pril 2001 Toughened Oxide Composites Based on Porous Alumina-Platelet Interphases area under the load-displacement curve by the cross-sectional area along the porous region at the center of the interphase is observed of the sample. The relative WoF values can be compared,(see Fig. 2(b)) because the specimens had essentially the same dimensions. The exceptions were some specimens that had different dimensions, as indicated at the bottom of the respective tables in this paper, which Fracture Behavior of the Mullite Laminates, Relative to the Mullite Content in the Interphase arize the mechanical-property data (B) Microstructure Characterization: The microstructure of The results of flexural testing for the 3Al,O3 2SiO2-matrix laminates, which have a variable 3Al2O3 2SiO, content in the the platelet powders, the surfaces of the sintered samples, and their A1,O3-platelet interphases, are listed in Table I. The strength side views and fracture surfaces after bend testing were observed increased as the 3A1,0,. 2SiO, content in the interphase increased sing scanning electron microscopy(SEM)(Model DS-130, Inter this result was attributed to a denser, albeit still porous, micro national Scientific Instruments, Santa Clara, CA) structure that formed during sintering. Load-displacement curve for the 3Al,O3 2SiO, laminates, as a function of the 3Al2O3 2SiO2 content in the interphase, are shown in Fig 3. The bend bars with I. Results 3Al20, 2SiO2 contents of 5, 10, and 20 vol% produced fracture (I Microstructure of the Platelets and the interphase curves with only a few steps before failure(see Fig. 3(a)). The The typical morphology of the Al2O3 platelets was hexagonal stepwise load drops are characteristic of "graceful failure"and indicate crack deflection and debonding at the interphase. Slightly crystals -1 um thick with diameters in relatively narrow size- more steps were observed in the laminates that had a distribution ranges: 3-5, 5-10, 10-15, and 20-25 Hm. In this 3Al20, 2SiO2 content of <5 vol% in the interphase(see Fig 3(b)) study, the 5-10 um and 10-15 um size ranges were determined to The laminate that had no 3A1,0 2SiO2 had the most steps in its be the most useful. The surfaces of the rree oe were sometime load-displacement curve, in contrast to the laminates that had a pited in the middle but were relatively free of pits on the outer 3A1,0, 2SiO, content of >5 vol%(see Fig 3(a)). However,the surfaces. Figure 2 shows representative micrographs of a lami- of the laminate that did not contain any 3Al,O3 2SiO2 nated 3A12O3 2SiO2 matrix with an Al,O3-platelet interphase. The platelets are randomly oriented and clearly exhibit a nonsintere in the interphase was rather low. The WoF at room rature was less for 3Al2O3 2SiO2 additions of >5 vol% region between the 3A10, 2Si0, matrix and the Al,. -platelet (despite relate e ye thic ker interphases compared to the mat phase is relatively dense. After fracture testing, crack deflection interphase)than in laminates that had I and 3 vol%of 3Al2O3, 2SiO2 in the interphase. This observation implies that, at room temperature, interphases with a 3Al2O32SiO2 content of >5 (a) vol% were not weak enough for crack deflection to occur. Micrographs of the tensile-side(failed-side) view of the lami- nates with 3Al,O3 2SiO, contents of 5 and 3 vol% are shown in Figs. 4(a)and (b), respectively. In the 3-vol%0-3Al2O3 2SiO interphase specimen, the crack deflection noticeably proceeded along the weak Al,O3-platelet interphase, despite the interphase Interphase layers being relatively thin(-80 um). This observation implies that the crack deflection is strongly dependent on the microstruc- ure of the porous interphase, rather than its thickness ( Fracture Behavior of the Laminates, Relative to the Matrix: Interphase Thickness Ratio aminates with 3A1,O3, or Al,O, matrixes in different matrix interphase thickness ratios are presented in Tables ll and Ill. The strength and Matrix WoF each increased as the matrix thickness increased In partic ular. the laminates that had a"bimodal" design showed noticeabl 899225K58um higher strength and wOF. In 3Al,O3, 2SiO -matrix laminates, the sample with a 4: I thickness ratio had a higher strength and WoF than did that of the laminate that had a 6: I thickness ratio, despite (b) a thicker interphase. This result is attributed to the effect of the interphase. The effects of thickness ratio and 3Al2O3'2SiO Crack content were optimized in the bimodal 3Al,O, ', laminate that had alternating matrix layers, with matrix interphase ratios of 3 Table I. Variation in Strength and work of fracture for Mullite -Matrix Laminates, According to the Mullite Content Interphase in the Alumina-Platelet(10-15 um)Interphases Thickness Flexural strength Work of fracture, WOF Matrix 889625KV588u 246222 Fig. 2. SEM micrographs of the ite with alumina-platelet- TDensified matrix interphase thickness ratio, ' For specimens 30 mm(length)x4.0 weakened interphases ((a) cross section and (b) crack profile)area under the load–displacement curve by the cross-sectional area of the sample.43 The relative WOF values can be compared, because the specimens had essentially the same dimensions. The exceptions were some specimens that had different dimensions, as indicated at the bottom of the respective tables in this paper, which summarize the mechanical-property data. (B) Microstructure Characterization: The microstructure of the platelet powders, the surfaces of the sintered samples, and their side views and fracture surfaces after bend testing were observed using scanning electron microscopy (SEM) (Model DS-130, Inter￾national Scientific Instruments, Santa Clara, CA). III. Results (1) Microstructure of the Platelets and the Interphase The typical morphology of the Al2O3 platelets was hexagonal crystals ;1 mm thick with diameters in relatively narrow size￾distribution ranges: 3–5, 5–10, 10–15, and 20–25 mm. In this study, the 5–10 mm and 10–15 mm size ranges were determined to be the most useful. The surfaces of the platelets were sometimes pitted in the middle but were relatively free of pits on the outer surfaces. Figure 2 shows representative micrographs of a lami￾nated 3Al2O3z2SiO2 matrix with an Al2O3-platelet interphase. The platelets are randomly oriented and clearly exhibit a nonsintered, porous microstructure (see Fig. 2(a)). However, the interfacial region between the 3Al2O3z2SiO2 matrix and the Al2O3-platelet phase is relatively dense. After fracture testing, crack deflection along the porous region at the center of the interphase is observed (see Fig. 2(b)). (2) Fracture Behavior of the Mullite Laminates, Relative to the Mullite Content in the Interphase The results of flexural testing for the 3Al2O3z2SiO2-matrix laminates, which have a variable 3Al2O3z2SiO2 content in the Al2O3-platelet interphases, are listed in Table I. The strength increased as the 3Al2O3z2SiO2 content in the interphase increased; this result was attributed to a denser, albeit still porous, micro￾structure that formed during sintering. Load–displacement curves for the 3Al2O3z2SiO2 laminates, as a function of the 3Al2O3z2SiO2 content in the interphase, are shown in Fig. 3. The bend bars with 3Al2O3z2SiO2 contents of 5, 10, and 20 vol% produced fracture curves with only a few steps before failure (see Fig. 3(a)). The stepwise load drops are characteristic of “graceful failure” and indicate crack deflection and debonding at the interphase. Slightly more steps were observed in the laminates that had a 3Al2O3z2SiO2 content of ,5 vol% in the interphase (see Fig. 3(b)). The laminate that had no 3Al2O3z2SiO2 had the most steps in its load–displacement curve, in contrast to the laminates that had a 3Al2O3z2SiO2 content of .5 vol% (see Fig. 3(a)). However, the strength of the laminate that did not contain any 3Al2O3z2SiO2 powder in the interphase was rather low. The WOF at room temperature was less for 3Al2O3z2SiO2 additions of .5 vol% (despite relatively thicker interphases, compared to the matrix thickness, and, hence, more likely crack deflection along the interphase) than in laminates that had 1 and 3 vol% of 3Al2O3z2SiO2 in the interphase. This observation implies that, at room temperature, interphases with a 3Al2O3z2SiO2 content of .5 vol% were not weak enough for crack deflection to occur. Micrographs of the tensile-side (failed-side) view of the lami￾nates with 3Al2O3z2SiO2 contents of 5 and 3 vol% are shown in Figs. 4(a) and (b), respectively. In the 3-vol%-3Al2O3z2SiO2 interphase specimen, the crack deflection noticeably proceeded along the weak Al2O3-platelet interphase, despite the interphase layers being relatively thin (;80 mm). This observation implies that the crack deflection is strongly dependent on the microstruc￾ture of the porous interphase, rather than its thickness. (3) Fracture Behavior of the Laminates, Relative to the Matrix:Interphase Thickness Ratio The results of the flexural testing of laminates with 3Al2O3z2SiO2 or Al2O3 matrixes in different matrix:interphase thickness ratios are presented in Tables II and III. The strength and WOF each increased as the matrix thickness increased. In partic￾ular, the laminates that had a “bimodal” design showed noticeably higher strength and WOF. In 3Al2O3z2SiO2-matrix laminates, the sample with a 4:1 thickness ratio had a higher strength and WOF than did that of the laminate that had a 6:1 thickness ratio, despite a thicker interphase. This result is attributed to the effect of the addition of 1 vol% of 3Al2O3z2SiO2 powder to the platelet interphase. The effects of thickness ratio and 3Al2O3z2SiO2 content were optimized in the bimodal 3Al2O3z2SiO2 laminate that had alternating matrix layers, with matrix:interphase ratios of 3:1 Fig. 2. SEM micrographs of the composite with alumina-platelet￾weakened interphases ((a) cross section and (b) crack profile). Table I. Variation in Strength and Work of Fracture for Mullite-Matrix Laminates, According to the Mullite Content in the Alumina-Platelet (10–15 mm) Interphases Mullite content (vol%) Thickness ratio† Flexural strength (MPa) Work of fracture, WOF‡ (kJ/m2 ) 0 2:1 70 0.4 1 4:1 77 0.6 3 6:1 88 0.6 5 2:1 86 0.4 10 2:1 97 0.4 20 2:1 128 0.4 † Densified matrix:interphase thickness ratio. ‡ For specimens 30 mm (length) 3 4.0 mm (thickness) 3 3.0 mm (width). April 2001 Toughened Oxide Composites Based on Porous Alumina-Platelet Interphases 769