**2. Constrained sintering by rigid inclusions: Solid-state sintering of ceramiccomposites**

In nature the combination of various materials is a constant search for an optimization of properties and functions. By analyzing the different tissues and organs that constitute the human body is possible to verify that they typically result from a combination of different types of biological materials. As an example, natural bone is an ideal nanocomposite, which consists of approximately 30% of matrix material and 70% of nanosized mineral (matrix material- collagen fibers (polymer) plus mineral - HA crystals (ceramic)).

In the development of biomaterials that strategy is also widely used. A variety of materials that result from the combination of two (or more) materials, with composition and structure and different properties that display the higher end of its individual components are composites. One of the most important characteristics of composites is the possibility of modifying the properties with the change of only one of several processing variables such as size, shape, distribution and orientation of the constituents, among others. The principal material of a composite that involves the "reinforcement" is called matrix, with regard to the extent in the body, and the second component (e.g. SiC particles or fibers, ZrO2 particles, etc.) is usually referred as the filler (or inclusion).

When attempting to join different materials, several factors need to be considered to ensure the integrity of the resulting structure. These include the physical and chemical properties of each component (reactivity, coefficient of thermal expansion, tensile and compression strength, etc.) and the processing conditions (temperature and atmosphere).

Kim, 2003; W. Li & Gao, 2003), alumina (J. Li et al., 1995), and apatite (Agathopoulos et al., 2003), among others. However, the combination of HA and ZrO2 have attracted a great attraction for applications to bone tissues exposed under high friction and high impact due to the potential mechanical properties such as high fracture toughness and hardness as well as bioinertness of the ZrO2 component. A higher strength and fracture toughness can be significantly enhanced either by stress-induced tetragonal (*t*-ZrO2) to monoclinic (*m*-ZrO2) phase transformation toughening, or by a deflection toughening mechanism. The mechanical behavior is strongly dependent on the structure of the material. Moreover there are basic principles that govern the behavior and properties of a material, which are described as a ratio of it internal structure, processing and their properties. When one aspect

The processing of materials is another key parameter that will have implications on the final properties of the biomaterial. Many researchers have observed that the mechanical strength and fracture toughness of HA based ceramics can be improved by the use of different sintering techniques which include hot isostatic pressing (HIP), spark plasma sintering (SPS) and microwave sintering. Also a sintering aid based on a low sintering temperature may be helpful if an easy and cost-effective sintering technique has to be used. On the other hand, to minimize the thermal decomposition and improve the densification of the HA/ZrO2 composites, various atmospheres assisted sintering can be employed. Moreover, the addition of a low melting secondary phase to achieve liquid phase sintering for better densification, incorporation of sintering additives to enhance densification through grain boundary strengthening, and use of nanoscale ceramic powders from sol-gel process for

better densification contributed to desirable mechanical and biological properties.

material- collagen fibers (polymer) plus mineral - HA crystals (ceramic)).

strength, etc.) and the processing conditions (temperature and atmosphere).

etc.) is usually referred as the filler (or inclusion).

**composites** 

**2. Constrained sintering by rigid inclusions: Solid-state sintering of ceramic-**

In nature the combination of various materials is a constant search for an optimization of properties and functions. By analyzing the different tissues and organs that constitute the human body is possible to verify that they typically result from a combination of different types of biological materials. As an example, natural bone is an ideal nanocomposite, which consists of approximately 30% of matrix material and 70% of nanosized mineral (matrix

In the development of biomaterials that strategy is also widely used. A variety of materials that result from the combination of two (or more) materials, with composition and structure and different properties that display the higher end of its individual components are composites. One of the most important characteristics of composites is the possibility of modifying the properties with the change of only one of several processing variables such as size, shape, distribution and orientation of the constituents, among others. The principal material of a composite that involves the "reinforcement" is called matrix, with regard to the extent in the body, and the second component (e.g. SiC particles or fibers, ZrO2 particles,

When attempting to join different materials, several factors need to be considered to ensure the integrity of the resulting structure. These include the physical and chemical properties of each component (reactivity, coefficient of thermal expansion, tensile and compression

of this relationship is changed the others will be affected accordingly.

One of the most popular paths to the formation of polycrystalline ceramic composites involves powder compaction into a porous body which would receive sintering treating rather than melting or other methods. The reasons for that are two: First, ceramics usually melt at high temperatures, which makes melting difficult and inefficient. Second, ceramics are brittle which is not suitable for processing by thermo-mechanical forming. Sintering commonly refers to the process by which a system of powder particles in contact is consolidated to form dense polycrystalline aggregates through physical and chemical changes when subjected to appropriate temperatures. Almost all aspects of the sintering process have been addressed in many books and publications (Exner, 1980; Exner & Arzt, 1996; German, 1996; Kingery et al., 1976; Kingery & Berg, 1955; Coble & Gupta, 1967).

Sinterization is one of the most important manufacturing and delicate steps that a ceramic body is subject. The most important because it determines, in large part, the final properties of the ceramic, such as microstructure, mechanical properties and final crystalline phases and in the case of bioceramics, their behavior in service. Is one of the more delicate because the achievement of the desired final properties is dependent of the processing parameters such as material properties (e.g. melting temperature, Tm), impurities, temperature and time, initial compact density, initial particle size distribution, and applied pressure. Usually a selfsupporting compact is obtained from powders that are compacted uniaxially or isostatically to form a "*green body*", followed by a densification stage involving sintering at appropriate temperatures. So, temperature, time, initial particle size of the material and sintering atmosphere, nature of additives, etc. are the most important factors which can influence the sintering process and the resulting microstructure.

The heat-assisted treatment of sintering usually improves the mechanical strength of the material through the formation of a solid bond between the particles. This process is generally accompanied with densification (i.e. elimination of the pores) and strengthening of the compact, a process driven by interfacial energy (γ). From a macroscopic point of view, during sintering not only dimensional changes of the products occurs, i.e. the compacts usually contract as the material densifies, but also their shape may vary as a consequence of anisotropic sintering contraction, phase transformations, etc. Moreover, anisotropic shrinkage behavior of composite powders can cause formation of coalesced pores and other microstructural changes in sintered compacts. Hence, based on the application and the expected properties of the final product, there are different techniques to control the mechanism and stages of the sintering process.

The driving force to bond individual particles together, during sintering, is the reduction of free energy of surfaces of powder compacts. This can be achieved by the elimination of pores and reduction of grain boundaries via grain growth. Since γsv(surface free energy of the solid-vapor interface) is normally greater than γss (surface free energy of the solid-solid interface), the solid-vapor interfaces tend to be replaced by solid-solid interface when enough energy is provided. Through the driving force, the neck growth between adjacent particles results in a bonded network polycrystalline microstructure with the pores trapped inside, as shown in Fig.1a.

Through further heating of the compact, the residual pores become smaller in size and number, and isolated from each other (Fig.1b). These isolated pores could completely disappear in the final stages of sintering depending on processing parameters.

New Challenges in the Sintering of HA/ZrO2 Composites 183

Also, differences in rates of sinterization between inclusions and matrix may develop stresses which cause sintering defects such as cracks and isolated pores. In fact, the matrix sintering is inhibited when it tries to densify and contract around the inclusions. In this situation, the mean stress will be compressive on the inclusion and tensile on the matrix, thereby opposing densification. The internal stress field will also depend on the shape of the second phase and its volume fraction. It was also reported that the densification rate decreases with increasing volume fraction of inclusions and that densification apparently stops before the theoretical density is reached. It was shown (Brinker & Scherer, 1990) that when inclusions inhibit densification, coarsening mechanisms such as surface diffusion are favored over sintering. Although several improvements in the mechanical properties (e.g. toughness) of ceramic matrices can be achieved by introducing certain reinforcing particles, they tend to inhibit shrinkage so that the full density of the compact may never be achieved,

unless a pressurized process, such as hot-pressing or other approaches is applied.

coarsening competition, and producing ceramics with desirable microstructures.

regions of the matrix limits the final densification of the composite.

Fig. 2. SEM microstructure of a HA compact after sintering at 1300 ºC for 2 h.

of the phases (matrix and dispersed phase) properties.

The success in powder consolidation is intimately related to the control of the competition between densification and coarsening. The growth of larger particles, as shown in Fig.2, and shrinkage of small ones (coarsening) is observed often during sintering of crystalline materials. Control of grain growth is thus an essential aspect of controlling the densification-

The volume fraction of a particulate second phase present in a composite is an important factor in sintering and resultant shrinkage, and the shape and size of the reinforcing phase may also influence the sintering behavior of the matrix material. The effect of heterogeneities, especially, rigid inclusions, on the matrix densification has been a subject of several studies (Bordia & Scherer, 1988; Scherer, 1987). It was shown that at the first stage of densification, the volume fraction, shape, and distribution of the inclusions affects the morphology of the network formed by the differential densification of the matrix. The matrix properties determine whether, in later stages of densification, the interstitial porous regions can reach full density or not. It was concluded that grain growth in the dense

Many properties of composites are predicted on the basis of the rule of mixtures, which is a method based on an assumption that a composite property is the volume weighed average

**Abnormal grain growth** 

**2** μ**m** 

Fig. 1. (a) Sintering model geometry of sphere particles. (b) SEM microstructure of a HA compact, sintered at 1100 ºC for 1 h.

During sintering, material moves through mechanisms of viscous flow or diffusion (or both) in order to eliminate porosity and reduce the γsv. Diffusion is among the most important phenomena observed in sintering of ceramic materials. Diffusive mass transport takes place when there is a gradient in the chemical potential and when the species in question has sufficient mobility. There are a number of competing paths for mass transport during ceramic sintering (Garay, 2010), such as grain boundary diffusion, volume diffusion and surface diffusion. Usually surface diffusion, leads to coarsening, which is the growth of the neck between particles leading to the reduction of the specific surface area. The other transport mechanisms, normally volume diffusion and grain boundary diffusion lead to densification.

Most of ceramic particles, including oxides, need elevated temperatures to be sintered to high levels of densification. Densification of covalent-ionic ceramics without additives is extremely difficult due to the low diffusivity in the solid state following from the nature of bonding in such materials. The sinter temperature is usually at *T* > 0.5*Tm* [K]. Hence, the high temperature ensures that the atoms have sufficient mobility to diffuse to the pores and densify ceramic powders. On the other hand, abnormal grains grow due to Ostwald ripening (German, 1996) when traditional sintering techniques are used, can also occur. Temperature, however, is not the only factor that affects the densification process. A high "*green density*", uniform packing, small particle size, spherical particles and narrow size distribution, for narrow-pore size distribution (Yeh & Sacks, 1990) are also important. In the formation of polycrystalline matrix composites, the presence of second-phase inclusions (e.g. particles) leads to a drastic reduction in the matrix densification rate. Thus considerable difficulties are often encountered in the formation of polycrystalline matrix ceramic composites by conventional, pressure less sintering.

Recent developments in bioceramic composites produced a new set of challenges for sintering theory, particularly when differential densification occurs due to constrained sinterization, i.e parts of a structure densify at different rates and temperatures than other parts (Green et al., 2008) because matrix densifies around rigid inclusions. Inclusions and heterogeneities not only reduce the rate of densification, but also cause differential sintering which generates defects that reduce the strength of the final composite due to differential shrinkage. It was shown that the large inclusions can effectively retard the matrix densification and affect the composite microstructure characteristics (Sudre & Lange, 1997).

Neck

Pore

Particle

Fig. 1. (a) Sintering model geometry of sphere particles. (b) SEM microstructure of a HA

(a) (b)

**2** μ**m**

During sintering, material moves through mechanisms of viscous flow or diffusion (or both) in order to eliminate porosity and reduce the γsv. Diffusion is among the most important phenomena observed in sintering of ceramic materials. Diffusive mass transport takes place when there is a gradient in the chemical potential and when the species in question has sufficient mobility. There are a number of competing paths for mass transport during ceramic sintering (Garay, 2010), such as grain boundary diffusion, volume diffusion and surface diffusion. Usually surface diffusion, leads to coarsening, which is the growth of the neck between particles leading to the reduction of the specific surface area. The other transport mechanisms, normally volume diffusion and grain boundary diffusion lead to

Most of ceramic particles, including oxides, need elevated temperatures to be sintered to high levels of densification. Densification of covalent-ionic ceramics without additives is extremely difficult due to the low diffusivity in the solid state following from the nature of bonding in such materials. The sinter temperature is usually at *T* > 0.5*Tm* [K]. Hence, the high temperature ensures that the atoms have sufficient mobility to diffuse to the pores and densify ceramic powders. On the other hand, abnormal grains grow due to Ostwald ripening (German, 1996) when traditional sintering techniques are used, can also occur. Temperature, however, is not the only factor that affects the densification process. A high "*green density*", uniform packing, small particle size, spherical particles and narrow size distribution, for narrow-pore size distribution (Yeh & Sacks, 1990) are also important. In the formation of polycrystalline matrix composites, the presence of second-phase inclusions (e.g. particles) leads to a drastic reduction in the matrix densification rate. Thus considerable difficulties are often encountered in the formation of polycrystalline matrix ceramic

Recent developments in bioceramic composites produced a new set of challenges for sintering theory, particularly when differential densification occurs due to constrained sinterization, i.e parts of a structure densify at different rates and temperatures than other parts (Green et al., 2008) because matrix densifies around rigid inclusions. Inclusions and heterogeneities not only reduce the rate of densification, but also cause differential sintering which generates defects that reduce the strength of the final composite due to differential shrinkage. It was shown that the large inclusions can effectively retard the matrix densification and affect the composite microstructure characteristics (Sudre & Lange, 1997).

compact, sintered at 1100 ºC for 1 h.

composites by conventional, pressure less sintering.

densification.

Also, differences in rates of sinterization between inclusions and matrix may develop stresses which cause sintering defects such as cracks and isolated pores. In fact, the matrix sintering is inhibited when it tries to densify and contract around the inclusions. In this situation, the mean stress will be compressive on the inclusion and tensile on the matrix, thereby opposing densification. The internal stress field will also depend on the shape of the second phase and its volume fraction. It was also reported that the densification rate decreases with increasing volume fraction of inclusions and that densification apparently stops before the theoretical density is reached. It was shown (Brinker & Scherer, 1990) that when inclusions inhibit densification, coarsening mechanisms such as surface diffusion are favored over sintering. Although several improvements in the mechanical properties (e.g. toughness) of ceramic matrices can be achieved by introducing certain reinforcing particles, they tend to inhibit shrinkage so that the full density of the compact may never be achieved, unless a pressurized process, such as hot-pressing or other approaches is applied.

The success in powder consolidation is intimately related to the control of the competition between densification and coarsening. The growth of larger particles, as shown in Fig.2, and shrinkage of small ones (coarsening) is observed often during sintering of crystalline materials. Control of grain growth is thus an essential aspect of controlling the densificationcoarsening competition, and producing ceramics with desirable microstructures.

The volume fraction of a particulate second phase present in a composite is an important factor in sintering and resultant shrinkage, and the shape and size of the reinforcing phase may also influence the sintering behavior of the matrix material. The effect of heterogeneities, especially, rigid inclusions, on the matrix densification has been a subject of several studies (Bordia & Scherer, 1988; Scherer, 1987). It was shown that at the first stage of densification, the volume fraction, shape, and distribution of the inclusions affects the morphology of the network formed by the differential densification of the matrix. The matrix properties determine whether, in later stages of densification, the interstitial porous regions can reach full density or not. It was concluded that grain growth in the dense regions of the matrix limits the final densification of the composite.

Many properties of composites are predicted on the basis of the rule of mixtures, which is a method based on an assumption that a composite property is the volume weighed average of the phases (matrix and dispersed phase) properties.

Fig. 2. SEM microstructure of a HA compact after sintering at 1300 ºC for 2 h.

New Challenges in the Sintering of HA/ZrO2 Composites 185

internal stress caused by the interface effect, especially the interaction between internal stress/interface and cracks. The reason for the improvement of fracture toughness by this mechanism is the growth of surface area of crack and the change of the stress field

In general, the mechanical properties decrease significantly with increasing content of porosity and grain size, while a high crystallinity, a low porosity and small grain size tend to give a higher stiffness, a higher compressive and tensile strength and a greater fracture toughness. Recently it were obtained HA/YSZ composites with improved mechanical properties, flexural strength of ~155 MPa and fracture toughness of ∼2*.*1 MP m1*/*2, due to the contribution of 25 wt% YSZ component (Sung et al., 2007). The increased fracture toughness would result from the stress-induced tetragonal to monoclinic phase transformation in the YSZ component of HA/YSZ composites. In spite of a better mechanical behaviour, HA/YSZ composites must display uniform microstructures with a high degree of dispersion and without decomposition of the HA, during the sintering process. There are, however, some problems concerning the ceramic processing of HA/ZrO2 composite that should be solved. These problems are related to the influence synthesis and sintering conditions on phase stability of HA and *t*-ZrO2. The conventional mechanical mixing of HA and YSZ powders has been reported for the preparation of HA/YSZ composites through sintering (H.W. Kim et al., 2005; Kumar et al., 2005; Rapacz-Kmita et al., 2006). However, this mechanical mixing would cause low sintering density and/or non-uniform YSZ phase distribution in the sintered HA matrix due to the large particle size and/or segregation of YSZ particles, as

Non-uniform distribution of the YSZ phase in the HA matrix, as shown in Fig.3, would seriously deteriorate the mechanical properties of HA/YSZ composites. Rather, the addition of ZrO2 particles reduces the size of HA grains, ZrO2 can thus act as an effective grain growth inhibitor to the HA grains. However, such composites cannot be sintered at temperatures same as the sintering of pure HA, while the sintering temperature for pure HA does not exceed the 1000 ºC, the specimens made of HA–20 vol.% YSZ must be necessarily heated up to 1100 ºC to obtain near full dense compacts (Khalil et al., 2007). Furthermore, the higher sintering temperature of HA/ZrO2 composite, frequently, causes an extreme grain coarsening, also illustrated in Fig.2, and HA phase decomposition, leading to a considerable deterioration of mechanical properties. Also, at high temperature the calcium, which is the main component of the HA, diffuses into YSZ and changes its tetragonal structure into the stable cubic phase, hindering the transformation toughening mechanism (Wu & Yeh, 1988). Furthermore, calcium diffusion also favours the decomposition of HA, leading to the formation of β-tricalcium phosphate (β-TCP, Ca3(PO4)2) and calcium zirconate (CZ, CaZrO3). The presence of such phases in the composite drastically reduces it mechanical behaviour (LeGeros, 1993). The formation of other phases among HA also interferes with the sinterability of the composites due to differential shrinkage, leading to overall reduced density as compared to that of pure HA. For instance, the sinterability of a HA–20wt.%ZrO2 (PSZ; 3 mol% Y2O3-doped) composite was affected by the differential shrinkage due to HA thermal decomposition (Wu & Yeh, 1988). Therefore, if there is no reaction between HA and ZrO2, the strength and toughness of HA are indeed improved significantly. HA and ZrO2 readily react with each other above 1000 ºC (Shen et al., 2001). However, reducing the temperature, the densification is

distribution due to the crack propagating by deflection.

shown in both micrographs of Fig.3.

Description of sintering of particle-reinforced composites, for a large extent of densification and if no interfacial reaction occurs between the matrix and reinforcement, can be made in terms of a simple rule of mixtures approach (Edrees et al., 1998), as shown by equation (1):

$$
\left(\frac{\Delta V}{V\_0}\right)\_\mathcal{C} = \left(\frac{\Delta V}{V\_0}\right)\_m (1 - p) \tag{1}
$$

where the subscripts *c* and *m* refer to composite and matrix, respectively, and where *p* is the volume fraction of the filler; therefore, the shrinkage of the composite after sintering is equal to the shrinkage that would be found in the isolated matrix times the matrix volume fraction of the composite.
