image credits: shutterstock.com/g/yuangeng%20zhang
Although there has been an extraordinary amount of work in the area of erosion of brittle materials over the past two decades, there has been little work on direct comparisons of the erosion behaviour of the most commonly used advanced ceramics and the effects of various impact angles and erodent particle properties on the erosion mechanism of ceramic materials.
In this study, the erosion behaviour of three advanced ceramics, namely high purity alumina (AD 998), MgO partially stabilised zirconia (Mg-PSZ A and B) and Ca α-sialon (CA1005, CA2613 and CA 3618), have been investigated under airborne erosion using garnet and silicon carbide grits. The effects of impact angle and erodent particle properties on the erosion rate were determined. The effect of ceramic microstructure on the erosion mechanism was also examined.
The Effect of Impact Angle
In all cases, the erosion rate peaked at normal impact and decreased steadily as the impingement angle became more oblique. The distinctive differences in erosion rates for shallow and normal impact angles can be attributed to the different material removal mechanisms in these two cases. SEM examination of the eroded surfaces provides some insight into the material removal process.
A comparison between the morphological features in completely eroded surfaces of material AD998 subjected to garnet erosion at 30° and 90° impacts, shows two different methods of materials removal. In the case of the shallow angle eroded surface, the dominant material removal mechanism was grain ejection and plastic deformation, while in the case of the high angle eroded surface, the material removal mechanism was mainly grain ejection. Grain ejection is caused by grain boundary microcracking and plastic deformation is believed to be the result of the smearing of the deformed materials. These differences may be explained as follows.
Low Angle Impact
At low angle impact, the kinetic energy of the impinging particles contributes mainly to the ploughing mechanism and very little to normal repeated impact. The ploughing mechanism is associated with the plastic smearing and cutting of the materials, while the repeated impact mechanism is responsible for initiating and propagating the grain boundary microcracks. Ceramics have high hardness, and thus they are not easily plastically deformed. Hence the material removal rate is low.
High Angle Impact
At high angle impact, the kinetic energy of the impinging particles contributes mainly to repeated impact. Ceramics have low fracture toughness, and therefore cracks readily propagate to form a crack network. The subsequent impacts will easily remove the surface material via the ejection of the upper layer grains. Hence the material removal rate is high.
Effect of Erodent Particle Properties
The cumulative volume loss due to erosion of all target materials by (a) garnet and (b) SiC erodent at 90° as a function of the amount of erodent particles impinging on the surface was measured. In both cases, the ranking of the material, from the erosion resistance point of view, was essentially the same. α-sialons possessed the best erosion resistance and then the two Mg-PSZ materials. Alumina ceramic showed the lowest erosion resistance. However, two interesting features were revealed:
1. In the case of erosion using garnet erodent, the material removal rate of alumina ceramic was about 13 and 200 times higher than those of the two zirconia and the three sialon materials, respectively. However, such differences were significantly reduced when the SiC erodent was used, where the material removal rate of the alumina ceramic was only 3 and 6 times higher than those of zirconia and sialon ceramics, respectively.
2. In all cases, SiC resulted in significantly higher material removal rate than garnet particles.
The difference in erosion rates between the garnet and SiC erodents can be accounted for by the variations in the properties, such as hardness and fracture toughness, of the erodent materials.
Erosion by Garnet Particles
SEM examination of eroded surfaces indicated that erosion by garnet erodent was relatively smooth and contained few isolated pits. Higher magnification SEM study of the plastically deformed regions, revealed that it contained fine debris, deformed materials and locally melted materials. EDX analysis of these deformed and melted materials showed high levels of Fe, indicating that they derived or partially derived from garnet erodent which contained FeO. Higher magnification images of the damaged surface inside the pits, revealed clear grain facets, indicating that grain ejection is the main material removal mechanism.
Erosion by Silicon Carbide
On the other hand, the damaged surface due to erosion by SiC erodent appeared to be much rougher than that eroded by garnet particles. A higher magnification view of the damaged surface showed that grain ejection is the main material removal mechanism.
It is clear that when softer garnet particles impinge on the harder ceramic target surfaces, they undergo fragmentation, deformation as well as local melting. At the same time, the grain boundary microcracks of the underlying ceramic grains are also developed. Further impacts will eventually result in material removal by spalling of the smeared materials. However, when harder SiC particles impinge on the relatively softer ceramic target surfaces, they undergo fragmentation and also produce substantial grain boundary microcracking on the target surface. If the target ceramic exhibits weak grain boundary strength or fine equiaxed grains, the subsequent impacts will result in a high material removal rate via grain ejection mechanism.
Effect of Ceramic Microstructure
Of the three α-sialon ceramics eroded by SiC particles, it can be seen that, material CA2613 exhibits the highest erosion resistance while CA1005 shows the lowest erosion resistance. Given the mechanical properties, such as hardness and toughness, of the three sialon materials shown in Table 1, the difference in erosion rates among the sialon ceramics can not be simply accounted for by their mechanical properties. Therefore, such a difference must be due to the variations in their microstructures.
Table 1. Properties of the target ceramics.
||Ca α -Sialon
SEM examination of the CA2613 and CA3618 sialon materials eroded using SiC at 90°shows that plastic smearing and transgranular fracture are the main erosion mechanisms. This is, however, quite different from the case of CA1005 subjected to SiC erosion, where the main erosion mechanism is grain dislodgment. The differences in erosion rates, as well as the morphological features for the three sialon compositions, can be explained in the following scenario.
As stated earlier, when ceramics are subjected to SiC erosion, substantial grain boundary microcracks are developed on the target surface. Subsequently, the brittle nature of the ceramics allows the cracks readily to propagate to form crack networks. If the target contains equiaxed grains, the crack network readily results in the dislodgment of the surface grains and hence the material removal rate is high. However, the grain dislodgment is hindered if the target contains randomly orientated elongated grains. This is because the interlock effect of the elongated grains results in transgranular fracture of the surface grains and hence the material removal rate is relatively low.
The erosion behaviour of several advanced ceramics has been examined. The effects of impact angle and erodent particle properties on the material removal mechanism of these materials have also been investigated. Three important conclusions can be drawn from the findings of this study:
1. At low angles of impact, ceramics exhibit excellent erosion resistance.
2. Garnet erodent produces less erosion of ceramic materials than SiC.
3. Ceramics with elongated grains coupled with strong grain boundary strength exhibit better erosion resistance, especially at high angle impact.
Note: More details on the experimental procedure and a complete list of references are provided in the original text.
Primary author: Y Zhang, Y.B. Cheng and S. Lathabai
Source: Abstracted from The Journal of the Australasian Ceramics Society, Vol. 37, No. 1, pp. 39-44 2001.
For more information on this source please visit The Australasian Ceramic Society.