M.B. Ruggles-Wrenn et al. Composites Science and Technology 68(2008)1588-1595 1593 s5 mm 5 mm 5 mm Fig 8. Fracture surfaces obtained in creep tests conducted at 45 MPa at 1200C in:(a)air,(b) steam and (c) argon 50 um 100 um 100um Fig 9. SEM micrographs of the fracture surfaces obtained in creep tests conducted at 45 MPa at 1200C in:(a) air, (b)steam and (c)argon. occurs primarily through the matrix, with only minimal typical for dense-matrix CMCs. As seen in the SEM micro- fiber fracture. a diffuse localized deformation band is seen graphs(Fig. 11), the amount of matrix material remaining in all specimens(Fig 8). Following localization, substantial bonded to the fibers is greater than that in the specimens additional straining occurs within the localized band prior subjected to creep at 45 MPa. The difference is particularly to specimen failure Deformation within the localized band pronounced in the case of specimens tested in steam is accommodated by the extensive fragmentation of the Recent studies [35-37] investigated effects of thermal matrix and rotation of the fiber tows towards the loading aging on the physical and mechanical properties of compos- direction. Damage within this band is highly dilatational, ites consisting of Nextel720 fibers and a porous-matrix of as evidenced by the pronounced through-thickness swell- mullite and alumina. For a composite with a pure alumina ing, which is attributed to the comminution of the matrix matrix, a porosity reduction of 6% was observed after a during deformation [25, 18]. Note that the SEM micro- 10 min exposure at 1200C [36, 37]. For a composite with a graphs(Fig. 9)reveal only small amounts of matrix parti- mullite/alumina matrix, strengthening of the matrix and cles bonded to the fiber surfaces. The test environment the fiber-matrix interfaces was observed following aging at appears to have little effect on the failure mechanism 1200C [35]. Additional sintering of the matrix during the and/or fracture surface appearance aging treatments was considered to be associated predomi- By contrast, the fracture surfaces obtained for specimens nantly with Al2O3. The short duration of the creep tests at subjected to prior creep at 35 MPa(Figs. 10 and ll)reveal 45 MPa does not allow for any substantial strengthening that the failure mechanism in this case includes extensive of the matrix to occur. However, it is likely that additional fiber fracture. The fracture surface of the specimen pre- sintering of the matrix occurred during the 100 h creep tests crept in air(Fig. 10a)shows areas of fiber failure, while at 35 MPa. The resultant strengthening of the matrix is man- retaining some of the V-shape fracture typically seen in ifested in the retained properties of the composite. Results in the porous-matrix composites with +45 orientation. Spec- Table 2 show that after 100 h of prior creep at 35 MPa, the imens subjected to prior creep in steam and in argon frac- modulus and the tensile strength increase, and the inelastic tured along the planes nearly orthogonal to the loading straining capability of the composite decreases. The direction. Fracture surfaces in Fig. 10b and c suggest that strengthening is also manifested in the change in the failure specimens pre-crept in steam and in argon failed cata- mechanism. The failure of the composite subjected to prior strophically at the maximum load, with the majority of creep at 35 MPa in steam and in argon is dominated by fiber the fibers breaking in the process. Such failure process is fracture, while the as-processed material fails predominantlyoccurs primarily through the matrix, with only minimal fiber fracture. A diffuse localized deformation band is seen in all specimens (Fig. 8). Following localization, substantial additional straining occurs within the localized band prior to specimen failure. Deformation within the localized band is accommodated by the extensive fragmentation of the matrix and rotation of the fiber tows towards the loading direction. Damage within this band is highly dilatational, as evidenced by the pronounced through-thickness swell￾ing, which is attributed to the comminution of the matrix during deformation [25,18]. Note that the SEM micro￾graphs (Fig. 9) reveal only small amounts of matrix parti￾cles bonded to the fiber surfaces. The test environment appears to have little effect on the failure mechanism and/or fracture surface appearance. By contrast, the fracture surfaces obtained for specimens subjected to prior creep at 35 MPa (Figs. 10 and 11) reveal that the failure mechanism in this case includes extensive fiber fracture. The fracture surface of the specimen pre￾crept in air (Fig. 10a) shows areas of fiber failure, while retaining some of the V-shape fracture typically seen in the porous-matrix composites with ±45 orientation. Spec￾imens subjected to prior creep in steam and in argon frac￾tured along the planes nearly orthogonal to the loading direction. Fracture surfaces in Fig. 10b and c suggest that specimens pre-crept in steam and in argon failed cata￾strophically at the maximum load, with the majority of the fibers breaking in the process. Such failure process is typical for dense-matrix CMCs. As seen in the SEM micro￾graphs (Fig. 11), the amount of matrix material remaining bonded to the fibers is greater than that in the specimens subjected to creep at 45 MPa. The difference is particularly pronounced in the case of specimens tested in steam. Recent studies [35–37] investigated effects of thermal aging on the physical and mechanical properties of compos￾ites consisting of NextelTM720 fibers and a porous-matrix of mullite and alumina. For a composite with a pure alumina matrix, a porosity reduction of 6% was observed after a 10 min exposure at 1200 C [36,37]. For a composite with a mullite/alumina matrix, strengthening of the matrix and the fiber-matrix interfaces was observed following aging at 1200 C [35]. Additional sintering of the matrix during the aging treatments was considered to be associated predomi￾nantly with Al2O3. The short duration of the creep tests at 45 MPa does not allow for any substantial strengthening of the matrix to occur. However, it is likely that additional sintering of the matrix occurred during the 100 h creep tests at 35 MPa. The resultant strengthening of the matrix is man￾ifested in the retained properties of the composite. Results in Table 2 show that after 100 h of prior creep at 35 MPa, the modulus and the tensile strength increase, and the inelastic straining capability of the composite decreases. The strengthening is also manifested in the change in the failure mechanism. The failure of the composite subjected to prior creep at 35 MPa in steam and in argon is dominated by fiber fracture, while the as-processed material fails predominantly Fig. 8. Fracture surfaces obtained in creep tests conducted at 45 MPa at 1200 C in: (a) air, (b) steam and (c) argon. Fig. 9. SEM micrographs of the fracture surfaces obtained in creep tests conducted at 45 MPa at 1200 C in: (a) air, (b) steam and (c) argon. M.B. Ruggles-Wrenn et al. / Composites Science and Technology 68 (2008) 1588–1595 1593