J. She et al. Materials Science and Engineering 4325(2002)19-24 a) (b) Oum Fig. 6. Fracture topography of (a) the surface region and(b) the interior region for a porous mullite/mullite composite after six infiltration cycles of infiltration cycles. This indicates an increased elastic specimens(the presence of an unstable crack propaga- modulus of the composites after multiple infiltrations tion during tests is not a result of multiple crack due to the introduction of alumina, which has a higher initiation but a feature of fiber-reinforced ceramic-ma Youngs modulus than mullite trix composites with weak fiber/matrix interfaces) In addition, the flexural strength of the infiltrated However it must be remembered that the measured composites was calculated from the maximum load in the load-displacement response. The results are sum- 300 marized in Fig. 7. Regardless of the fact that the matrix density increases with the cycles of infiltration, all the specimens have almost an identical strength of 190 E 240 MPa. This suggests that the fracture stress of the composites is governed mainly by the fibers Fig. 8 shows the fracture energy of porous mullite/ mullite composites after multiple infiltrations. Consid- ering the fact that the mechanical behavior of the infiltrated composites changes gradually from the sur- fracture energy was estimated directly from strength A 60 face to the interior due to a varying microstructure, the measurements without the introduction of a notch or a precrack on the tensile surface of the specimen. In this 0 case, multiple crack initiation may occur and thus 6 influence the determination of fracture energy. Interest Number of Infiltration Cycles ingly, such a crack initiation was not observed, as evidenced in Fig 4 where the stress-displacement curve Fig. 7. Flexural strength versus the number of infiltration cycles for remains linear up to the maximum stress for all theJ. She et al. / Materials Science and Engineering A325 (2002) 19–24 23 Fig. 6. Fracture topography of (a) the surface region and (b) the interior region for a porous mullite/mullite composite after six infiltration cycles. of infiltration cycles. This indicates an increased elastic modulus of the composites after multiple infiltrations due to the introduction of alumina, which has a higher Young’s modulus than mullite. In addition, the flexural strength of the infiltrated composites was calculated from the maximum load in the load–displacement response. The results are sum￾marized in Fig. 7. Regardless of the fact that the matrix density increases with the cycles of infiltration, all the specimens have almost an identical strength of
190 MPa. This suggests that the fracture stress of the composites is governed mainly by the fibers. Fig. 8 shows the fracture energy of porous mullite/ mullite composites after multiple infiltrations. Consid￾ering the fact that the mechanical behavior of the infiltrated composites changes gradually from the sur￾face to the interior due to a varying microstructure, the fracture energy was estimated directly from strength measurements without the introduction of a notch or a precrack on the tensile surface of the specimen. In this case, multiple crack initiation may occur and thus influence the determination of fracture energy. Interest￾ingly, such a crack initiation was not observed, as evidenced in Fig. 4 where the stress–displacement curve remains linear up to the maximum stress for all the specimens (the presence of an unstable crack propaga￾tion during tests is not a result of multiple crack initiation but a feature of fiber-reinforced ceramic-ma￾trix composites with weak fiber/matrix interfaces). However, it must be remembered that the measured Fig. 7. Flexural strength versus the number of infiltration cycles for porous mullite/mullite composites