Cyclic fatigue crack growth behavior in four kinds of aluminas, monolithic silicon carbide and TiB2 particulates reinforced silicon carbide composite ceramic is investigated, paying special attention to crack closure, and the effect of grain size in alumina ceramics is discussed. Cyclic fatigue tests were carried out by using four-point bend specimens and the load-strain and load-differential strain curves were monitored. These curves show non-linearities due to crack closure and hysteresis loop, probably due to frictional sliding of bridging grains. The crack opening load is determined from the load- differential strain curve by using a method introduced in this study. The newly defined crack opening ratio, U*, is nearly constant with increase of the maximum stress intensity factor, Kmax, at a given stress ratio, R, and can be expressed as a function of stress ratio, R. Crack in these ceramics shows a susceptibility to cyclic fatigue loading. Fracture surfaces and morphology of crack path in these ceramics were observed by SEM. It is shown that intergranular fracture mode is dominant for the cyclic fatigue crack and grains at the crack interface are bridged with the surrounding matrix. Thus, fatigue crack growth for monolithic silicon carbide and silicon carbide composite ceramics, as well as for aluminas, can be explained by grain bridging mechanism Growth rates can be successfully described by the relationship $ da/dN = C [ΔK_eff /(1-K_max/K_IC)]^m $ proposed in this study to account for the effects of crack closure and maximum stress intensity factor. The effect of grain size on growth rates in alumina ceramics can be well explained by the above relationship. The growth rate curve based on the proposed relationship shows a sigmoidal form for ceramic materials ,like metals.
Back-face elastic strain(BFS) compliance in four-point bend(FPB) specimen was calibrated by finite element analysis and experiment. The normalized BFS compliance values, E'BCW, can be well expressed with a 6th-order polynomial function of crack ratio, a/W. The back-face strain analysis shows that the gage length less than 0.2W is useful for measurement of back face strain compliance of the FPB specimen.
