S.R. Choi et al. /Journal of the European Ceramic Society 25(2005)1629-1636 b) Fig 3. Fracture patterns for(a) Nicalon/BSAS, (b )SiCr/MAS, (c)SiCr/SiC (enhanced), and(d)Cr/SiC (enhanced")ceramic matrix composites tested in tension at elevated temperatures in air. The upper and lower pictures for a given composite material indicate the specimens tested at the lowest (=0.005 MPa/ and the highest(=5 MPa/s)test rates, respectively specimen-thickness direction. A change in surface(matrix) The strength dependency on test rate exhibited by these color of test specimens from dark grey to white was more pro- composites at elevated temperatures(see Fig. 2)is very simi nounced at 0.005 MPa/s than 5 MPa/s, an evidence of more lar to that commonly observed in advanced monolithic ceram aggressive high-temperature reaction/oxidation involved cs at elevated temperatures. The strength degradation with the lower test rate. attributed to increased test time. fracture decreasing stress rate in monolithic ceramics is known to oc. patterns for the SiCr/MAS composite indicated some fiber cur by a slow crack growth process(also known as delayed pullout with jagged faceted matrix cracking often propagat- failure, subcritical crack growth or fatigue)and is expressed ng along the test-specimen length One specimen tested at as follows the fast test rate of 5 MPa/s failed close to the transition and grip regions. No significant difference in the mode of frac (1) ture was observed between SiCr/SiC (enhanced)and Cr/SiC (standard or enhanced), where almost all the specimens tested where n and D are slow crack growth parameters, and of at either a high or low test rate exhibited relatively flat, straight (MPa)and o(MPa/s)are fracture strength and applied stress fracture surfaces with little fiber pullout, termed brittle frac- rate, respectively. Eq- (1)is based on the conventional power ture. A similar brittle mode of fracture was also observed law crack velocity formulation as expressed previously for the 2D standard SiCf/Sic composite. Black KI to bluish discoloration in the heated region of tested spec U=A mens was obvious for either standard or enhanced Cr/sic. accompanying a weight loss after testing due to oxidation: where u, Kl, and Kle are crack velocity, mode I stress intensity he more weight loss occurred at the lower test rate, and vice factor and fracture toughness, respectively. A is also called versa. Detailed oxidation and stress rupture/life behaviors of slow crack growth parameter. The parameter D is associated this Cr/Sic composite system have been explored previously with n, A, KIc, and inert strength of a material. The parameters in a low partial pressure of oxygen environment.. 7,8 nand D in Eq (1)can be obtained by a linear regression1632 S.R. Choi et al. / Journal of the European Ceramic Society 25 (2005) 1629–1636 Fig. 3. Fracture patterns for (a) Nicalon/BSAS, (b)SiCf/MAS, (c) SiCf/SiC (“enhanced”), and (d) Cf/SiC (“enhanced”) ceramic matrix composites tested in tension at elevated temperatures in air. The upper and lower pictures for a given composite material indicate the specimens tested at the lowest (=0.005 MPa/s) and the highest (=5 MPa/s) test rates, respectively. specimen-thickness direction. A change in surface (matrix) color of test specimens from dark grey to white was more pro￾nounced at 0.005 MPa/s than 5 MPa/s, an evidence of more aggressive high-temperature reaction/oxidation involved at the lower test rate, attributed to increased test time. Fracture patterns for the SiCf/MAS composite indicated some fiber pullout with jagged faceted matrix cracking often propagat￾ing along the test-specimen length. One specimen tested at the fast test rate of 5 MPa/s failed close to the transition and grip regions. No significant difference in the mode of frac￾ture was observed between SiCf/SiC (enhanced) and Cf/SiC (standard or enhanced), where almost all the specimens tested at either a high or low test rate exhibited relatively flat, straight fracture surfaces with little fiber pullout, termed brittle frac￾ture. A similar brittle mode of fracture was also observed previously for the 2D standard SiCf/SiC composite.1 Black to bluish discoloration in the heated region of tested speci￾mens was obvious for either standard or enhanced Cf/SiC, accompanying a weight loss after testing due to oxidation: the more weight loss occurred at the lower test rate, and vice versa. Detailed oxidation and stress rupture/life behaviors of this Cf/SiC composite system have been explored previously in a low partial pressure of oxygen environment.4,7,8 The strength dependency on test rate exhibited by these composites at elevated temperatures (see Fig. 2) is very simi￾lar to that commonly observed in advanced monolithic ceram￾ics at elevated temperatures. The strength degradation with decreasing stress rate in monolithic ceramics is known to oc￾cur by a slow crack growth process (also known as delayed failure, subcritical crack growth or fatigue) and is expressed as follows:9–11 log σf = 1 n + 1 log σ˙ + log D (1) where n and D are slow crack growth parameters, and σf (MPa) and σ˙ (MPa/s) are fracture strength and applied stress rate, respectively. Eq. (1) is based on the conventional power￾law crack velocity formulation as expressed: v = A
KI KIc n (2) where v, KI, and KIc are crack velocity, mode I stress intensity factor and fracture toughness, respectively. A is also called slow crack growth parameter. The parameter D is associated with n, A,KIc, and inert strength of a material. The parameters n and D in Eq. (1) can be obtained by a linear regression anal-