Advanced ceramic materials have attracted increasing attention throughout the 1980s from many disciplines including chemistry, physics, metallurgy and materials science and this multidisciplinary approach is illustrated by the diverse range of journals and conferences where information is disseminated. In addition the discovery of high-temperature ceramic superconductors in 1986 has raised the profile of advanced ceramics activities not only within the scientific community but also among the general public. Attendance at conferences and surveys of scientific literature show that chemical synthetic methods have played an increasing role, over the past fifteen years, in improving the properties of ceramic materials. Books concerned with fabrication and physical properties of ceramics do not, in my opinion, highlight chemical aspects of ceramic preparations which are not the principal interest of physical, organic and inorganic chemistry textbooks.

My discussions with undergraduate and postgraduate students in chemistry and materials science as well as university lecturers and those in industry concerned with research into and manufacture of advanced ceramics produced two conclusions. Firstly, there did not seem to be a short volume available which acted as a bridge between pure chemistry and conventional ceramic studies such as fabrication. Also, although scientific publications and conference proceedings proliferate it was not obvious how a comprehensive view of the rapid inroads chemistry is making into ceramic synthesis could be obtained. I see this book as that bridge between pure chemical and conventional ceramic studies. I have included a chapter on fabrication for continuity but this is not the main theme.

Introduction

The international advanced ceramics industry is concerned with basic research and ceramic fabrication as well as manufacture of powders and fibres while the success of research can be measured by its application to large-scale economic production of ceramics which function in particular working environments. Advanced ceramic materials are defined in this introductory chapter and their variety and uses are explained. The ceramics industry is large and an indication of its volume production and monetary value is also given here. Finally, the recent discovery of high-temperature oxide superconductors has had a tremendous impact on worldwide ceramic activities and a section is included on the properties and potential applications of these advanced ceramic materials.

From traditional to advanced ceramics

Ceramics are the group of non-metallic inorganic solids and their use by man dates from the time of ancient civilisations. In fact, the word ceramic is of Greek origin and its translation (keramos) means potter's earth. Traditional ceramics are those derived from naturally occurring raw materials and include clay-based products such as tableware and sanitaryware as well as structural claywares like bricks and pipes. Also in this category are cements, glasses and refractories. Examples of the latter are chrome–magnesite refractories used in the steel-making industry and derived from magnesite (MgCO3) and chrome ore. Advanced ceramics are developed from chemical synthetic routes or from naturally occurring materials that have been highly refined. A variety of names has been used to describe ceramic systems.

Introduction

Precipitation from solution, powder mixing and fusion are all conventional techniques for synthesis of traditional and advanced ceramics on both the laboratory and industrial scale. These methods, with selected examples, are described in this chapter together with limitations of their use for advanced ceramic materials. There is increasing demand for alternative routes to ceramic materials that impart superior properties compared with those attainable from conventional syntheses and the reasons for this continuing search for other synthetic pathways are also described here.

Precipitation from solution

Alumina occurs as the mineral bauxite and is refined in the Bayer process whereby ore is initially dissolved under pressure in sodium hydroxide so that solid impurities (SiO2, TiO2, Fe2O3) separate from sodium aluminate solution (Evans & Brown, 1981). This solution is either seeded with gibbsite crystals (α-Al2O3.3H2O) or undergoes auto- precipitation to bayerite (β-Al2O3.3H2O) after its neutralisation with CO2 gas. Temperature, alumina supersaturation and amount of seed affect particle size during crystallisation but, as for other precipitation reactions, the product is agglomerated (section 9.3). Alumina production from the Bayer process was 31.9 × 106 Mg in 1985; 90 weight % of this was reduced to metal and only 5 weight % found ceramic and refractory applications (MacZura, Carbone & Hart, 1987).

Introduction

The title of this chapter refers to synthesis of advanced ceramic powders by using non-aqueous liquid media, which can be inert or constitute a reactant, for example ammonia, and at the time of writing is particularly associated with silicon nitride. There has been increasing interest during the 1980s in developing low-temperature liquid-phase syntheses, which has resulted in the industrial production of certain powders. Reactions between ammonia and silicon tetrachloride are described in this chapter, together with those involving other chlorosilanes, ammonia and amines. Finally, the preparation of non-oxide powders, as well as Si3N4, with liquid-phase reactions are outlined. Work described here and in chapter 9 highlights the importance of controlling reaction conditions in order to reduce impurity levels such as chlorine in Si3N4. The significance of these levels may not be appreciated when synthesis is viewed strictly from a chemical approach, and without also considering the ceramic applications of, for example, high-temperature structural components.

The reaction between silicon tetrachloride and ammonia in the liquid phase

Persoz (1830) carried out initial studies on the liquid-phase reaction between silicon tetrachloride and ammonia. He considered that the white precipitate obtained on passing NH3(g) through SiCl4 dissolved in benzene at ca. 273 K was silicon tetramide, Si(NH2)4, a conclusion also drawn by Lengfeld (1899) and later by Vigoureux & Hugot (1903).

Introduction

The phrase sol–gel describes several types of processes in different areas of chemistry and materials development, while the term ‘gel’ has been used (Flory, 1974) to embrace a wide range of substances in systems as diverse as lamellar mesophases, inorganic clays and oxides, phospholipids, disordered proteins and three-dimensional or network polymers. There are three types of sol–gel processes associated with corresponding transitions. Examples of the first are the reversible gelation of certain polysaccharide solutions, for example, agarose and the vulcanisation of rubber, but these transitions will not be considered in this book. The other two sol–gel processes have attracted intense interest since the mid-1970s because of their use in the synthesis of ceramic materials. Sol–gel processing of metal–organic compounds, namely, alkoxides, is described in chapter 5, while the early study of colloids, their stability and use in sol–gel transitions for both nuclear and industrial ceramics are the subject of this chapter.

The nature of colloids

The scientific study of colloids dates back to 1845 when Selmi prepared silver chloride dispersions (sols) followed by Prussian blue sols in 1847, which he referred to as demulsions and pseudosolutions, respectively. These systems were considered to be in the same category as starch, cholic acid and albumin, but distinct from true solutions, following experiments on the effect of salt on their sedimentation. Faraday (1857) observed light-scattering from ruby-coloured gold sols made by reduction of gold chloride with phosphorus together with the effect of salt on their stability and colour.

Introduction

The previous six chapters have been concerned with non-conventional preparative routes to ceramic materials, that is, techniques which avoid powder mixing and milling or coprecipitation of hydroxides and oxalates. Synthetic routes not associated with sol–gel processing of both colloids and metal–organic compounds, non-aqueous liquid-phase reactions, polymer pyrolysis, hydrothermal synthesis and gas-phase reactions are the subject of this chapter. These methods are the citrate gel process, alkoxide pyrolysis, freeze-drying and rapid expansion of supercritical solutions. Different drying processes, which have been mentioned in earlier sections, are discussed here with particular reference to their role in modifying the sintering behaviour of ceramic powders.

The citrate gel process

The citrate gel process, which was developed by Marcilly & co-workers (1970), can be illustrated by the synthesis of Gd2O3-stabilised zirconia (van der Graaf & Burggraaf, 1984). In order to form an organometallic complex, citric acid was added to mixed zirconyl and Gd(III) nitrate solutions (acid/metal mole ratio = 2:1) whose pH had been increased with NH4OH to between 6 and 7.5. Other organic acids containing at least one hydroxyl and one carboxylic group such as tartaric, lactic and glycollic acid can be used. Rapid partial dehydration produced highly viscous mixtures that could be dried to amorphous gels at 373 K. The viscous ceramic precursor swelled on further heating because of decomposition of NH4NO3 after which the organometallic complex decomposed exothermically.

Introduction

Solid ceramic bodies are generally produced by using the process of powder compaction followed by firing at high temperature. Sintering or densification occurs during this heat treatment and is associated with joining together of particles, volume reduction, decrease in porosity and increase in grain size. The phase distribution or microstructure within the ceramic is developed during sintering and fabrication techniques used for shaping ceramics are described in this chapter. The aim of these techniques is to produce microstructures suitable for particular applications. Hence a fine-grained distribution is required for strength. Controlled grain size is necessary where optical properties such as translucency are required. Strength and toughness of ceramic systems are also discussed here with particular reference to the role of powder preparation on the strength of ceramics.

Solid-state sintering

The driving force for sintering is reduction in surface free energy associated with a decrease of surface area in powder compacts due to removal of solid–vapour interfaces. Vapour-phase nucleation is described in chapter 9 by using the Kelvin equation (9.1), which is also applicable to mass-transport processes in a consolidated powder (Kingery, 1983). The vapour-pressure difference across a curved interface can enhance evaporation from particle surfaces and condensation at the neck between two particles, particularly for particle diameters of several micrometres or less, such as occur in ceramic fabrication. Although this evaporation–condensation process produces changes in pore shape and joins particles together, the centre-to-centre distance between particles remains constant so that shrinkage and densification do not occur.

Introduction

Sol–gel processing of colloids described in chapter 4 allows the preparation of homogeneous compositions and crystalline phases at temperatures lower than required for conventional powder mixing. However, sol particles contain between 103 and 109 molecules and improved mixing of components can occur by interaction at the molecular rather than colloidal level. This is achieved during the hydrolysis of metal–organic compounds, namely alkoxides. Sol–gel processing of alkoxides has attracted intense interest in the past 15 years because it offers non-melt routes to high-purity glasses and crystalline ceramics. The synthesis and hydrolytic reactions of alkoxides are described in this chapter, together with their use in the preparation of a wide range of ceramic materials.

The synthesis of metal alkoxides

Metal alkoxides have the general formula M(OR)Z and can be considered to be derivatives of either an alcohol, ROH where R is an alkyl chain, in which the hydroxyl proton is replaced by a metal M, or of a metal hydroxide, M(OH)Z.

Ebelman (1846) made the first synthesis of an alkoxide, silicon tetraisoamyloxide, by reacting silicon tetrachloride with isoamyl alcohol. This was followed (Ebelman & Bouquet, 1846) by the preparation of boron methoxide, ethoxide and amyloxide using boron trichloride and the corresponding alcohol. Electronegativity of the main element is an important factor in the choice of synthetic route so that strongly electropositve metals with valencies up to three react directly with alcohols to give alkoxides together with the liberation of hydrogen.