1.1.1 Bioinert ceramics

Alumina and zirconia have been used as an important alternative to surgical metal alloys in total hip prostheses and as tooth implants. The main advantages of using ceramics over the traditional metal and polymer devices are lower wear rates at the articulating surfaces and the release of very low concentrations of ‘inert’ wear particles. For example, using femoral heads of alumina ceramic bearing against alumina cup sockets significantly reduces wear debris when against ultrahigh molecular weight polyethylene cups. Excessive wear rates can contribute to loosening and eventual implantation failure. Alumina ceramics have been used successfully for many years. Zirconia ceramics have advantages over alumina ceramics in terms of higher fracture toughness and higher flexural strength, combined with a relatively lower Young’s modulus (Table 1.1). Therefore, zirconia ceramics were developed for bearing surfaces in total hip prostheses. However, concerns about in-service failures (particularly the premature fracture of a batch of ceramic femoral heads) resulted in a Food and Drug Administration (FDA) recall. For this reason, the use of zirconia for strengthening and toughening of alumina matrix composites has been developed. One example is Biolox® delta (CeramTec), which has FDA approval for use in femoral head components.

Nanotechnology has also been applied to improve the properties of implant materials with the aim of extending the longevity of implant devices in the body, with no revision surgery necessary at the later time. To improve the fracture toughness of alumina ceramics, nanophase alumina with grain size of 23 nm were synthesised. The modulus of elasticity of nanophase alumina decreased by 70% (Webster et al., 1999). The fracture toughness of alumina can then be controlled through the use of nanophase formulations; furthermore, enhanced biological responses of osteoblast cells to the nanophase materials were found, indicating the improved osseointegration potential for nanophase alumina (Webster et al., 2000).

Alumina and zirconia have good biocompatibility, and adequate mechanical strength, but are relatively biologically inactive (nearly inert) and lack direct bonding with host tissue. Bioactive materials are conceptually different from bioinert materials in that chemical reactivity is essential. A series of bioactive ceramics, glasses and glass-ceramics are capable of promoting the formation of bone at their surface and of creating an interface, which contributes to the functional longevity of tissue.

1.1.2 Bioactive ceramics

Bioactive ceramics include several major groups, such as calcium phosphate ceramics, bioactive glasses and glass-ceramics.

1.2.1 HA and substituted HA

HA possesses a hexagonal lattice and a P6₃/m space group. This space group is characterised by a six-fold c-axis perpendicular to three equivalent a-axes at angles 120° to each other with cell dimension of a = b = 0.9418 nm and c = 0.6884 nm (Posner, 1969).

The mineral phase of bone, biological apatite, is not stoichiometric HA. The apatite is hospitable to a variety of cationic and anionic substitutions, and the type and amount of these ionic substitutions in the apatite phase varies from the wt% level (e.g. 3–8 wt% CO₃). to the ppm–ppb level(e.g.Mg^2 + or Sr^2 +). Substitution in the apatite structure for (Ca), (PO₄) or (OH) groups results in changes in properties, such as lattice parameters, morphology and solubility, without significantly changing the hexagonal symmetry.

The substitution of fluoride (F⁻ for OH⁻) has the consequence of increasing the crystallinity, crystal size and the stability of the apatite, which in turn reduces solubility. Fluoride substitution has been implicated in caries prevention, where its presence in enamel crystals increases stability. This helps to resist dissolution in the acidic oral environment (LeGeros and LeGeros, 1993).

Carbonate, CO₃, can substitute for either the hydroxyl (OH) groups, or the phosphate groups, and the resulting apatite is designated as Type A or Type B respectively. An important effect of carbonate substitution in HA is on crystal size and morphology. An increase in carbonate content leads to changes in the size and shape of apatite crystal (LeGeros et al., 1967) and the carbonate substituted apatites are more soluble than carbonate-free synthetic apatites.

Although silicon has only been found in trace quantities in bone mineral (up to a level of ~ 0.5 wt%), it has been shown to have a crucial role in bone mineralisation, and believed to be essential in skeletal development (Carlisle, 1970, 1972). In vitro and in vivo bioactivity was enhanced with the incorporation of silicate groups into the HA lattice (Gibson et al., 1999a; Patel et al., 2002, 2005). The silicate substitution in HA inhibited densification and grain growth at higher sintering temperatures (Gibson et al., 1999b, 2002), thus increasing the total surface area/volume ratio of grain boundaries (Porter et al., 2003, 2004), this may also have an effect on the in vivo responses. The surface charge of silicate substituted HA (SiHA) was significantly more negative than that of pure HA (Botelho et al., 2002), and may contribute to the faster bone-like apatite formation in vitro induced by SiHA. A higher concentration of protein absorption onto SiHA than HA was related to the formation of a silicate network structure, which interacts with integrins, thus triggering a signalling cascade and leading to consequent cell attachment, proliferation and differentiation. The addition of silicate in SiHA provides an extra chemical cue to stimulate and enhance bone formation, which promotes the attachment and proliferation of human osteoblast (HOB) cells (Fig. 1.1). The enhanced bioactivity of SiHA is the result of the increase in availability of Si as well as the favourable topography from increased grain boundaries with decrease in the grain size. Therefore, SiHA is a highly attractive alternative to conventional HA in bone replacement. SiHA as bone graft, commercially known as Actifuse™, has been used successfully for spinal fusion.

Porous strontium (Sr) and magnesium (Mg) co-substituted HA has been exploited to prolong the beneficial Mg release during the bone regeneration process as well as to utilise the anti-osteoporotic and cariostatic properties of Sr ions (Landi et al., 2013). Enhanced adhesion of osteoblasts was found on HA doped with yttrium (Y) compared with those doped with cadmium (Cd), zinc (Zn) or magnesium (Mg) (Webster et al., 2002).

Ti-substituted HA nanoparticles were found to inhibit the growth of four bacterial strains, including multi-antibiotic resistant epidemic MRSA (EMRSA) 15 and EMRSA 16 ‘superbugs’, which is the first step in the development of multifunctional dental and orthopaedic prostheses (Huang et al., 2011). Recent studies have shown that zinc-substituted HA (ZnHA) resulted in increased proliferation and differentiation of human adipose-derived mesenchymal stem cells, while there is a significant decrease in the number of viable Staphylococcus aureus bacteria after being in contact with ZnHA (Thian et al., 2013).

Besides the substitutions mentioned above, there are other substitutions, both cationic (substituting for calcium with barium, lead) and anionic (substituting for the phosphate with vanadates, borates, manganates). Shepherd et al. summarised some of the key effects of substitutions with magnesium, zinc, strontium, silicon and carbonate on physical and biological characteristics of HA, such as thermal stability, solubility, osteoclastic and osteoblastic response in vitro and degradation and bone regeneration in vivo (Shepherd et al., 2012).

1.2.2 Bioactive glasses

Polycrystalline ceramics are solids in which the atoms or ions are arranged in regular array. In contrast, the regularity (order) is only short range in glass (amorphous materials), because a glass formed when a molten ceramic composition is rapidly cooled while the atoms do not have time to arrange themselves in a periodic structure. A number of bioactive glasses have been developed and investigated for tissue engineering and probably the best known of these is Bioglass® (Hench, 2006), while a three-dimensional (3D) SiO₂ network is modified by incorporation of Na₂O, CaO and P₂O₅. The composition of Bioglass® makes the surface highly reactive when exposed to an aqueous medium, leading to in vitro and in vivo bioactivity.

Furthermore, sol–gel-derived bioactive glasses have a porous texture in the nanometre range, giving them a surface area of 150–600 m² g⁻¹, which is two orders of magnitude higher than that of melt-derived glasses. Dissolution is therefore more rapid for sol–gel glasses at similar composition, and more silanol groups are on sol–gel glass surfaces to act as nucleation sites for formation of apatite layer, leading to high bioactivity. A comprehensive review of bioactive glass has recently been done by Jones (2013).

1.2.3 Bioactive glass-ceramics

A glass-ceramic is polycrystalline solid prepared by the controlled crystallisation or devitrification of a parent glass. It generally consists of fine grains (with crystal sizes ranging from 0.1 to 10 μm) and has a small volume of residual glass sited at the grain boundary. One advantage of glass-ceramics is that the crystallisation and formation of the crystal phases can be controlled to develop materials with a combination of special properties, such as bioactivity, machinability and improved mechanical properties. A-W glass-ceramic is the most extensively studied glass-ceramic for use as bone substitutes (Kokubo, 1991). It combines high bioactivity with desirable mechanical properties and has been successfully in the load-bearing spinal area of the body.

1.4.1 Mechanical properties

The mechanical properties of a material are important as they determine its structural applications. Table 1.1 summarises some of mechanical properties of bioactive and bioinert ceramics.

Strength can be measured in a number of different ways, such as uniaxial tension, three-point bending, four-point bending and uniaxial compression. Tensile strength testing is typically used for characterising ductile metals. Ceramics are not normally characterised by tensile testing due to the high cost of test specimen fabrication and the requirement for extremely good alignment of the load during testing. Compressive strength is the crushing strength of a material, and is commonly measured for ceramics, especially those that must support loads. The strength of ceramics materials is generally characterised by bending testing, also referred to as flexural testing. Apart from understanding these uniaxial stresses of a material, biaxial strength provides the data under a biaxial stress condition (e.g. under both tensile and shear stresses) since many applications for materials impose multi-axial stress fields. The strength value is dependent on the type of test conducted, flaw size distribution of the material and the stress distribution in the test specimen. Based on the concept of the failure of the weakest link, the strength distribution of ceramic materials can be described as a Weibull modulus, a dimensionless number used to characterise the variability in measured strength of brittle materials (ceramics) which arises from the presence of flaws with a distribution in size and orientation. The higher the Weibull modulus is, the more consistent the material, which means uniform defects are evenly distributed throughout the entire volume.

Fracture toughness is one of the key measures of the mechanical properties of ceramic materials. Due to the brittle nature of ceramics, it is essential to consider fracture in terms of crack surface displacement and the stresses at the tip of the crack using a fracture mechanics approach. The stress concentration at a crack tip is denoted in terms of the stress intensity factor K. Additional subscripts refer to the direction of load application with respect to the position of the crack. In the tensile and bending tests the load is perpendicular to the crack, and Mode I is most frequently encountered for ceramic materials. K_IC is the stress intensity factor at which the crack will propagate and lead to fracture. It is also referred to as fracture toughness and is considered as a basic property of ceramic materials. The higher the fracture toughness, the more difficult it is to initiate and propagate a crack.

The hardness of a material is a measure of its resistance to localised deformation by indentation or scratching. A small indenter is pressed into the surface of the material and the size of the indent is measured to calculate a hardness value. The hardness and fracture toughness of a ceramic influence its wear behaviour. The wear properties of bioceramics are important for their application in joint replacement, as wear debris can trigger the inflammatory responses in the surrounding tissue and eventually lead to failure of the implant.

1.4.2 Surface properties

The physicochemical properties of implant materials, such as surface energy and charge, hydrophilicity or hydrophobicity, tend to affect cellular response by influencing protein absorption and cell attachment. Therefore, the surface properties of materials can have a great impact on cellular responses. The nature and development of a stable interface between an implanted bioceramic and bone, which is crucial for the clinical success of the implant, are affected by many factors. It is becoming recognised that a key aspect is the surface modification of the bioceramic, which occurs due to interaction with the local environment. For materials, such as bioactive glasses, a series of surface reactions occur when immersed in physiological solution. Exchanges of Na⁺ and K⁺ with H⁺ and H₃O⁺ ions from solution at the glass surface lead to the loss of soluble silica and the formation of Si-OH (silanols) at the glass–solution interface. This stage is followed by the migration of calcium and phosphate ions through the silica-rich layer to the surface and the formation of an amorphous calcium phosphate layer, which can then crystallise by the incorporation of hydroxyl, carbonate and fluoride ions to create an apatite layer. In the presence of osteogenic precursors, bioactive glasses favour the formation of osteoblasts which govern the further steps of bone development (Hench, 2006).

Enhancing the interaction between tissue and biomaterials is another way to achieve a desirable tissue response to implant materials. Cells recognise surface features and react to them, resulting in contact guidance (Curtis and Wilkinson, 1997). A key design parameter for achieving maximal cell responses is material topography. There is increasing evidence that surface topography both on the micro and nano-scale are important in determining the cell response to biomaterials (Gray et al., 1996; Dalby et al., 2002, 2007). The creation of micro- and nano-scaled surface topography using HA has been attempted by electrohydrodynamic spraying and print-patterning with the aim of up-regulating cell activity (Huang et al., 2004; Ahmad et al., 2006; Munir et al., 2011).

In order to understand the mechanism of bonding of a bioactive material with host tissue, it is necessary to characterise the surface of a material in vitro, both acellular (in physiological solutions) and cell cultures, as the specific tissue compatibility of a material is highly dependent on the composition and structure of surface layers. Such in vitro analysis will help in understanding the potential in vivo host tissue responses and will provide the material characteristics information for developing new tissue engineering scaffold materials.

There are two approaches, solution analysis and surface analysis, for studying the mechanism and reaction at material surface. In solution analysis, the constituents released into the surrounding environment (potentially those will be released into the tissue) are examined. The leaching of alkali and alkaline earth ions from bioactive glass and ceramic surface, the dissolution of silica and the precipitation of calcium phosphates are revealed by this method. The dissolution products of bioactive glasses have a positive effect on the expression of genes regulating osteogenesis (Xynos et al., 2000, 2001). Therefore, the quantification of the release products from bioactive glasses and ceramics is important for their application as scaffolds for the formation of bioengineered bone tissue.

Surface analysis is used to examine the surface of the material using instrumental tools such as AFM, Auger electron spectroscopy, infrared reflection spectroscopy, thin-film X-ray diffraction and SEM with X-ray analysis. Solution analysis can provide information about the total depths of reaction on the surface of a material; however, if precipitation occurs in the system (on the material surface or elsewhere), it will yield misleading information about the surface reactions of a material. By using instrumental tools, the surface structure regarding either the top 1.5 μm or only the top 0.5–5 nm of the sampling depth can be obtained. Combining surface or solution analyses, the characteristics of a reactive material throughout top and deep zones of the surface can be revealed.

1.4.3 Bioactivity and biocompatibility

1.5.1 Preparation of HA ceramics

There are numerous methods for the preparation of synthetic apatites, which can be grouped as aqueous reactions, solid-state reactions and hydrothermal reactions. The aqueous reactions may be divided into chemical precipitation and hydrolysis methods. Chemical precipitation is the most commonly used method, because of its simplicity and ability to produce a wide variety of particle sizes and morphologies.

Methods based on those described by Akao et al. (1981) (equation 1.1) and Hayek and Newesely (1963) (equation 1.2) are the most frequently used. They consist of the dropwise addition of phosphate solution into a stirred solution of calcium solution. The addition of ammonium hydroxide is needed to keep the pH of the reaction alkaline to ensure the formation of HA after sintering the precipitate.

$10\mathrm{Ca}{\left(\mathrm{OH}\right)}_2+6{\mathrm{H}}_3\left({\mathrm{PO}}_4\right)\to {\mathrm{Ca}}_{10}{\left({\mathrm{PO}}_4\right)}_6{\left(\mathrm{OH}\right)}_2+18{\mathrm{H}}_2\mathrm{O}$ [1.1]

[1.1]

$\begin{array}{l}10\mathrm{Ca}{\left({\mathrm{NO}}_3\right)}_2+6\left({\mathrm{NH}}_4\right){}_2\mathrm{H}\left({\mathrm{PO}}_4\right)+2{\mathrm{H}}_2\mathrm{O}\to {\mathrm{Ca}}_{10}\left({\mathrm{PO}}_4\right){}_6\left(\mathrm{OH}\right){}_2\hfill \\ +12{\mathrm{NH}}_4{\mathrm{NO}}_3+8{\mathrm{HNO}}_3\hfill \end{array}$ [1.2]

[1.2]

The concentrations of reagents must be such that the Ca/P molar ratio is maintained at 1.67 for stoichiometric HA. The concentration of calcium can be adjusted if substitution for calcium (strontium, magnesium, etc.) is required. Similarly, the phosphate concentration can be adjusted and replaced with required amount of carbonate or silicate when carbonate or silicate substitution is desired. Fluoride or chloride substitute apatite can be prepared by addition of fluoride or chloride ions in the reactions.

The next step of ceramic processing is to break down the materials received from chemical synthesis, which is a solid aggregation of particles in a dried, filtered precipitate. The agglomerates have a deleterious effect on the properties of the ceramic, and therefore need to be broken down by crushing and grinding. Milling is then used to further reduce the particle size. Particle size reduction is vital for good sinterability, but the ability to handle the powder is equally important, as it ensures the powder flows properly and can be compacted practicably. Spray drying is a widely used method to transform powder to soft agglomerates for easy handling. Calcination, a heat treatment, is another means of improving powder handling.

The HA powder obtained can then be made into dense or macroporous products using compaction (die pressing, isostatic pressing, slip casting, etc.) followed by solid-state sintering. The properties of powder, such as morphology, surface area, mean particle size and particle size distribution, need to be adequately characterised, as this will greatly influence the handling and processing.

Consolidation or compacting is the final stage in the powder preparation. The compacted or compressed body is usually sintered at temperatures of 950 to 1300 °C. The processing of densifying a powder compact without the presence of a liquid phase is called solid-state sintering. During solid-state sintering the material moves to eliminate the pores and open channels that exist between the grains of the compact, the crystals become tightly bonded together at their grain boundaries and the density, strength, toughness and corrosion resistance of sintered material increase greatly.

1.5.2 Porous ceramics

Porous ceramics have attracted great interest as scaffolds for tissue engineering, particularly bioactive ceramics and glasses, as they are able to bond to the host tissues. To be able to regenerate a tissue, a scaffold should act as a template for tissue to grow in three dimensions. The template must be a network of large pores (macropores, at least 100 μm in size) and the pores must be connected to each other, thus allowing essential nutrients to reach the whole network and stimulating blood vessels to grow inside the pore network.

Macroporosity can be introduced by mixing the powder with a volatile component, e.g. hydrogen peroxide or naphthalene, or adding polymethyl methacrylate (PMMA) beads to the powder slurry, using porous polymer (polyurethane foam) as templates for impregnating with ceramic suspensions. The volatile components or polymer phases with low decomposition temperature are removed during the evaporating and sintering process. The ceramics component with porous structure remain.

Ceramic slurries can also be foamed to obtain a porous structure. The incorporation of bubbles is achieved by injection of gases though the fluid medium, mechanical agitation, blowing agents and evaporation of compounds. A surfactant is generally used to stabilise bubbles formed in the liquid phase by reducing the surface tension of the gas–liquid interface. The gel-casting method has been used to produce macroporous HA with interconnected pores. The compressive strengths of HA foams were above 10 MPa, which is similar to that of trabecular bone (Sepulveda et al., 2000).

Porous HA has also been produced by hydrothermal transformation from reef-building corals. These methods employ the use of elevated temperatures, pressures and controlled atmospheres to convert calcium carbonate skeleton into HA. The route has the benefit of preserving the original architecture, the corals serving as a template to make a porous structure (Roy and Linnehan, 1974).

The design and control of the internal architectural of the porous HA structure influence the tissue regeneration (Chu et al., 2002). Rapid prototyping (RP) has emerged as a new processing technique for making scaffold, which allows highly complex structures to be built as a series of thin 2D slices using computer-aided design (CAD) and computer-aided manufacturing (CAM) programs. This technique allows properties such as porosity, interconnectivity and pore size to be predefined. Rapid prototyping, especially 3D printing, has been developed for making custom-made 3D porous HA scaffolds for bone replacement, and the repair of osseous defects from trauma or disease. A 3D HA structure with controlled patterns or porosity was built by direct-write assembly (Michna et al., 2005). A complex-shaped porous HA ceramic with fully interconnected channel was generated from HA powder (Seitz et al., 2005), which was printed with a binder solution layer by layer. Unglued powder was removed and the obtained ceramic green body was consolidated by sintering at high temperature. It is possible to design and manufacture parts according to individual patient’s anatomy. Patient-derived cells can then be seeded onto the scaffolds for tailor-making tissue engineering implants.

1.5.3 Glass

Melting and sol–gel processing are two well-known methods for producing glasses. Traditional glass synthesis consists of melting the precursor mixture and quenching. New components can be added to the system to tailor the glass composition for clinical applications, i.e. to increase or decrease bioactivity, decrease the glass-forming temperatures, etc. An alternative approach is the use of sol–gel techniques to prepare glasses. This route can produce high purity glasses, which are more homogeneous than those obtained by melting, and require relatively low processing temperatures. It involves the transformation of a sol (suspension of colloidal particles) to a gel (3D, interconnected network). Colloids are solid particles with diameter less than 100 nm. Thermal treatment increases the density, strength and hardness of the gels and converts the network to a glass with properties similar to melt-derived glass. An advantage of the sol–gel process is the ability to control the surface chemistry of the material by the thermal treatment and make it possible to expand the bioactive compositional ranges studied in the phase diagram of melted glasses. The glasses obtained exhibit higher surface area and porosity, the critical factors in their bioactivity.

Glass in the fibre form can be produced by the melt-spinning approach. The fibre diameter is limited to micrometre-scale (e.g. tens to hundreds of micrometres). Electrospinning is the technique able to produce various polymer fibres in the range of 10–1000 nm, which has been applied to produce bioactive glass fibres in the micro-/nanoscale using sol–gel glass. Nanoglass fibres in the forms of bundled filaments, fibrous membranes and 3D scaffold have been produced (Kim et al., 2006).

Porous bioactive glasses have been produced by foaming of melt-derived and sol–gel-derived bioactive glasses (Sepulveda et al., 2002; Jones and Hench, 2003; Chen et al., 2006). During the foaming process of sol–gel glass, air was entrapped in the sol under vigorous agitation as viscosity increased and the silica network formed. As the porous foam became a gel, the bubbles were stabilised. The gel was then subjected to a controlled thermal processing of ageing (60 °C), drying (130 °C) and sintering to remove organic species (500–800 °C). The resulting bioactive glass foam scaffolds had macropores up to 600 μm in diameter and compressive strength up to 2.5 MPa (Jones, 2005). An overview of structural characteristics and mechanical properties of highly porous bioactive ceramic or glass for bone tissue engineering was summarised by Chen et al. (2006).

1.5.4 Glass-ceramics

Glass-ceramics are produced by the transformation of the glass into a ceramic. The glass is firstly heated at the temperature of 450–700 °C to produce a large number of nuclei, then the temperature is increased to 600–900 °C to promote crystal growth. The resulting microstructure is finegrained with a uniform size distribution.

1.5.5 Coating processing

Although the relatively poor mechanical properties of calcium phosphate ceramics limit their clinical applications to non-major load-bearing parts of the skeleton, calcium phosphate-coated metallic implants are used in the load-bearing parts. The most popular commercial routes of calcium phosphate coatings are based on plasma spraying (de Groot et al., 1987). The ceramic powder is suspended in the carrier gas and fed into the plasma, where it can be fired at a substrate. However, the high temperature involved in the processing can lead to the changes of phase purity and crystallinity of calcium phosphate ceramics. A variety of thin film and low temperature techniques have been developed to deposit a bioactive layer on the surface of bioinert materials, such as electrostatic atomisation spray deposition (Huang et al., 2005; Leeuwenburgh et al., 2005), sol–gel deposition (Gross et al., 1998), biomimetic deposition (Abe et al., 1990; Habibovic et al., 2002) and magnetron sputtering (Jansen et al., 1993; Thian et al., 2007). Condensation from a vapour by sputtering has been used to produce nanosized coatings, which are uniform, nonporous and very fine grains. Nanocrystalline SiHA coating by radiofrequency (RF) magnetron sputtering has been found to promote better biological responses (Thian et al., 2006).