A. Morales-Rodrfgnez er al. /Journal of the European Ceramic Society 27(2007)3301-3305 3303 Microstructural observations have been devoted to charac e the fiber fracture surfaces, pull-out phenomena and matrix cracking to account for the composite mechanical response. The main results obtained are summarized as follows, (i)The 72% and 82% of fibers in RTT specimens exhibit mirror surfaces after static fatigue at 70% and 90% or respectively, whereas about the half of fibers in MTT deformed composites presents mirror surfaces(56%). The previous result suggests that deformation under constant load enhances SCG processes in fibers. The mirror sur faces are well-defined and, in general, the original defect in these Nicalon fibers is associated to the fiber surface. but is too small to be observed(Fig 3). Measurements of mir- ror surfaces radii indicate that similar critical crack sizes are observed in both testing conditions (ii)The pull-out length is shorter after static fatigue RTT=(34+2)um, than for original specimens, Fig. 4. Pulled-out fibers emerging from the fracture plane in specimen #4. MTT=(58+4) um. Interfacial debonding between fibers and matrix has been systematically observed for pulled-out fibers. Fig. 4 shows a typical micrograph of pulled-out fibers from a RTT-deformed composite (iii) The distribution of the intercracking distance of the Sic matrix has been studied to compare multicracking damage of composites. Fig. 5 shows that the mean matrix intercracking distances decreases with increasing final stress levels acting on the specimens. Data from two HA日 additional tests performed to compare the influence of monotonic and static loading on the matrix damage are also included in Fig. 5(specimens were loaded below to final fracture stress values until 80% R in case of MTT(#7)and r during 90h for the static f the intercracking distribution was studied for these non- fractured specimens). No differences with global behavior have been deduced from these both loading conditions induce similar matrix damage char- d (um) correlation between maximum final stress applied and matrix damage: the higher istics where the key role is played by the final stress 4. discus 4.1. In situ fiber strength and fiber/ matrix interfacial shear stress of 2 D-SiCySiC composites from the measured mirror radii using the following empirical Fig 3. Distinct mirror surface composites where it is where Im is the mirror radius, or the in situ fiber tensile strength ot possible to identify the initial defect. Note the interfacial debonding between and Am is the mirror constant. The value of Am=3.5MPam-I fiber and matrix has been considered in this study following the work of ThoulessA. Morales-Rodr´ıguez et al. / Journal of the European Ceramic Society 27 (2007) 3301–3305 3303 Microstructural observations have been devoted to character￾ize the fiber fracture surfaces, pull-out phenomena and matrix cracking to account for the composite mechanical response. The main results obtained are summarized as follows: (i) The 72% and 82% of fibers in RTT specimens exhibit mirror surfaces after static fatigue at 70% and 90% σR, respectively, whereas about the half of fibers in MTT￾deformed composites presents mirror surfaces (56%). The previous result suggests that deformation under constant load enhances SCG processes in fibers. The mirror sur￾faces are well-defined and, in general, the original defect in these Nicalon fibers is associated to the fiber surface, but is too small to be observed (Fig. 3). Measurements of mir￾ror surfaces radii indicate that similar critical crack sizes are observed in both testing conditions. (ii) The pull-out length is shorter after static fatigue, lRTT = (34 ± 2)m, than for original specimens, lMTT = (58 ± 4)m. Interfacial debonding between fibers and matrix has been systematically observed for pulled-out fibers. Fig. 4 shows a typical micrograph of pulled-out fibers from a RTT-deformed composite. (iii) The distribution of the intercracking distance of the SiC matrix has been studied to compare multicracking damage of composites. Fig. 5 shows that the mean matrix￾intercracking distances decreases with increasing final stress levels acting on the specimens. Data from two additional tests performed to compare the influence of monotonic and static loading on the matrix damage are also included in Fig. 5 (specimens were loaded below to final fracture stress values until 80% σR in case of MTT (#7) and 70% σR during 90 h for the static fatigue test (#8) and then, the intercracking distribution was studied for these non￾fractured specimens). No differences with global behavior have been deduced from these experiences meaning that both loading conditions induce similar matrix damage char￾Fig. 3. Distinct mirror surface developed in 2D-SiCf/SiC composites where it is not possible to identify the initial defect. Note the interfacial debonding between fiber and matrix. Fig. 4. Pulled-out fibers emerging from the fracture plane in specimen #4. Fig. 5. Matrix-intercracking distance (d) vs. final stress level plot showing a correlation between maximum final stress applied and matrix damage: the higher stress, the shorter intercracking distance. acteristics where the key role is played by the final stress level carried. 4. Discussion 4.1. In situ fiber strength and fiber/matrix interfacial shear stress of 2D-SiCf/SiC composites In situ fiber strength in the composites was evaluated from the measured mirror radii using the following empirical relationship7: σfr1/2 m = Am, (1) where rm is the mirror radius, σf the in situ fiber tensile strength and Am is the mirror constant. The value of Am = 3.5 MPa m−1/2 has been considered in this study following the work of Thouless