and berlinite A1P04 are observed at each firing temperature in constant proportions. Compared with monoaluminum phosphate reacted with 200°C CSG, there is little difference, with substantial meta-B phase formed at 500°C. It is indicated that when MAP is 71 used as the phosphating material in the CSG system, the "pre-fire" temperature does not greatly affect product formation. When the results are compared with H3P04-reacted material, it is observed that there is much stronger formation of metaphosphate phases rather than A I P O 4 formation. In this case it is indicated that monoaluminum phosphate is less reactive with alumina phases, and converts to acid phosphate intermediates and aluminum metaphosphates. 5.5 S o l - G e l Phases I m p r e g n a t e d w i t h P h o s p h o r i c A c i d 5.5.1 Heat-Treated Composite Sol-Gel CSG pellets prepared by gel-casting were fired at 200°C for 1 hour and impregnated with phosphoric acid, as described in Section 4.2.5. The impregnated samples were fired at temperatures between 300-600°C. SEM analysis was done of cross-sections of the impregnation layer for a sample fired at 500°C. The penetrated layer has a pocketed, porous appearance, and is distinct in appearance from the interior bulk material (Fig. 5.5-la). By EDS map (Fig. 5.5-lb), the penetration of phosphate can be easily seen, and the distribution is homogeneous throughout the penetration depth. The impregnation depth of these samples was found to be -600 p:m. This is a model system, and the phosphated ceramic layer analyzed is thicker than phosphated CSG coatings on metals, which are -50 (j.m thick. This system allows analysis of the penetration layer on a larger scale. The penetration layer is observed to have a high density on the surface, and increased porosity with increased distance from the surface. The may be the result of the release of gasses. Cross-sectional hardness values reflect change in porosity with penetration depth. In Figure 5.5-2, it is seen that microhardness decreases until reaching the limit of impregnation. The inconsistency of point values reflects the non-homogeneity of the impregnated, reacted layer. Below the impregnation point, the hardness decreases from -2.5 GPa to <0.3 GPa (for hardness values and estimated error, see Appendix V.4). 72 F i g u r e 5.5-1 a-b Analysis of the penetration layer of CSG (200°C) impregnated with H3PO4 by (a) S E M and (b) elemental phosphorus (light dots) EDS map. 3.5 i 0 H 1 1 1 1 1 1 1 1 0 100 200 300 400 5 0 0 6 0 0 700 800 Distance f r o m Surface /u.m F i g u r e 5.5-2 Microhardness values versus depth of penetration for 200°C and 500°C-fired CSG samples impregnated with H3PO4 and fired at 500°C. Surface hardness values were measured for 200°C CSG samples impregnated and fired at temperatures between 300-600°C (Figure 5.5-3). Impregnated samples fired at temperatures >200°C show more than ten-fold increase in hardness. The peak value is at 500°C, with the hardness decreasing slightly at 600°C. Experiments were also carried out where the samples 73 were impregnated with phosphate, and then left for 12 hours at 85°C before firing at 300°C. The result was a 37% increase in hardness from the value of the sample fired immediately after impregnation at 300°C (Appendix V.6). This shows that more time which is allowed for phosphate bonding to take place improves the hardness properties of the material. The results are compared with those of 500°C CSG impregnated with phosphoric acid. o C O co o O 3.5 2.5 2 H 1.5 0.5 -\ 0 200 C Pre-fire 500 C Pre-fire 200 300 400 500 Firing Temperature /C 600 F i g u r e 5.5-3 Hardness values for 200°C and 500°C-fired CSG samples impregnated with H3PO4 heat-treated at temperatures between 300-600°C. CSG pellets prepared by gel-casting were fired at 500°C for 1 hour and impregnated with phosphoric acid. The impregnated samples were fired at temperatures between 300-600°C. SEM analysis was carried out on cross-sections of the impregnation layer for a sample fired at 500°C. By EDS map (Figure 5.5-4a-b), the penetration of phosphate looks to be even and homogeneous to a depth of -600 um. The penetration depth is similar to that of the impregnated 200°C pre-fired CSG (Section 5.4.3), however the appearance by S E M is different. There is less overall porosity in the penetration layer and there is very little increase in porosity 74 with increasing distance from the surface. It is indicated that reaction of fired CSG (500°C) with H3PO4 preceded with little evolution of gaseous products. (a) • i mm 5 0 0 ( i m F i g u r e 5.5-4 a-b Analysis of the penetration layer of CSG (500°C) impregnated with H3PO4 by (a) S E M and (b) elemental phosphorus EDS map. In the cross-sectional hardness values (Figure 5.5-2), it is seen that hardness is approximately constant at -2.3 GPa until around 600 urn from the surface, when the hardness decreases to <0.3 GPa at 720 |im. This result differs from that of the 200°C CSG impregnated with PA. While the other system displayed a higher surface hardness, internal hardness results were more erratic, reflecting a non-homogeneous structure. Surface microhardness values were determined for samples fired at temperatures between 300-600°C. In Figure 5.5-3, it is shown surface microhardness reaches a maximum value with firing temperature. Unlike the 200°C system, higher hardness values were not retained at 500 and 600°C. The peak value is at 400°C and at a lower hardness than the peak value for the 200°C reacted material. In the experiments where the samples were impregnated with phosphate, and then left for 12 hours at 85°C before firing at 300°C, there was an 18% increase in hardness (Appendix V.6). Microhardness may be related to amorphous content and 75 aluminum phosphate crystallization. Reaction of 500°C fired alumina sol has been shown by M A S N M R to form some amorphous phase when reacted with H3PO4 at 500°C (Fig. 5.4-7a). Amorphous products have been shown to be present at 200°C and 300°C when CSG (500°C) is reacted with H3PO4 (Fig. 5.4-10a). Crystallization of meta-B and reduction in amorphous phase at 400°C and 500°C corresponds with reduction in microhardness values. Therefore, while progress of chemical bonding with increase in heat-treatment temperature may increase microhardness, subsequent crystallization with further increase in temperature could decrease microhardness. Within an optimum firing range (300-400°C), it is thought that a disordered arrangement of metaphosphate type material is present which may improve the protective properties (in terms of strength, wear resistance, and toughness) of CB-CSG ceramics. 5.5.2 Aluminum Phosphate Coatings Composite sol gel alumina coatings which were applied to different metal surfaces and phosphated with phosphoric acid or monoaluminum phosphate were provided by Q. Yang, Department of Metals and Materials, at U B C . The CSG coatings had been fired to 300°C on the metal before phosphating, and then subsequently fired to 300°C for 5 minutes. The X R D spectrum of a CSG/H3PO4 coating on steel is shown in Figure 5.5-5a. Substantial corundum remains, and products formed are cristobalite and tridymite forms of AIPO4, and some amorphous material indicated by a hump at 20 =22°. Zirconium oxide (Zr0 2 is also included in this coating, as indicated by a peak at 20 =30.3°. In Figure 5.5-5b, the X R D spectrum of a CSG/A1(H2P04)3 coating on aluminum shows the same products. The use of monoaluminum phosphate or phosphoric acid as the bonding phase does not seem to affect the types of aluminum phosphate phases formed in these systems. Since substantial corrundum remains, it can be assumed that the systems are under conditions of low loading of phosphate (Al/P ratio >2). Remaining anhydrous alumina (boehmite) in the system may increase alumina reactivity, 76 since the CSG is only fired to 300°C before phosphating, and favor the formation of AIPO4 products. The appearance of tridymite AIPO4 has not been observed in other experiments. The amorphous phase present may be an aluminum metaphosphate-type phase which has remained amorphous under the low firing temperature and short reaction time conditions. o c r i s t o b a l i t e T t r i d y m i t e .e (a) 10 Ml « c o r u n d u m • a l u m i n u m 20 \ JV^.l J V A . . . z i r c o n i a ( t e t r a g o n a l ) 30 29 / degree 40 50 29 / degree F i g u r e 5.5-5 a -b XRD Spectra of (a) CSG on steel, fired at 300°C, phosphated with H3PO4 and fired at 300°C, and (b) CSG on aluminum, fired at 300°C, phosphated with A1(H2P04)3 and fired at 300°C. 77 CHAPTER 6 SUMMARY AND CONCLUSIONS 6.1 Summary From analysis of the reaction products of various aluminas reacted with phosphoric acid or monoaluminum phosphate, it was found that two reaction sequences occur. In the first reaction sequence, monoaluminum phosphate can be formed by alumina reaction with phosphoric acid at approximately 100-200°C and converted to aluminum triphosphate hydrate and amorphous material at 300°C, aluminum metaphosphate-B (chain form) at 350°C, and aluminum metaphosphate-A (ring form) at 500°C. The results were corroborated by the thermal treatment of 61% monoaluminum phosphate solution. This reaction sequence dominates under high phosphate loading conditions, such as for alumina sol (pre-fired at 500°C) and phosphoric acid reacted in a 1:2 A1:P ratio. In the second reaction sequence, the reaction of alumina and phosphate forms aluminum orthophosphate phases, either berlinite, tridymite, or cristobalite. Aluminum orthophosphate phases are the primary phases formed when alumina and phosphate are reacted in an Al/P ratio > 1, though a small amount of aluminum metaphosphate phases were still observed to crystallize at temperatures >300°C. Composite sol-gel (CSG) ceramic containing hydrated alumina phase (boehmite) also favors the formation of aluminum orthophosphates. S E M and EDS analysis of mixed aluminum orthophosphate and metaphosphate reaction products indicated a trend where larger (50-100 \im) particles of low phosphate content are surrounded by clusters of fine particles of higher phosphate content. The large particles show a dense morphology which the fine particles are a platelet structure at 200-300°C and form spherical particles at temperature >400°C. The fine particle morphology corresponds to 78 aluminum triphosphate hydrate and aluminum metaphosphate, respectively, formed from the thermal treatment of monoaluminum phosphate. The reaction products of monoaluminum phosphate as a phosphate binder were compared with the products formed with phosphoric acid. It was observed that at a higher phosphate loading (Al/P=l), phosphoric acid binder produces more phosphate-rich products (ATH, meta-B), while at a lower phosphate loading (Al/P=2), monoaluminum phosphate binder produces more phosphate-rich products (compared with phosphoric acid binder). Monoaluminum phosphate is less reactive than phosphoric acid and therefore product formation is less affected by the A1:P ratio. The hydration of alumina reacted with monoaluminum phosphate was also shown to have little effect of the formation of aluminum phosphate reaction products. Drying of alumina sol was shown to form microcrystalline boehmite of -22 nm particle size. With thermal treatment, boehmite is dehydrated and converted to y-alumina. This result 27 was confirmed by the appearance of octahedral, pentacoordinate, and tetrahedral sites in the A l MAS N M R spectrum. The reaction products of 1:1 alumina sol (heat-treated at 500°C) and phosphoric acid were compared with the products of reaction of 1:1 a-alumina and CSG (heat-treated at 500°C) with phosphoric acid and reacted at 500°C. While all reactions showed approximately the same consumption of alumina to form aluminum phosphate products (-70%), ratios of aluminum orthophosphates to metaphosphates were very different for a-alumina (91:9) and alumina sol (45:55). CSG showed a slightly higher ratio of metaphosphate products than a-alumina. The presence of amorphous material was also observed in the 2 7 A l M A S N M R spectrum of reacted alumina sol. It is clear that the use of alumina sol reacted with phosphoric acid affects the type and amorphous character of the aluminum phosphate products formed. M A S N M R analysis was found to be an excellent technique for identifying and quantifying aluminum phosphate systems, and especially systems containing amorphous content. 79 Microhardness could be related to amorphous content and aluminum phosphate crystallization. For composite sol-gel (pre-fired at 500°C) impregnated with phosphoric acid and heat treated, microhardness was found to peak at 2.5 GPa at a firing temperature of 400°C and then decrease with increasing temperature. Crystallization of meta-B and reduction in amorphous phase (devitrification) at 400°C and 500°C corresponds to a reduction in microhardness. Amorphous material present in the low-fired (300-400°C) samples could be contributing to the improvement of other mechanical properties, such as wear and toughness of CSG. Some observations were made concerning processing of aluminum phosphate ceramics. From a comparison of slip-casting and gel-casting of phosphate-bonded a-alumina, a greater extent of bonding with, gel-casting was surmised due to higher hardness, higher phosphate content, and an observably more cohesive microstructure. Phase transformations of monoaluminum phosphate were found to occur at lower temperatures over longer firing times. Some real coating materials were analyzed and related to the data compiled on these systems. Composite sol-gel coatings were set on metal substrates, fired to 300°C for 5 minutes, and phosphated with a thin impregnation layer of phosphoric acid or monoaluminum phosphate with subsequent firing at 300°C. Both systems indicated that little phosphate was present (Al/P>2), since substantial corundum remains. The X R D spectra show aluminum orthophosphate products and also indicate amorphous content, which may be high-phosphate phases which have not yet crystallized due to the low firing-temperature, short reaction-time conditions. 80 6.2 Conclusions From analysis of the literature, previous studies, and the present investigation, the following conclusions are drawn: 1. Reaction between alumina and phosphoric acid or monoaluminum phosphate, heat-treated in the temperature range of 100-500°C, results in the formation of aluminum orthophosphate (AIPO4) phases and /or aluminum metaphosphate (A1(P03)3) phases. 1.1 Aluminum orthophosphate phases are produced directly from reaction of alumina and phosphate (phosphoric acid or monoaluminum phosphate) at <200°C. 1.2 Monoaluminum phosphate (A1(H2P04)3), formed by reaction of alumina and phosphoric acid at 200°C (or initially present as a reactant), converts to aluminum triphosphate hydrate (AlHjPsOio^O) and amorphous phases at 300°C, then to aluminum metaphosphate-B at 350°C, and aluminum metaphosphate-A at 500°C. 2. The formation of aluminum phosphate products is dependent on the alumina to phosphate reactant ratio, temperature and time of heat-treatment. 2.1 High phosphate-content phases (monoaluminum phosphate, which converts to aluminum triphosphate hydrate, and aluminum metaphosphates) are exclusively favored by high phosphate conditions (Al/P<0.5). 2.2 Monoaluminum phosphate heat-treated at 300°C is amorphous for <15 minutes reaction time. 2.3 Aluminum metaphosphates may also be formed at lower temperatures over longer firing times, such as at 300°C for two hours. 2.4 Aluminum orthophosphate phases are the majority phases produced under low phosphate conditions (A1/P>1). 81 2.5 Formation of aluminum orthophosphate phases occurs immediately, even at relatively low temperature (<200°C), and is largely complete after <5 minutes firing time. 3. The type of amorphous alumina formed by heat-treatment of alumina sol affects the formation of crystalline and amorphous aluminum phosphate products. 3.1 Alumina sol forms -22 nm particle size microcrystalline boehmite phase when dried, which remains present during thermal treatment until 300°C. At temperatures >400°C, the boehmite phase is converted to amorphous y-alumina. 3.2 Low-fired (200°C) alumina sol phase (boehmite) reacted with phosphoric acid favor the formation of aluminum orthophosphate phases. The presence of remaining hydrated alumina (boehmite) in the composite sol-gel system promotes reaction with phosphoric acid. 3.3 Heat-treated (500°C) alumina sol phase (y-alumina) and phosphoric acid fired in a 1:1 ratio forms considerably more amorphous phase and monoaluminum phosphate products when heat-treated between 200-500°C, compared with a-alumina reaction with phosphoric acid. The amorphous y-alumina formed favors amorphous and metaphosphate products. 4. Reaction products of alumina and monoaluminum phosphate are less affected by the type of alumina reacted with phosphate and reaction conditions compared to alumina reaction with phosphoric acid. 4.1 Monoaluminum phosphate is generally less reactive and can act as a self-binder since it can directly convert to metaphosphate-type phases. 4.2 Under both high (Al/P=l) and low (Al/P-2) relative phosphate loading conditions, the reaction of alumina and monoaluminum phosphate produces phosphate-rich products. 82 5. Analysis of morphology and microhardness of the products show that physical and mechanical properties are affected due to aluminum phosphate formation and transformations with heat treatment. 5.1 S E M investigation of reacted material at high magnification indicates particles containing aluminum orthophosphate phases with smaller particles of higher phosphate phases, such as aluminum triphosphate hydrate or aluminum metaphosphates, surrounding (depending on temperature of heat-treatment). 5.2 Impregnation of low-fired (200°C) composite sol gel may produce dehydration side reactions to produce gaseous products. More porosity is observed below the surface than for pre-fired (500°C) composite sol-gel. 5.3 Microhardness of composite sol-gel material increases more than 10-fold with impregnation with phosphoric acid and firing above 200°C. 5.4 It is indicated that conversion from amorphous phase to crystalline aluminum metaphosphate within the phosphated composite sol-gel systems decreases the hardness of the material. There is a decrease in microhardness at temperatures of 350-500°C which can be correlated with the temperature of formation of crystalline aluminum metaphosphates. 6. Solid state M A S N M R is an excellent technique for identifying and quantifying aluminum phosphate solid products, especially if there is the presence of amorphous phase. 6.1 2 7 A l M A S N M R is used to identify phases of heat-treated alumina sol and products of alumina and phosphate, and is the only technique which can be accurately used to calculate product ratios. 83 6.2 3 I P M A S N M R confirmed results from 2 7A1 spectra and is used to identify the presence of both crystalline and amorphous aluminum phosphate products and to identify structures of aluminum metaphosphates in A and B forms. From analysis of real composite sol-gel coatings deposited on metal substrates, pre-fired at 300°C, phosphate impregnated and fired at 300°C, it is postulated that there is little reacted phosphate (Al/P>2) since there is substantial corrundum remaining and amorphous phase present. 84 CHAPTER 7 RECOMMENDATIONS FOR FUTURE WORK In the l o w - t e m p e r a t u r e r a n g e s t u d i e d ( 1 0 0 - 6 0 0 ° C ) , f u r t h e r a n a l y s i s s h o u l d be c a r r i e d out u n d e r l o w e r p h o s p h a t e c o n d i t i o n s ( A l / P > 2 ) i n o r d e r to m o r e c l o s e l y m a t c h c o n d i t i o n s o f c o a t i n g p r e p a r a t i o n . S i n c e c e r a m i c c o a t i n g s o n m e t a l s n e c e s s a r i l y r e q u i r e sho r t f i r i n g t i m e s ( 5 - 1 5 m i n u t e s ) at r e l a t i v e l y l o w t e m p e r a t u r e s , r e s e a r c h c o u l d be m o r e f o c u s e d o n p h a s e f o r m a t i o n o v e r sho r t f i r i n g i n t e r v a l s a n d at l o w e r t e m p e r a t u r e ( ~ 3 0 0 ° C ) . F u r t h e r a n a l y s i s s h o u l d b e c a r r i e d ou t u s i n g m a g i c a n g l e s p i n n i n g n u c l e a r m a g n e t i c r e s o n a n c e ( M A S N M R ) s p e c t r o m e t r y i n o r d e r to a c c u r a t e l y q u a n t i f y the a m o u n t o f e a c h o f the p h a s e s f o r m e d a n d i d e n t i f y a m o r p h o u s states f o r m e d at l o w e r t e m p e r a t u r e s ( < 5 0 0 ° C ) a n d f o r a n a l y s i s o f r ea l c o a t i n g s . It is a l s o r e c o m m e n d e d that the r e s e a r c h is c o m p l e m e n t e d w i t h f u r t h e r s t u d i e s o f m e c h a n i c a l p r o p e r t i e s s u c h as s t u d i e s o f h a r d n e s s , w e a r a n d c o r r o s i o n . I f the resu l t s a re c o r r e l a t e d to m e c h a n i c a l p r o p e r t i e s , the i n f o r m a t i o n c o u l d b e v a l u a b l e i n o p t i m i z i n g the p r o c e s s i n g a n d p r o p e r t i e s o f c o m p o s i t e a l u m i n a s o l - g e l c o a t i n g s o n m e t a l s . M o r e g e n e r a l r e c o m m e n d a t i o n s f o r f u t u r e w o r k i n c l u d e the i n v e s t i g a t i o n o f d i f f e r e n t " f i l l e r " p h a s e s i n s y s t e m s s t i l l i n c o r p o r a t i n g a l u m i n a s o l a n d p h o s p h a t e as a b o n d i n g p h a s e , a n d d e t a i l e d a n a l y s i s o f p r o d u c t m o r p h o l o g y ( s u c h as p o r o s i t y ) f o r i m p r o v e m e n t s i n c o r r o s i o n a n d w e a r a p p l i c a t i o n s . A l t e r n a t e f i l l e r c a n d i d a t e s are a l u m i n u m n i t r i d e ( A 1 N ) , s i l i c o n c a r b i d e ( S i C ) , a n d z i r c o n i a (ZrO"2), w h i c h c a n b e u s e d to ad jus t t h e r m a l c o n d u c t i v i t y a n d t h e r m a l e x p a n s i o n p r o p e r t i e s , a n d m e t a l f i l l e r p h a s e , u s e d to i m p r o v e f r a c t u r e - t o u g h n e s s . 85 APPENDIX I CALCULATION OF CRYSTALLITE SIZE BY THE SCHERRER EQUATION Very fine crystallites <100 nm in size lack enough diffractive planes necessary to cancel scattering, which results in a peak-broadening effect in X R D spectra. By the Scherrer equation, crystallite size broadening (p\) is related to the mean crystallite dimension (x) by the following formula [34]: x = KX E q n A l p\cos0 Line broadening p\ is in radians, and given by (B-b), where B is the breadth of the line at half intensity maximum of the peak and b is inherent instrumental broadening. K is the shape factor, taken at 0.9, and X is the wavelength of incident X-rays, which is 0.154 nm for the spectrometer used.. For the dried alumina sol spectrum was analyzed for the main peak at 13.7 20. It was found that B=7.04xl0"3 rad, and b=7.04xl0"4 rad for (larger-sized) boehmite used to obtain a value for inherent instrumental broadening. By calculations using the Schere equation and the parameters given, the mean crystallite dimension is found to be T=22 nm. APPENDIX II SUPPLEMENTARY INFRARED SPECTRA Figure AH. 1 A16 a-Alumina (0.4 |im) c o c ,Mr-K \ / i i 11 'xl I i j 4000 3000 2000 1500 1000 cm-Figure AH2 Alumina Sol, air-dried ! ! Jl i i ! 500 cm" Figure AIJ.3 Composite Sol-Gel, air-dried C/3 c >-E-1 s o 4 0 0 0 3000 2 0 0 0 1500 1000 c m " 1 500 Figure AU.4 1:1 Wet-Mixed Alumina Sol and Phosphoric Acid o o o o o o o H 85 C 3 0 0 C 3 5 0 C_ 5 0 0 4 0 0 0 3 0 0 0 2 0 0 0 1500 1000 c m - l APPENDIX III SUPPLEMENTARY X-RAY DIFFRACTION SPECTRA Figure A H I . l A l 6 a - A l u m i n a J L 10 20 30 40 50 2-Theta-Scale 60 70 Figure AIH.2 Compos i te S o l - G e l 85 C 100 c / \ 300 CL A 400 C 500 C - U W . , J T I f 1/ 1^ 60 70 10 20 30 40 50 2-Theta-Scale 89 APPENDIX IV SUPPLEMENTARY MAS NMR SPECTRA Figure ArV. 1 A l 2 7 Spectrum of A16 Figure ATV.2 A l 2 7 Spectrum of Alumina a-Alumina Sol, Heat-Treated at 500°C iv 50 40 30 20 10 6 AO -20 ppm 90 80 70 60 50 40 30 20 10 6 - l 'o-20-30 ppm' Figure ArV.3 A l Spectrum of Composite Sol-Gel, Heat-Treated at 500°C iv —r— 50 l b 30 20 ib" -10 -20 ppm Note: Refer to Table 9.4-2 for assignment of peaks. 90 APPENDIX V MICROHARDNESS VALUES AND ESTIMATED ERROR Table AV. 1 Surface Microhardness of 11:1 a-Al 20 3 and H3PO4 Processing Firing Temperature /°C Indentation Load /g Microhardness /GPa slip-cast 80 50 0.0265 ± 0.000 slip-cast 100 50 0.036 ±0.001 slip-cast 300 300 0.40 ± 0.02 slip-cast 500 500 0.45 ± 0.02 gel-cast 80 100 0.30 ±0.17 gel-cast 100 100 0.42 ± 0.09 gel-cast 300 500 2.02 ±0.08 gel-cast 500 500 2.48 ± 0.07 Table AV.2 Surface Microhardness of Composite Sol-Gel (CSG) Firing Temperature /°C Indentation Load /g Microhardness /GPa 80 100 0.019 ±0.008 200 100 0.133 ±0.007 300 100 0.187 ±0.013 400 100 0.23 ± 0.03 500 300 0.26 ± 0.02 600 300 0.287 ± 0.009 Table AV.3 Surface Microhardness of Wet-Mixed 11:1 CSG and H 3 P0 4 Firing Temperature /°C Indentation Load /g Microhardness /GPa 80 500 0.7 ±0.3 300 500 2.4 ±0.3 400 500 2.14 ±0.09 500 500 1.76 ±0.09 600 500 1.7 ±0 .2 91 Table AV.4 Cross-Sectional Microhardness of Heat-Treated CSG Impregnated with H 3 P0 4 Pre-fire Temperature /°C Firing Temperature /°C Distance from Surface /urn Indentation Load /g Microhardness /GPa 200 500 0 500 3.1 ±0.7 200 500 80.7 500 2.3 ±0.4 200 500 161 500 2.2 ±0.3 200 500 242 500 2.4 ±0 .4 200 500 323 500 2.3 ±0 .4 200 500 403 500 2.1 ±0.3 200 500 483 500 2.5 ±0.4 200 500 565 500 1.72 ± 0.19 200 500 645 500 1.7 ±0.3 200 500 726 500 0.221 ±0.015 500 500 0 500 2.3 ±0.3 500 500 80.7 500 2.32 ±0.15 500 500 161 500 2.24 ±0.13 500 500 242 500 2.3 ±0.2 500 500 323 500 2.15 ±0.13 500 500 403 500 2.10±0.19 500 500 483 500 2.1 ±0.2 500 500 565 500 2.03 ±0.14 500 500 645 500 1.8 ±0.3 500 500 726 500 0.23 ±0.03 Table AV.5 Surface Microhardness of Heat-Treated CSG Impregnated with H3PO4 Pre-fire Temperature /°C Firing Temperature /°C Indentation Load /g Microhardness /GPa 200 300 500 2.4 ±0.3 200 400 500 2.6 ±0.6 200 500 500 3.1 ±0.7 200 600 500 2.9 ±0.4 500 300 500 2.0 ±0.3 500 400 500 2.60 ±0.13 500 500 500 2.3 ±0.3 500 600 500 1.8 ±0.2 92 Table AV.6 Surface Microhardness of Heat-Treated CSG Impregnated with H3PO4 with a 12 hour Delay before Firing Pre-fire Temperature /°C Firing Temperature /°C Indentation Load /g Microhardness /GPa 200 300 500 3.27 + 0.18 500 300 500 2.4 ± 0.4 93 APPENDIX VI OTHER DATA AVI. 1 Particle Size Measurement of A16 SG a-Alumina Powder * CONDITIONS DISP.. VISC. DISP, DENS. SflHR.. DENS.. D (RAX) D (MIN) SPEED TIME 0.96 EcP] LI 3.9? ES/cc] [9/cc] 5.80 mi e. 18 im 960 [r.Pm/ain] 6 H 6 PI 58 S 16 Dm] * DATA TIME RBSORBRNCE 0.5 • DISTRIBUTION GRAPH (8V VOL.) . CktD]=0PTI0NfiL) D, (HEDifiN) SD Sv 0.41 .CJ»«I 8.64 IM 4.387. -lnWl 94 REFERENCES 1 W.D. Kingery, "Fundamental study of phosphate bonding in refractories:I-m" J. Am. Ceram. Soc, 33 [8] 239-250 (1950). 2 S.T. Wilson, B . M . Lok, C.A. Messina, T.R. Cannan, and E . M . Flanigen, "Aluminophosphate molecular sieves: a new class of micoroporous crystalline inorganic solids," J. Am. Chem. Soc, 104, 1146-47 (1982). 3 Chung, United States Patent, "Phosphate binders for metal-matrix composites," Patent Number 5536686, Jul. 16, 1996. 4 F. William, F. Laqaurent, United States Patent, "Aluminum phosphates and binder compositons/ceramic materials comprised thereof," Patent Number 5496529, Mar. 5, 1996. 5 Armstrong World Ind Inc, Great Britian Patent, "Phosphate ceramic materials," Patent Number 2100246, Dec. 22, 1982. 6 D. Singh, A.S. Wagh, M . Tlustochowicz, S.Y. Jeong, "Phosphate ceramic process for macroencapsulation and stabilization of low-level debris wastes," Waste Management, 18, 135-143 (1998). 7 R.J. Brook, R.W. Cahn, P. Haassen, E. Kramer, Materials Science and Technology, Processing of Ceramics: Parti, Weinheim, New York, pp 72, 298 (1996). 8 Publication by Chemat Technology, Inc. (www.chemat.com) 9 R.N. Rothon, "Solution Deposited Metal Phosphate Coatings," Thin Solid Films, 11, 149-53 (1981). 1 0 D A . Barrow, T.E. Petroff, and M . Sayer. "Thick ceramic coatings using a sol-gel based ceramic-ceramic 0-3 composite," Surface and Coatings Technology, 16-11, 113-118 (1995). 1 1 Barrow et al, United States Patent, "Method for producing thick ceramic films by a sol-gel coating process," Patent Number 5585136, Dec. 17, 1996. 1 2 T. Troczynski, Q. Yang, "Process for making chemically bonded sol-gel ceramics," United States Patent Application, serial number 091383074, filed August 26, 1999. 1 3 Q. Yang, "Composite sol-gel ceramics," PhD Thesis, Department of Metals and Materials Engineering, University of British Columbia, May 1999. 1 4 Q. Yang and T. Troczynski. "Dispersion of alumina and silicon carbide powders in alumina sol," J. Am. Ceram. Soc, 82 [7] 1928-30 (1999). 1 5 S. Wilson, H . M . Hawthorne, Q. Yang and T. Troczynski. "Sliding and abrasive wear of composite sol-gel alumina coated A l alloys," accepted for publication in Surface and Coatings Technology, July, 2000. 95 1 6 R . M . German. "Sintering Theory and Practice," John Wiley & Sons, Inc, New York, 1996. 1 7 D . M . Roy. "New strong cement materials: chemically bonded ceramics" Science, [6] 651-8 (1987). 1 8 R.K. Brow, T .M. Alam, D.R. Tallant, and R. J. Kirkpatrick. "Spectroscopic studies on the structures of phosphate sealing glasses", MRS Bullietin, November 1998. 1 9 M.J . O'Hara, J.J. Duga and H.D. Sheets, Jr., "Studies in phosphate bonding," Am. Ceram. Soc. Bull., 51 [7] 590-95 (1972). 2 0 F.J. Gonzalez and J.W. Halloran, "Reaction of orthophosphoric acid with several forms of aluminum oxide," Ceramics Bulletin, 59 [7] 727-738 (1980). ' 2 1 Cassidy, J.E., "Phosphate bonding then and now," Ceramic Bulletin, 56 [7] 640-643. 2 2 J.E. Lyon, T.U. Fox, and J.W. Lyons, "Inhibited phosphoric acid for use in high-alumina refractories," Am. Ceram. Soc. Bull., 59 [7] 727-38 (1980). "" S.J. Likasiewicz and J.S. Reed, "Phase development on reacting phosphoric acid with various Bayer-process aluminas," Am. Ceram. Soc. Bull., 66 [7] 1134-38 (1987). 2 4 W.R. Beck, "Crystallographic inversions of the aluminum orthophosphate polymorphs and their relation to those of silica," J. Am. Ceram. Soc, 32 [4] 147-51 (1949). 2 5 J.V. Bothe, Jr., and Paul W. Brown, "Reactivity of Alumina toward phosphoric acid," J. Am. Ceram. Soc, 76 [9] 2353-58 (1993). 2 6 J .M. Chiou, D.D.L. Chung, "Improvement of the temperature resistance of aluminium-matrix composites using an acid phosphate binder," J. Mat. Sci. 28 (1993) 1435-1487. B.D. Craig and L.F. Francis, "Alumina/epoxy interpenetrating phase composite coatings: I, Processing and microstructural development," J. Am. Ceram. Soc, 81 [12] 3109-16 (1998). 2 8 H. Remy, "Treatise on Inorganic Chemistry, Volume I," Elsevier Publishing Co., Amsterdam, 633-636 (1956). 29 Ya. V . Klyucharev and L .K. Skoblo, "Composition of products in the hardening of aluminophosphate binders in corundum refractories," Zh. Prikl. Khim., 38 [3] 530-535 (1965). 3 0 J.V. Bothe, Jr., and P.W. Brown, "Low-temperature formation of aluminum orthophosphate," J. Am. Ceram. Soc, 76 [2] 362-68. 3 1 A.S. Yutina, Z.D. Zhukova and S.V. Lysak, "Reaction of phosphoric acid with some forms of alumina," Izv. Akad. Nauk SSSR, Neorg. Mater., 2 [11] 2020-28 (1966). 96 S.M. Bradley, R.A. Kydd, and F.F. Howe, "The structure of A l gels formed through the base hydrolysis of A l 3 + aqueous solutions," Journal of Colloid and Interface Science, 159, 405-412 (1993). 3 3 B .E. Yoldas, "Alumina sol preparation from aldoxides," Ceramic Bulletin, 54 [3] 289-290 (1975). 3 4 Kurt H. Stern. Metallurgical and Ceramic Protective Coatings, Chapt 6, Chapman & Hall, London, 1996, pp 113-151. 3 5 C.J. Brinker, C.S. Ashley, R.A. Cairncross, K.S. Chen, A.J . Hurd, S.T. Reed, J.S. Samuel, P.R. Schunk, R.W. Schwarz and C.S. Scotto, "Sol-gel derived ceramic films- fundamentals and applications," in Metallurgical and Ceramic Protective Coatings, ed. K . H . Stern, Chapman & Hall, London, 112-151 (1996). 3 6 A . Stoch, C. Paluszkiewicz, T. Gibala, and A. Bolek, "FTIR study of silica sol-gel deposited films on anodically-oxidized aluminum," Journal of Molecular Structure, 293, 287-290 (1993). 3 7 C.J. Brinker, and G.W. Scherer, "Sol-Gel Science, The Physics and Chemistry of Sol-Gel Processing," Academic Press, New York (1990). 3 8 S.S Prakash, C.J. Brinker, A.F. Hurd and S.M. Rao, "Silica aerogel films prepared at ambient pressure by using surface derivatization to induce reversible drying shrinkage derivatization to induce reversible drying shrinkage," Nature, 31 A, 439-43 (1995). 3 9 W.H. Gitzen, L.D. Hart, and G. Maczura, "Phosphate-bonded alumina castables: Some properties and applications," Bull. Am Ceram. Soc, 53 [6] 217-23 (1956). 4 0 R.J. Bratton and S.K. Lau, Advances in Ceramics, Vol . 3., American Cermics Society, Columbus OH (1981). 4 1 C G . Cuizard, A .C . Julbe, and A. Ayral, "Design of nonosized structures in sol-gel derived porous solids. Applications in catalyst and inorganic membrane preparation," J. Mater. Chem., %, 55-65 (1999). 4 2 J. Livage, F. Babonneau, M . Chatry and L. Coury, "Sol-gel synthesis and N M R characterization of ceramics," Ceramics International,23, 13-18 (1993). 4 3 C.J. Brinker, D.E. Clark, and D.R. Ulrich, "Better Ceramics Through Chemistry," North Holland (1984). 4 4 C.J. Brinker, D.E. Clark, and D.R. Ulrich, "Better Ceramics Through Chemistry JJ," Materials Research Society (1986). 4 5 C.J. Brinker, D.E. Clark, and D.R. Ulrich, "Better Ceramics Through Chemistry m," Materials Research Society (1988). 4 6 D.R. Clark, "Interpenetrating phase composites," J. Am. Ceram. Soc, 75 [4] 739-59 (1992). 97 S. Kwon and G. Messing, "Sintering of mixtures of seeded boehmite and ultrafine a-Alumina, J. Am. Ceram. Soc, 83 [1] 82-88 (2000). 4 8 G.L. Messing and M . Kumagai, "Low-temperature sintering of oc-alumina-seeded boehmite gels," Am. Ceram. Soc Bull., 73 [10] 88-91 (1994). 4 9 Q. Yang and T. Troczynski. "Dispersion of alumina and silicon carbide powders in alumina sol," J. Am. Ceram. Soc, 82 [7] 1928-30 (1999). 5 0 D.A. Barrow, T.E. Petroff, and M . Sayer. "Thick ceramic coatings using a sol-gel based ceramic-ceramic 0-3 composite," Surface and Coatings Technology, 16-11, 113-118 (1995). 5 1 Barrow et al, United States Patent, "Method for producing thick ceramic films by a sol gel coating process," Patent Number 5585136, Dec. 17, 1996. 5 2 H. Kominami, K. Matsuo, and Y . Kera, "Crystallization and transformation of aluminum orthophosphates in organic solvent containing a small amount of water," J. Am. Ceram. Soc, 19 [9] 2506-508 (1996). 5 3 R. Iwamoto, C. Fernandez, J.P. Amoureux, and J. Grimblot, "Characterization of Mo-P-alumina sol-gel catalysts by solid-state 3 1 P and 2 7 A l magic-angle spinning nuclear magnetic resonance and two-dimensional 2 7 A l multiple-quantum magic-angle spinning nuclear magnetic resonance," J. Phys. Chem. B, 102, 4342-4349 (1998). 5 4 J. Quartararo, M . Guelton et al., "Sol-gel synthesis of alumina modified by phosphorus: a solid state N M R characterization study," J. Mater. Chem., 9, 2637-2646 (1999). 5 5 H . van der Meer, "The crystal structure of a monoclinic form of aluminum metaphosphate, A1(P0 3) 3," Acta Cryst. B32, 2423 (1976). 5 6 R.A. Nyguist et al., "Handbook of infrared and raman spectra of inorganic compounds and organic salts, volume 4," Academic Press, San Diego, 1997. 5 7 L. Pauling, J. Shermany, "The crystal structure of aluminum metaphosphate, A1(P0 3) 3," Zeitschrift fur Kristallographie, Kristallgeometrie, Kristallphysik, Kristallchemie, 96, 481-487 (1937). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0078780/manifest

Comment

Related Items