ICF10B

EXPERIMENTAL PROCEDURE All the matrices of the three test composites were reinforced by ceramic-grade NicalonÔ fibers with a fiber volume fraction of about 0.39. The nominal fiber diameters ranged from 10 to 15 mm. The three composite materials tested included NicalonÔ unidirectionally (1D) fiber-reinforced calcium aluminosilicate (designated SiC/CAS-II), NicalonÔ cross-plied (2D) magnesium aluminosilicate (designated SiC/MAS-5), and NicalonÔ plain-woven (2D) silicon carbide composites (designated SiC/SiC). Both SiC/CAS-II and SiC/MAS-5 were fabricated by Corning, Inc. through hot-pressing followed by ceraming of the composites by a thermal process. The silicon carbide matrix in the SiC/SiC composites was fabricated by the DuPont Company through chemical vapor infiltration (CVI) into the fiber perform. SiC/CAS-II and SiC/MAS-5 laminates were 18 and 16 plies thick, respectively, with a nominal thickness of about 3 mm. The plainwoven laminates of the SiC/SiC composite were supplied 12 plies (normally 3.5 mm thick). More detailed information regarding the test composite materials can be found elsewhere [5]. The SiC/CAS-II material has been used in a previous, preliminary study on test rate-effect on tensile strength [6]. The dogboned tensile test specimens measuring 152.4 mm (length) x 12.7 mm (width) were machined from the composite laminates, with the gage section of about 30 mm long, 10 mm wide and 3.0-3.5 mm thick (as-furnished). The design of the dogboned tensile test specimen was the result of previous finite element analysis [7]. Monotonic tensile testing was conducted in air at 1100oC for both SiC/CAS-II and SiC/MAS-5 and at 1200oC for SiC/SiC, using a servohydraulic test frame (Model 8501, Instron, Canton, MA). A total of three to four different loading rates (in load control), corresponding to stress rates ranging within 50-0.005 MPa/s, were employed with typically 3 test specimens tested at each loading rate. Detailed experimental procedure on tensile testing and related induction-heating equipment can be found elsewhere [5]. Preload or accelerated testing technique, applied primarily to monolithic ceramics and glass [8], was also conducted at test temperatures using 0.5 MPa/s (for SiC/CAS-II) or 0.005 MPa/s (for SiC/MAS-5 and SiC/SiC) in an attempt to better understand the governing failure mechanism of the materials. Predetermined preloads, corresponding to about 80 to 90 % of the failure strength at 0.5 MPa/s or 0.005 MPa/s with zero preload (regular testing), were applied quickly to the test specimens prior to testing and their corresponding strengths were measured. Typically two to three test specimens were used in preload testing. Tensile testing was performed in accordance with an ASTM Test Method, ASTM C 1359 [9]. RESULTS Constant Stress-Rate Testing Results of monotonic tensile strength testing with different test rates are presented in Figure 1, where log (ultimate strength) was plotted as a function of log (applied stress rate) for each composite material. Each solid line in the figure indicates a best-fit regression line based on the log (ultimate strength) versus log (applied stress rate) relation. The decrease in ultimate strength with decreasing stress rate, which represents a susceptibility to damage accumulation or delayed failure, was significant for all the composite materials. The strength degradation was about 51, 31 and 62 %, respectively, for SiC/CAS-II, SiC/MAS-5 and SiC/SiC when stress rate decreased from the highest to the lowest. Fracture patterns for the SiC/CAS-II composite showed some fiber pullout with jagged faceted matrix cracking often propagating along the test-specimen length. For a given stress rate, however, the difference in strength between different fracture patterns was not obvious. No appreciable difference in the mode of failure was observed for SiC/MAS-5 and SiC/SiC, where most specimens tested at either high or low stress rate exhibited relatively flat fracture surfaces – possibly termed brittle fracture. Preload Testing The results of preload tests are also shown in Figure 1, where the ultimate strength with 80 to 90 % preloads is compared with that in regular testing with zero preload. The difference in strength between two preloads (0 and 80-90 %) was negligibly small for each material: 211 MPa (for 0 % preload) and 209 MPa (for a 85 % preload) for SiC/CAS-II; 142 MPa (0 and 80 % preload) for SiC/MAS-5; 77 MPa (0 %) and 80 MPa (90 %) for SiC/SiC. Hence, the maximum strength difference, exhibited by SiC/SiC, amounts to only about 4 %. This indicates that any significant damage that would control ultimate strength of the material did not occur before the applied loads up to 80 to 90 % of fracture load. Conversely, the damage to control final failure

RkJQdWJsaXNoZXIy MjM0NDE=