JAm. Ceru. Soc,871259-126402004) urna High-Temperature Fatigue Strength of Crack-Healed Al2O3 Toughened by Sic Whiskers Sang-Kee Lee, Koji Takahashi, *,T,s Masahiro Yokouchi,*, f Hiroyoshi Suenaga, and Kotoji Ando Department of Energy and Safety Engineering, Yokohama National University, 79-5, Hodogaya, Yokohama, 240-8501, Japan Kokan Keisoku KK, Kawasaki-ku, Kawasaki, 210-0855, Japan Al,O3 reinforced by SiC whiskers(Al O,SiC-w)was hot- size that can be completely healed, the high-temperature pressed to investigate the fatigue strength of crack-healed strength of crack-healed zones, o,>- assessment of the cyclic specimens at high temperature Semielliptical surface cracks of and static fatigue strengths of crack-healed ceramic compo- 100 um surface length were introduced on each specimen nents,- and crack-healing behavior under static or cyclic urface. These specimens were crack-healed at 1300C for 1 h stress.2 It has been shown that Si, N, /SiC and mullite/SiC in air, and static and cyclie fatigue strengths were systemati- have excellent crack-healing abilities. It was also reported that cally investigated at room temperature, 900 and 1100oc by three-point bending. The static and cyclic fatigue limits of the ability similar to Si,N /SiC8,22 and mullite/SiC.23,2 crack-healed specimens were more than 70% of the average However, higher fracture toughness is desirable for the struc- ending strength at each testing temperature. Crack-heales tural integrity of Al2O, ceramics. It is well known that whisker specimens of Al2O/SiC-w were not sensitive to statie and reinforcement is very effective for increasing the fracture tough cyclic fatigue at room temperature and high temperatures. ness of structural ceramics. -If whisker-reinforced Al,, could Therefore the combination of crack-healing and whisker express excellent crack-healing ability, it would be desirable for reinforcement can play an important role in increasing static the structural integrity of Al2O, components. In a previous and cyclic fatigue strengths at high temperature. tudy, we sintered Al,O, reinforced by SiC whiskers(AlOy/ Sic-W) and studied basic crack-healing behaviors, We found that the combination of crack-healing and whisker reinforcement is . Introduction very useful for inducing self-crack-healing ability and increasing A LUMINA(Al2,)is a very popular ceramic used in various monotonic strength and fracture toughness fields. However, it has three weak points: low bendin It is essential to determine fatigue behavior under the appropri- strength(400 MPa), low fracture toughness(3 MPam),and ate loading, such as static, dynamic, or cyclic loadings indispens- a low heat-resistance limit(-900oC)for bending strength. These weaknesses limit the application of Al,O, for important compo- tructural purposes. A considerable number of studies have been nents. These weaknesses can be overcome in the following two conducted with regard to the fatigue behavior of engineering ways:(1)by toughening Al, O, with fibers or whiskers, and (2) by ramics 30-35 Few critical studies have been conducted regarding inducing a self-crack-healing ability static and cyclic fatigue strength of alumina ceramics at high Niihara proposed a new concept related to"nanocomposite temperatures, although activity in this field has recently been ceramics.Al,O3/SiC nanocomposite sintered by Niihara et al. ncreasing. In a previous study, we examined the high- temperature strength of Al,O /SiC-W and found that the lim showed excellent bending strength, fracture toughness, and heat- temperature for the bending strength of crack-healed specimen resistance limit for bending strength. Many studies have reinforcement and many useful results have been reported. -ger conducted with the aim of toughening Al,O, by fiber or whisker was 1100C. However, the fatigue strength of the crack-healed specimen was not studied. This paper focuses on the fatigue However, the fracture toughness of Al,O3 is not high, thus behaviors of crack-healed Al,O SiC-w composite ceramics at Al2O3 is very sensitive to cracking and is less reliable than metal. room temperature and high temperatures It is well known that monolithic Al,O3 and Al,O3 reinforced by SiC particles have an interesting crack-healing ability. -4If this ability is used in structural components for engineering uses, great Il. Materials, Specimens, and Test Method benefits can be anticipated, such as increased reliability of the The alumina powder used in this investigation was AKP-20 structural ceramic components and reduced inspection, machining, mean particle size 0.4-0.6 um, purity 99.99%)from Sumitomo nd polishing costs of such compor lowever, before this Chemical (Tokyo, Japan). The SiC whiskers used were SCW No healing ability is realized in structural ceramics, many problem 1-0.8 (length 30-100 um, diameter 0.8-1.0 Hm)from Tateho must be resolved. These include the crack-healing behavior as a Chemical Industries(Ako, Japan). The quantity of Sic whiskers function of healing temperature and time, the effects of ior. 3-I8 the maximum crack added was 20 vol% relative to AL, O, powder. Isopropyl alcohol nt on crack-healing behavi was added to the mixture and the mixture was blended completely for 12 h using alumina balls and a mill pot. Thereafter, the mixture was placed in an evaporator to extract the solvent, and then in a T: Rouxel-contributing editor vacuum desiccator to produce a dry powder mixture. Rectangular plates with dimensions of 9 mm X 50 mm X 50 mm were hot-pressed at 1850.C and 40 MPa for I h in an argon environ ment. The value of fracture toughness(Kic) of the test material ript No. 10275. Received June 25, 2003, approved March 16, 2004. was Kic =5.7 MPam". The density of the test material measured by the Archimedes technique was 3.83 g/cm, which was 99.9% of the theoretical density of the material. The average grain size of to whom correspondence should be addressed. e-mail: kaka @ynu ac jp Al,, was 1-2 um, and most of the Sic whiskers were located in 59
High-Temperature Fatigue Strength of Crack-Healed Al2O3 Toughened by SiC Whiskers Sang-Kee Lee,† Koji Takahashi,* ,†,§ Masahiro Yokouchi,* ,† Hiroyoshi Suenaga,‡ and Kotoji Ando* ,† Department of Energy and Safety Engineering, Yokohama National University, 79-5, Hodogaya, Yokohama, 240-8501, Japan Kokan Keisoku K. K., Kawasaki-ku, Kawasaki, 210-0855, Japan Al2O3 reinforced by SiC whiskers (Al2O3/SiC-W) was hotpressed to investigate the fatigue strength of crack-healed specimens at high temperature. Semielliptical surface cracks of 100 m surface length were introduced on each specimen surface. These specimens were crack-healed at 1300°C for 1 h in air, and static and cyclic fatigue strengths were systematically investigated at room temperature, 900° and 1100°C by three-point bending. The static and cyclic fatigue limits of the crack-healed specimens were more than 70% of the average bending strength at each testing temperature. Crack-healed specimens of Al2O3/SiC-W were not sensitive to static and cyclic fatigue at room temperature and high temperatures. Therefore, the combination of crack-healing and whisker reinforcement can play an important role in increasing static and cyclic fatigue strengths at high temperature. I. Introduction ALUMINA (Al2O3) is a very popular ceramic used in various fields. However, it has three weak points: low bending strength (400 MPa), low fracture toughness (3 MPam1/2), and a low heat-resistance limit (900°C) for bending strength. These weaknesses limit the application of Al2O3 for important components. These weaknesses can be overcome in the following two ways: (1) by toughening Al2O3 with fibers or whiskers, and (2) by inducing a self-crack-healing ability. Niihara proposed a new concept related to “nanocomposite” ceramics.1,2 Al2O3/SiC nanocomposite sintered by Niihara et al. showed excellent bending strength, fracture toughness, and heatresistance limit for bending strength. Many studies have been conducted with the aim of toughening Al2O3 by fiber or whisker reinforcement and many useful results have been reported.3–8 However, the fracture toughness of Al2O3 is not high; thus Al2O3 is very sensitive to cracking and is less reliable than metal. It is well known that monolithic Al2O3 and Al2O3 reinforced by SiC particles have an interesting crack-healing ability.9–14 If this ability is used in structural components for engineering uses, great benefits can be anticipated, such as increased reliability of the structural ceramic components and reduced inspection, machining, and polishing costs of such components. However, before this healing ability is realized in structural ceramics, many problems must be resolved. These include the crack-healing behavior as a function of healing temperature and time,15,16 the effects of environment on crack-healing behavior,13–18 the maximum crack size that can be completely healed,19–21 the high-temperature strength of crack-healed zones,15,16,19–23 assessment of the cyclic and static fatigue strengths of crack-healed ceramic components,21–28 and crack-healing behavior under static or cyclic stress.21,26–28 It has been shown that Si3N4/SiC and mullite/SiC have excellent crack-healing abilities. It was also reported that Al2O3 14 reinforced by SiC particles had excellent crack-healing ability similar to Si3N4/SiC18,22 and mullite/SiC.23,25–27 However, higher fracture toughness is desirable for the structural integrity of Al2O3 ceramics. It is well known that whisker reinforcement is very effective for increasing the fracture toughness of structural ceramics.3–7 If whisker-reinforced Al2O3 could express excellent crack-healing ability, it would be desirable for the structural integrity of Al2O3 components. In a previous study,29 we sintered Al2O3 reinforced by SiC whiskers (Al2O3/ SiC-W) and studied basic crack-healing behaviors. We found that the combination of crack-healing and whisker reinforcement is very useful for inducing self-crack-healing ability and increasing monotonic strength and fracture toughness. It is essential to determine fatigue behavior under the appropriate loading, such as static, dynamic, or cyclic loadings indispensable in the engineering application of ceramic materials for structural purposes. A considerable number of studies have been conducted with regard to the fatigue behavior of engineering ceramics.30–35 Few critical studies have been conducted regarding static and cyclic fatigue strength of alumina ceramics at high temperatures, although activity in this field has recently been increasing.30,34 In a previous study,29 we examined the hightemperature strength of Al2O3/SiC-W and found that the limit temperature for the bending strength of crack-healed specimens was 1100°C. However, the fatigue strength of the crack-healed specimen was not studied. This paper focuses on the fatigue behaviors of crack-healed Al2O3/SiC-W composite ceramics at room temperature and high temperatures. II. Materials, Specimens, and Test Method The alumina powder used in this investigation was AKP-20 (mean particle size 0.4–0.6 m, purity 99.99%) from Sumitomo Chemical (Tokyo, Japan). The SiC whiskers used were SCW No. 1-0.8 (length 30–100 m, diameter 0.8–1.0 m) from Tateho Chemical Industries (Ako, Japan). The quantity of SiC whiskers added was 20 vol% relative to Al2O3 powder. Isopropyl alcohol was added to the mixture and the mixture was blended completely for 12 h using alumina balls and a mill pot. Thereafter, the mixture was placed in an evaporator to extract the solvent, and then in a vacuum desiccator to produce a dry powder mixture. Rectangular plates with dimensions of 9 mm 50 mm 50 mm were hot-pressed at 1850°C and 40 MPa for 1 h in an argon environment. The value of fracture toughness (KIC) of the test material was KIC 5.7 MPam1/2. The density of the test material measured by the Archimedes technique was 3.83 g/cm3 , which was 99.9% of the theoretical density of the material. The average grain size of Al2O3 was 1–2 m, and most of the SiC whiskers were located in T. Rouxel—contributing editor Manuscript No. 10275. Received June 25, 2003; approved March 16, 2004. *Member, American Ceramic Society. † Department of Energy and Safety Engineering, Yokohama National University. ‡ Kokan Keisoku K. K. § Author to whom correspondence should be addressed. e-mail: ktaka@ynu.ac.jp. J. Am. Ceram. Soc., 87 [7] 1259–1264 (2004) 1259 journal
1260 Journal of the American Ceramic Sociery-Lee et al. Vol 87. No. 7 grain boundaries. The hot-pressed plates were then cut into test IlL. Test Results and discussion specimens measuring 3 mm x 4 mm X 23 mm. The tensile surface of the test specimen was finished. Semielliptical surface (1 Morphology of Surface Cracks before and after acks were made at the center of the tensile surface of the test Crack-Healing specimen using a Vickers indenter at a load of 19.6 N. By this Figures 2(a)and (b) show SEM micrographs of precrack introduced by means of the Vickers indentation method, before nethod, semielliptical cracks of 100 um surface length were and after the crack-healing at 1300C for I h in air,respectively It can be clearly observed that crack closure and rebonding were crack-healed at 1300%C for I h in air. The rising rate of the occurred. In Fig. 2, the light regions are alumina and the dark urnace temperature was 10.C/min. This is the optimized healing lition for a surface crack of 100 um, reported in a previou air, oxidation products originating from Sic whiskers(gray phase) tudy. 2 All smooth specimens were also heat-treated at 1300C were clearly observed, The reaction products are considered to b or I h in air before testing. Bending tests and static and cyclic solidified droplets of a glassy phase atigue tests were conducted using a three-point loading system with a span of 16 mm. The test fixtures used for the bending tests (2) Fatigue Strength Behavior at Room Temperature were made of SiC ceramics. The static and cyclic fatigue tests were The results of the static and cyclic fatigue tests at room rformed in air at room temperature, and at elevated temperatures temperature are shown in Fig 3. On the left-hand side of Figs. 3(a) of 900 and 1100C. The applied stress at which a specimen did and(b), the results of monotonic bending tests are also shown. The ot fracture up to 10%s was defined as the static fatigue limit symbols●,Δ,and▲ show the bending strength of the smooth yclic fatigue tests were conducted at a stress ratio of R= 0.2 and specimens heat-treated at 1300.C for I h in air, as-cracked, and using a sine wave with a frequency of 5 Hz. The maximum applied crack-healed specimens, respectively. The bending strength(B) stress at which a specimen did not fracture up to N=2 X 10 of the heat-treated smooth specimens scattered to a certain extent, cycles was defined as the cyclic fatigue limit. The surface cracks, giving a mean fracture stress of about 1000 MPa. Because of before and after crack-healing, and the fracture surface were precracking of 100 um, the bending strength decreased down to observed using a scanning electron microscope(SEM) 4450 MPa. However, the average value of B for the specimen crack-healed at 1300%C for I h in air was about 1000 MPa. Thus gB of the crack-healed specimens recovered to a level similar to ckers that of the heat-treated smooth specimens indentation SiC-whisker Crack SUum uum Pre-crack XIde forum 50um Fig. 2. SEM micrographs showing a surface crack introduced by Vickers Fig. 1. SEM micrographs of surface cracks: (a) indentation and crack; (b) indentation: (a) surface crack before crack-healing,(b) surface crack after crack-healing at 1300oC for I h in air
grain boundaries. The hot-pressed plates were then cut into test specimens measuring 3 mm 4 mm 23 mm. The tensile surface of the test specimen was mirror finished. Semielliptical surface cracks were made at the center of the tensile surface of the test specimen using a Vickers indenter at a load of 19.6 N. By this method, semielliptical cracks of 100 m surface length were made, as shown in Fig. 1. The ratio of depth (a) to half surface length (c) of the crack (aspect ratio) was a/c 0.9. Specimens were crack-healed at 1300°C for 1 h in air. The rising rate of the furnace temperature was 10°C/min. This is the optimized healing condition for a surface crack of 100 m, reported in a previous study.29 All smooth specimens were also heat-treated at 1300°C for 1 h in air before testing. Bending tests and static and cyclic fatigue tests were conducted using a three-point loading system with a span of 16 mm. The test fixtures used for the bending tests were made of SiC ceramics. The static and cyclic fatigue tests were performed in air at room temperature, and at elevated temperatures of 900° and 1100°C. The applied stress at which a specimen did not fracture up to 106 s was defined as the static fatigue limit. Cyclic fatigue tests were conducted at a stress ratio of R 0.2 and using a sine wave with a frequency of 5 Hz. The maximum applied stress at which a specimen did not fracture up to Nf 2 106 cycles was defined as the cyclic fatigue limit. The surface cracks, before and after crack-healing, and the fracture surface were observed using a scanning electron microscope (SEM). III. Test Results and Discussion (1) Morphology of Surface Cracks before and after Crack-Healing Figures 2(a) and (b) show SEM micrographs of precracks introduced by means of the Vickers indentation method, before and after the crack-healing at 1300°C for 1 h in air, respectively. It can be clearly observed that crack closure and rebonding occurred. In Fig. 2, the light regions are alumina and the dark regions are SiC whiskers. After heat treatment at 1300°C for 1 h in air, oxidation products originating from SiC whiskers (gray phase) were clearly observed. The reaction products are considered to be solidified droplets of a glassy phase.29 (2) Fatigue Strength Behavior at Room Temperature The results of the static and cyclic fatigue tests at room temperature are shown in Fig. 3. On the left-hand side of Figs. 3(a) and (b), the results of monotonic bending tests are also shown. The symbols F, , and Œ show the bending strength of the smooth specimens heat-treated at 1300°C for 1 h in air, as-cracked, and crack-healed specimens, respectively. The bending strength (B) of the heat-treated smooth specimens scattered to a certain extent, giving a mean fracture stress of about 1000 MPa. Because of precracking of 100 m, the bending strength decreased down to 450 MPa. However, the average value of B for the specimen crack-healed at 1300°C for 1 h in air was about 1000 MPa. Thus, B of the crack-healed specimens recovered to a level similar to that of the heat-treated smooth specimens. Fig. 1. SEM micrographs of surface cracks: (a) indentation and crack; (b) fracture surface. Fig. 2. SEM micrographs showing a surface crack introduced by Vickers indentation: (a) surface crack before crack-healing, (b) surface crack after crack-healing at 1300°C for 1 h in air. 1260 Journal of the American Ceramic Society—Lee et al. Vol. 87, No. 7
July 2004 High-Temperature Fatigue Strength of Crack-Healed Al-O, Toughened by SiC Whiskers 1261 Mark () We determined the static fatigue limit(oo)of the heat-treated specimen to be about 800 MPa. The crack-healed specimens tested racked specimen(e=100um) under 700 and 800 MPa survived up to 10%s. One specimen leat-treated smooth specimen fractured under 750 MPa, however, the time to failure was quite long(5.6x 105 s). Thus, we determined the static fatigue limit 1400 (c,o) for the crack-healed specimen to be about 750 MPa. The cyclic fatigue test results are shown in Fig. 3(b). The pecimens that did not fracture up to N=2 x 10 cycles are marked by an arrow (-) The cyclic fatigue limit (oo) heat-treated specimen and the crack-healed specimen are about 800 and 850 MPa, respectively 会 The static and cyclic fatigue limits of the crack-healed speci- mens are similar to those of the heat-treated specimens, indicating that the room-temperature fatigue strength of the crack-healed zone is comparable to that of base material. The cyclic fatigue limi of crack-healed specimens is about 10% higher than that of the static fatigue limit. This reason will be discussed later in this paper The ratio of the static fatigue limit(ro) and the cyclic fatigue limit (Uo) to the average bending strength of crack-healed specimens (Un) is about 75% and 85%, respectively(see also Table I). Thus, fatigue limits of crack-healed specimens at room temperature are Healing condition: 1300C, Ih in air Bending tests for the specimens which survived the static and cyclic fatigue tests were also conducted at room temperature. The 10210410° o. for fatigue experimental results are shown on the right-hand sides of Figs. 3(a) and(b). The fatigue-tested specimens exhibited bending strengths Time to Failure, t (sec) similar to those of the monotonically tested specimens as shown on the left-hand side of Figs. 3(a)and(b). These experimental results (a) ndicate that significant crack growth from the crack-healed zone id not occu Figure 4 shows SEM micrographs of the fracture surface after static fatigue tests at room temperature. The slow crack growth can A As-cracked specimen((e=100um)RT be clearly observed, as shown in Fig. 4(a). Figure 4(b) shows the Heat-treated smooth specimen details of the crack-healed zone. The dark regions near SiC whiskers are oxidized, while the light ones not. Figure 4(c)shows 140 Crack-healed specimen the details of the slow crack growth zone. The fracture surface of the slow crack growth zone is quite rough and predominantly Cyclic fatigue test intergranular. Interactions between SiC whiskers and Al,O, ma- R=02,f=5Hz trix, such as bridging and/or pullout are clearly observed. Thus, Sic whiskers contribute to an increase in the crack growth resistance. As seen in Fig. 4(d), intergranular fracture cannot be 10 identified in the final fracture zone 切 In a previous study, we investigated the high-temperature bending strength of Al,,/SiC-W which was precracked(2c 100 um)and crack-healed at 1300C for I h in air. The bending strength of the crack-healed specimen was UB= 800 MPa and was almost constant up to 1100C. However, the UB showed a sudden decrease at 1.C and decreased with increasing testing temper Healing conditior ature up to 1300oC. Therefore, the heat-resistance limit tempera ture for bending strength of the crack-healed specimen was 00C. In the present study, the cyclic and static fatigue strengths 10 of crack-healed specimens were examined at 900"and 1100.C The results of static fatigue tests at high temperature are shown Number of cycles to failure, N(cycle) in Fig. 5. The monotonic bending strength of the heat-treated mooth specimens and the crack-healed specimens at 900% and 1100%C are shown on the left-hand side of Fig. 5. The average bending strength of the heat-treated smooth specimens at 900and Fig 3. Fatigue test results for Al20/SiC-w at room temperature: (a) 1100oC are about 850 MPa. However, the bending strengths of static fatigue test results for crack-healed Al,O,SiC-w,(b)cyclic fatigu crack-healed specimens are about 88% and 76% compared with test results for crack-healed Al2O /SiC-W. Data marked with an asterisk that of the heat-treated smooth specimen at 900 and 1100C, indicate that the fracture occurred outside of the crack-healed zone respectively, and decrease with increasing test temperature The values of the static fatigue limit (o,o) for the crack-healed pecimens at 900 and 1100.C are about 550 and 450 MPa tests are shown in Fig 3(a), respectively. The static fatigue limit of the crack-healed specimen th time to failure. The is about 70% in comparison to the bending strength at each testing specimens that did not fracture up to 10 s are marked by an arrow temperature(see also Table D)
The results of the static fatigue tests are shown in Fig. 3(a), correlating applied constant stress with time to failure. The specimens that did not fracture up to 106 s are marked by an arrow (3). We determined the static fatigue limit (t0) of the heat-treated specimen to be about 800 MPa. The crack-healed specimens tested under 700 and 800 MPa survived up to 106 s. One specimen fractured under 750 MPa; however, the time to failure was quite long (5.6 105 s). Thus, we determined the static fatigue limit (t0) for the crack-healed specimen to be about 750 MPa. The cyclic fatigue test results are shown in Fig. 3(b). The specimens that did not fracture up to Nf 2 106 cycles are marked by an arrow (3). The cyclic fatigue limit (f0) for the heat-treated specimen and the crack-healed specimen are about 800 and 850 MPa, respectively. The static and cyclic fatigue limits of the crack-healed specimens are similar to those of the heat-treated specimens, indicating that the room-temperature fatigue strength of the crack-healed zone is comparable to that of base material. The cyclic fatigue limit of crack-healed specimens is about 10% higher than that of the static fatigue limit. This reason will be discussed later in this paper. The ratio of the static fatigue limit (t0) and the cyclic fatigue limit (f0) to the average bending strength of crack-healed specimens (B) is about 75% and 85%, respectively (see also Table I). Thus, fatigue limits of crack-healed specimens at room temperature are quite high. Bending tests for the specimens which survived the static and cyclic fatigue tests were also conducted at room temperature. The experimental results are shown on the right-hand sides of Figs. 3(a) and (b). The fatigue-tested specimens exhibited bending strengths similar to those of the monotonically tested specimens as shown on the left-hand side of Figs. 3(a) and (b). These experimental results indicate that significant crack growth from the crack-healed zone did not occur. Figure 4 shows SEM micrographs of the fracture surface after static fatigue tests at room temperature. The slow crack growth can be clearly observed, as shown in Fig. 4(a). Figure 4(b) shows the details of the crack-healed zone. The dark regions near SiC whiskers are oxidized, while the light ones not. Figure 4(c) shows the details of the slow crack growth zone. The fracture surface of the slow crack growth zone is quite rough and predominantly intergranular. Interactions between SiC whiskers and Al2O3 matrix, such as bridging and/or pullout are clearly observed. Thus, SiC whiskers contribute to an increase in the crack growth resistance. As seen in Fig. 4(d), intergranular fracture cannot be identified in the final fracture zone. (3) Static Fatigue Strength at High Temperature In a previous study,29 we investigated the high-temperature bending strength of Al2O3/SiC-W which was precracked (2c 100 m) and crack-healed at 1300°C for 1 h in air. The bending strength of the crack-healed specimen was B 800 MPa and was almost constant up to 1100°C. However, the B showed a sudden decrease at 1100°C and decreased with increasing testing temperature up to 1300°C. Therefore, the heat-resistance limit temperature for bending strength of the crack-healed specimen was 1100°C. In the present study, the cyclic and static fatigue strengths of crack-healed specimens were examined at 900° and 1100°C. The results of static fatigue tests at high temperature are shown in Fig. 5. The monotonic bending strength of the heat-treated smooth specimens and the crack-healed specimens at 900° and 1100°C are shown on the left-hand side of Fig. 5. The average bending strength of the heat-treated smooth specimens at 900° and 1100°C are about 850 MPa. However, the bending strengths of crack-healed specimens are about 88% and 76% compared with that of the heat-treated smooth specimen at 900° and 1100°C, respectively, and decrease with increasing test temperature. The values of the static fatigue limit (t0) for the crack-healed specimens at 900° and 1100°C are about 550 and 450 MPa, respectively. The static fatigue limit of the crack-healed specimen is about 70% in comparison to the bending strength at each testing temperature (see also Table I). Fig. 3. Fatigue test results for Al2O3/SiC-W at room temperature: (a) static fatigue test results for crack-healed Al2O3/SiC-W, (b) cyclic fatigue test results for crack-healed Al2O3/SiC-W. Data marked with an asterisk indicate that the fracture occurred outside of the crack-healed zone. July 2004 High-Temperature Fatigue Strength of Crack-Healed Al2O3 Toughened by SiC Whiskers 1261
Journal of the American Ceramic Society-Lee et al. Vol 87. No. 7 Table L. Summary of Bending Strength and Fatigue Limit of Heat-Treated Smooth Specimens and Crack-Healed Specime Test temperature(°C Gn(MPa)t σn(MPa)2 p/oB (% Gn (Mpa) cd/aB(% Heat-treated smooth specimen 1000 800 Crack-healed specimen 75 750 8897 1100 500 h.Static fatigue limit. Cyclic fatigue limit. Room temperature he bending strengths of the specimens that survived static fatigue (4 Cyclic Fatigue Strength at High Temperature tests were also investigated at each testing temperature, ie, 900% and The cyclic fatigue strengths of heat-treated smooth specimens 1100.C. The bending strengths are shown on the right-hand side of and crack-healed specimen have also been investigated at 900 and Fig. 5. The crack-healed specimens that survived the static fatigue test 1 100C. The experimental results are shown in Fig. 6. The values showed bending strengths similar to monotonically tested specimens, of the cyclic fatigue limit(op) of the heat-treated smooth indicating that significant crack growth from the crack-healed zone or specimen and the crack-healed specimen at 900C are about 800 mal defects did not occur and 700 MPa, respectively. These values are more than 90% Slow crack growth Pre-crack A B Sic whisker Oxide 200um SXkax500 5 e.aky×s.ae5u Fig. 4. SEM micrographs of fracture surface of the test specimens fractured during static fatigue test at mperature(slow crack growth occurred during the fatigue test, applied stress =800 MPa, tr=110 s):(a) crack profile, (b) detail of crack-healed zone shown by A in(a),(c)detail of slow crack owth zone shown by B in(a),(d) detail of area outside the slow crack growth zone shown by C in(a)
The bending strengths of the specimens that survived static fatigue tests were also investigated at each testing temperature, i.e., 900° and 1100°C. The bending strengths are shown on the right-hand side of Fig. 5. The crack-healed specimens that survived the static fatigue test showed bending strengths similar to monotonically tested specimens, indicating that significant crack growth from the crack-healed zone or internal defects did not occur. (4) Cyclic Fatigue Strength at High Temperature The cyclic fatigue strengths of heat-treated smooth specimens and crack-healed specimen have also been investigated at 900° and 1100°C. The experimental results are shown in Fig. 6. The values of the cyclic fatigue limit (f0) of the heat-treated smooth specimen and the crack-healed specimen at 900°C are about 800 and 700 MPa, respectively. These values are more than 90% Fig. 4. SEM micrographs of fracture surface of the test specimens fractured during static fatigue test at room temperature (slow crack growth occurred during the fatigue test, applied stress 800 MPa, tf 110 s): (a) crack profile, (b) detail of crack-healed zone shown by A in (a), (c) detail of slow crack growth zone shown by B in (a), (d) detail of area outside the slow crack growth zone shown by C in (a). Table I. Summary of Bending Strength and Fatigue Limit of Heat-Treated Smooth Specimens and Crack-Healed Specimens Test temperature (°C) B (MPa)† t0 (MPa)‡ t0/B (%) f0 (MPa)§ f0/B (%) Heat-treated smooth specimen RT¶ 1000 800 80 800 80 900 850 800 94 1100 850 700 82 Crack-healed specimen RT¶ 1000 750 75 850 85 900 750 550 73 700 93 1100 650 450 69 500 77 † Average bending strength. ‡ Static fatigue limit. § Cyclic fatigue limit. ¶ Room temperature. 1262 Journal of the American Ceramic Society—Lee et al. Vol. 87, No. 7
July 2004 High-Temperature Fatigue Strength of Crack-Healed Al-O, Toughened by SiC Whiskers 1263 Specimen type O.Heat-treated smooth specitmen900C Heat-treated smooth specimen Crack-healed specimen 900C oHeat-treated smooth specitmen1100C oHeat-treated smooth spec trmen1100c Crack -healed specimen l100C Crack-healed specimen l100c 1400 1400 Static fatigue test 400 20[:nm Healing condition: 1300 C, Ih in air Monotonic 10 10 for fatigue 10 le to Failure, t,(sec) Fig. 5. Results of static fatigue test of crack-healed Al,O,SiC-W at 900 and 1100°C Fig. 6. Results of cyclic fatigue test of crack-healed Al2O: /SiC-Wat mpared with the bending strength(oB) of each specimen at 900C, i.e., UoB> 90%(see also Table I static stresses with a maximum stress equal to the maximum cyclic the case of the cyclic fatigue test at 1100oC, the values of the stress. These fatigue test results are reasonable considering that cyclic fatigue limit of the heat-treated smooth specimen and the clic loading with a short duration of maximum stress during crack-healed specimen are about 700 and 500 MPa, respectively. each cyclic provides a beneficial effect on lifetime in comparison The ratios of cyclic fatigue limit to the bending strength of each to static loading at the same maximum stress. specimen at 1100.C are more than 75%0, i.e., U/oB>75%. The In a previous study, - it was found that the combination of ratio of oa/oB decreases with increasing test temperature. How- crack-healing and whisker reinforcement was very useful for ever, it can be said that the ratio of o o /oB is sufficiently high, even the bending strength and the fracture te at 1100°C Al,O3. The results of this study show that the combination of The values of cyclic fatigue limit of crack-healed specimen are crack-healing and whisker reinforcement is very useful to increase lower than those of heat-treated smooth specimens at 900 and static and cyclic fatigue strengths at high temperatures 1100 C. These reductions are caused by differences in the bending strength, as shown on the left-hand side of Fig. 6 and in Table I Therefore, it is necessary to increase both the bending strength and IV. Conclusions the fatigue limit of crack-healed specimens at elevated tempera tures to the same level as heat-treated smooth specimens. This Al,O3 reinforced by SiC whiskers(Al,O,SiC-W)was sintered matter is the subject of further study. and the specimens were subjected to three-point bending. A The bending strength of the specimens which survived the semielliptical surface crack of 100 um surface length was made on cyclic fatigue tests were also investigated. The bending strength each specimen. These specimens were crack-healed at 1300%C for are shown on the right-hand side of Fig. 6. bending tests were h in air, and cyclic and static fatigue strengths were systemati- rformed at each testing temperature, i.e, 900 and 1100.C. cally investigated at room temperature, 900, and 1100C. The Again, the heat-treated specimens and the crack-healed specimens main conclusions obtained are as follows that survived the cyclic fatigue test showed bending strengths (1 The values of static(c o)and cyclic(up) fatigue limits for similar to monotonically tested specimens the crack-healed specimens were 750 and 850 MPa, respectively, Table I summarizes the bending strength, the static fatigue limit, at room temperature. The static and cyclic fatigue limits were and the cyclic fatigue limit at room temperature, 900, and about 75% and 85%, respectively, in comparison to the bending 1 100.C. The cyclic fatigue limit of crack-healed specimens is strength at room temp rature Moreover. the values of on and higher than that of the static fatigue limit, regardless of the test for the heat-treated specimens were 800 MPa. Thus, the static and temperature investigated in this study, i.e., room temperature, cyclic fatigue limits of the crack-healed specimens are similar to 900, and 1100C. This is due to the following. The cyclic(Uo) those of the heat-treated specimens, indicating that room- and static (o,o) fatigue limit are defined as the maximum applied ature fatigue strength of the crack-healed zone is compara- stress at which a specimen did not fracture up to Ne=2 X 10 ble to that of base materials cycles and t=10s. This Nr of the cyclic fatigue test is equivalent (2) The static fatigue limits of crack simens(co)at to 4x 10 s. Thus, the cyclic fatigue test is shorter than the static 00° and 1100°were550and450 spectively atigue test. Moreover, previous studies- showed that cyclic values were about 70% in comparison to trength(oB)at stresses extend the lifetime in Al,O, and Si3N4 compared with each temperature
compared with the bending strength (B) of each specimen at 900°C, i.e., f0/B 90% (see also Table I). In the case of the cyclic fatigue test at 1100°C, the values of the cyclic fatigue limit of the heat-treated smooth specimen and the crack-healed specimen are about 700 and 500 MPa, respectively. The ratios of cyclic fatigue limit to the bending strength of each specimen at 1100°C are more than 75%, i.e., f0/B 75%. The ratio of f0/B decreases with increasing test temperature. However, it can be said that the ratio of f0/B is sufficiently high, even at 1100°C. The values of cyclic fatigue limit of crack-healed specimen are lower than those of heat-treated smooth specimens at 900° and 1100°C. These reductions are caused by differences in the bending strength, as shown on the left-hand side of Fig. 6 and in Table I. Therefore, it is necessary to increase both the bending strength and the fatigue limit of crack-healed specimens at elevated temperatures to the same level as heat-treated smooth specimens. This matter is the subject of further study. The bending strength of the specimens which survived the cyclic fatigue tests were also investigated. The bending strengths are shown on the right-hand side of Fig. 6. Bending tests were performed at each testing temperature, i.e., 900° and 1100°C. Again, the heat-treated specimens and the crack-healed specimens that survived the cyclic fatigue test showed bending strengths similar to monotonically tested specimens. Table I summarizes the bending strength, the static fatigue limit, and the cyclic fatigue limit at room temperature, 900°, and 1100°C. The cyclic fatigue limit of crack-healed specimens is higher than that of the static fatigue limit, regardless of the test temperature investigated in this study, i.e., room temperature, 900°, and 1100°C. This is due to the following. The cyclic (f0) and static (t0) fatigue limit are defined as the maximum applied stress at which a specimen did not fracture up to Nf 2 106 cycles and tf 106 s. This Nf of the cyclic fatigue test is equivalent to 4 105 s. Thus, the cyclic fatigue test is shorter than the static fatigue test. Moreover, previous studies36–38 showed that cyclic stresses extend the lifetime in Al2O3 and Si3N4 compared with static stresses with a maximum stress equal to the maximum cyclic stress. These fatigue test results are reasonable considering that cyclic loading with a short duration of maximum stress during each cyclic provides a beneficial effect on lifetime in comparison to static loading at the same maximum stress. In a previous study,29 it was found that the combination of crack-healing and whisker reinforcement was very useful for increasing the bending strength and the fracture toughness of Al2O3. The results of this study show that the combination of crack-healing and whisker reinforcement is very useful to increase static and cyclic fatigue strengths at high temperatures. IV. Conclusions Al2O3 reinforced by SiC whiskers (Al2O3/SiC-W) was sintered and the specimens were subjected to three-point bending. A semielliptical surface crack of 100 m surface length was made on each specimen. These specimens were crack-healed at 1300°C for 1 h in air, and cyclic and static fatigue strengths were systematically investigated at room temperature, 900°, and 1100°C. The main conclusions obtained are as follows: (1) The values of static (t0) and cyclic (f0) fatigue limits for the crack-healed specimens were 750 and 850 MPa, respectively, at room temperature. The static and cyclic fatigue limits were about 75% and 85%, respectively, in comparison to the bending strength at room temperature. Moreover, the values of t0 and f0 for the heat-treated specimens were 800 MPa. Thus, the static and cyclic fatigue limits of the crack-healed specimens are similar to those of the heat-treated specimens, indicating that roomtemperature fatigue strength of the crack-healed zone is comparable to that of base materials. (2) The static fatigue limits of crack-healed specimens (t0) at 900° and 1100°C were 550 and 450 MPa, respectively. These values were about 70% in comparison to bending strength (B) at each temperature. Fig. 5. Results of static fatigue test of crack-healed Al2O3/SiC-W at 900° and 1100°C. Fig. 6. Results of cyclic fatigue test of crack-healed Al2O3/SiC-W at 900° and 1100°C. July 2004 High-Temperature Fatigue Strength of Crack-Healed Al2O3 Toughened by SiC Whiskers 1263
Journal of the American Ceramic Sociery-Lee et al. Vol 87. No. 7 ( The cyclic fatigue limits of the crack-healed specimens I6Y. W. Kim, K. Ando, and M. C. Chu, "Crack Healing Behavior of (oo) at 900 and 1100C were 700 and 500 MPa. The ratio of cyclic fatigue limit(o) to the bending strength of the crack- M. C. Chu, S. Sato, Y. Kobayashi, and K. Ando,"Damage healed specimens(oB) at 900oC was more than 90%, i.e., eoB Mat engthening Behaviour in Intelligent Mullite/SiC Ceramics, Fatigue Fract. Eng >90%o, while the U/on of the crack-healed ndo, T Ikeda, S. Sato, F, Yao, and Y. Kobayashi, "A Preliminary Study on was about 75%. Moreover the values of g Crack Healing Behaviour of SiaN,/SiC Composite Ceramics, Fatigue Fract. Eng. specimens at 900 and 1 100oC were 800 and 700 MPa, respec- Mater. Struct,21,119-22(1998) tively. The ratios of cyclic fatigue limit to the bending strength of S K. Lee, w. Ishida, and K. Ando, "Strength Properties of Crack-Healed Silicon ach specimen at 1100C are more than 75%, i.e., Ua oB>75% Carbide Ceramics"(in Jpn ),J. Soc. Mater. Sci. Jpn, 52, 674-80(2002 20Y. Z. Zhang, L. Edwards, and W. J. Plumbridge, "Crack Healing in a Silicon The ratio of o/oB decreases with increasing test temperature. Nitride Ceramics, J.Anm. Ceram. Soc., 81, 34-37(1998). However, it can be said that the ratio of o on is sufficiently high, 2K. Ando, K. Furusawa, M. C. Chu, T. Hanagata, K. Tuji, and S. Sato, even at 1100°C icon Carbide Ceramics and the 4) From the above results, crack-healed Al,O/SiC-W w Resultant Fatigue Strength, J. Am. Ceram Soc., 84[9]2073-78(200 not sensitive to static and cyclic fatigue at room tem 2Y. Korous, M. C. Chu, M. Nakatani, and K. Ando, "Crack Healing Behavior of SiC Ceramics, J. Am. Ceram Soc., 83(11 2788-92(2000 high temperature. Therefore, the combination of crack lg an 2F. Yao, K. Ando, M. C. Chu, and S. Sato,"Crack-Healing Behavior, High- whisker reinforcement is very useful for static and cyclic fatigue Strength of Sic-Reinforced Silicon Nitride Composite, J. high temperature F. Yao, K. Ando, M. C. Chu, and S Sato, "Static and Cyclic Fatigue Behaviour References of Crack-Healed Si, N//SiC Composite Ceramics, J. Eur. Ceran. Soc., 21, 991-97 Ando, M. C. Chu, K. Tsuji, T Hit irasawa, Y Kobayashi, and S Sato, "Crack A. Nakahira, G. Sasaki, and M. Hirabayashi, "Development of Strong posites, Proc. Int. Meet Aav. Mater. Mater. Res. Soc. JpvL, 4 Healing Behavior of Mullite/SiC Composite Ceramics, "J. Eur. Ceram. Soc., 22, 1313-1902002) 24-34(1989) K. Houjyou, M. C. Chu, S Takeshita, K. Takahashi, S Sakamoto, and Niihara,"New Design Concept of Structural Ceramics Nanocomposites, S. Sato, "Crack-Healing Behavior of Si3N//SiC Ceramics Under Stress and Fatigue 3. Zhao, L. C. Stearns, M. P. Harmer, H. M. Chan, and G. A Miller,"Mechan Strength at the Temperature of Healing(1000C),J. Eur. Ceram Soc., 22, 1339-46 Behavior of Alumina-Silicon Nanocomposites, "J. Am. Ceram Soc., 762)503- (2002 (1993) 27K. Ando, K. Takahashi, S Nakayama, and S. Saito, "Crack-Healing Behavior of Alumina. " J. Anm. Ceram. Soc,6712(C257C269(9 Si3N/SiC Ceramics under Cyclic Stress and Resultant Fatigue Strength at the Healing Temperature, "J. Am. Ceram Soc., 85[9] 2268-72(2002 SG. C. Wei and P. F. Becher "Development of SiC-Whisker-Reinforced Ceram- 2*Takahashi, B. S. Kim, M. C. Chu, S. Sato, and K. Ando, " Cracked-healing ics,"Am. Ceram. Soc. Bull, 64 [2]298-304(1985)- Behavior of SiaN /SiC Ceramics Under Stress and Static Fatigue Strength at th J. Homeny, W.L. Vaughn, and M. K. Ferber, "Silicon Carbide Whisker/Alumina lealing Temperatures of 800, 900, 1000C,J. Eur. Ceram. Soc., 23, 1971-78 Cm9楼( Irface Treatment on Fracture Toughness, 2003) ndo,“ Crack- Healin 7E. S. Fisher, M. H. Manghnani, and J. F. Wang, "Elastic Properties of Al2O3 and of Al2O, Toughened by SiC Whiskers, "J.A. Ceram Soc., 86[ 12]2143-47(2003) Si3N4 Matrix Composites with Sic Whisker Reinforcement,J. Am. Ceram Soc., 75 3K. Jakus, T. Service, and J. E. Ritter Jr, "High-Tem Fatigue Behavior olycrystalline Alumina, "J. Am. Ceran. Soc., 63[1]4-7(1980) H E. Kim and A.J. Moorhead, "Oxidation Behavior and Effects of Oxidation on T, Kawakubo and K, Komeya, "Static and Cyclic Fatigue Behavior of a Sn the Strength of Sic-Whisker-Reinforced Alumina, J. Mater. Sci., 29, 1656-61 Silicon Nitride at Room Temperature,J.An. Ceram. Soc., 706]400-405(1987) 3R. H. Dauskardt, M. R. James, J.R. Porter, and R.O.Ritchie,"Cyclic F. F. Lange and K. C. Radford, ""Healing of Surface Cracks in Polycrystalline Fatigue-Crack Growth in a SiC-Whisker-Reinforced Alumina Ceramic Composite: Al2O3,”J.Am. Ceran.Soc,s30-1l420-21(1970) gupta, "Crack Healing and Strengthening of Thermally Shocked Alumin Long- and Small-Crack Behavior, J. An Ceram Soc., 75 14]759-71(1992)- J.Am. Ceran.Soc,59[5-6259-62(1976) 3L. R. Deobald and A.S. Kobayashi, "Dynamic Fracture Characterization of IA. M. Thompson, H. M. Chan, M. P. Harmer, and R. F. Cook, "Crack Healing Al2O, and SICwIAl2O3, " J. Am. Ceram Soc., 75 [10]2867-70(1992 Stress Relaxation in Al2O SiC Nanocomposites, "J.Am. Ceram Soc., 783] S. Leung, E. G. Mehrtens, G. T. Stevens, S. Bandyopadhyay, and C. C. Sorrell "On High Temperature Mechanical and Fracture Properties of an AlO3/SiCw 12J. E. Moffatt, W. J. Plumbridge, and R. Hermann, "High Temperature Crack C= posite,1m(① Annealing Effect on Fracture Toughness of Alumina and Alumina-SiC Composite Fatigue in SiC-Whisker-Reinforced Silicon Nitride Composite, Mater. Sci. Eng L. A Chou, H M. Chan, and M. P. Harmer, ""Effect of Annealing Environmen on the Crack Healing and Mechanical Behavior of Silicon Carbide-Reinforced 36D. S. Jacobs and 1-w. Chen. "Mechanical and Environmental Factors in the Nanocomposites, J. An Cyclic and Static Fatigue of Silicon Nitride,JAm. Ceram Soc., 77[5] 1153-61 umina and Strength of Crack-Healed Member"(in Jpn ),J. Soc. Mate K. J. Lin and D. F. Socie, "Static and Cyclic Fatigue of Alumina at High Sc.Jm,526667-73(200 Temperature, J Am Ceram Soc., 74 [7] 1511-18(1991)- K. Ando, M. C. Chu, F. Yao, and S Sato, "Fatigue Strength of Crack Healed masuda, T. Sowa, M. Matsui, and 1. Oda, "Fatigue of Ceramics(P Si3N,SiC Composite Ceramics, Fatigue Fract Eng. Mater. Struct, 22 [10] 3HCyclic Fatigue Behavior of Sintered Si3 N4 of High Temperature,JCeram Jn,97,612-18(1989
(3) The cyclic fatigue limits of the crack-healed specimens (f0) at 900° and 1100°C were 700 and 500 MPa. The ratio of cyclic fatigue limit (f0) to the bending strength of the crackhealed specimens (B) at 900°C was more than 90%, i.e., f0/B 90%, while the f0/B of the crack-healed specimens at 1100°C was about 75%. Moreover, the values of f0 for the heat-treated specimens at 900° and 1100°C were 800 and 700 MPa, respectively. The ratios of cyclic fatigue limit to the bending strength of each specimen at 1100°C are more than 75%, i.e., f0/B 75%. The ratio of f0/B decreases with increasing test temperature. However, it can be said that the ratio of f0/B is sufficiently high, even at 1100°C. (4) From the above results, crack-healed Al2O3/SiC-W was not sensitive to static and cyclic fatigue at room temperature and high temperature. Therefore, the combination of crack-healing and whisker reinforcement is very useful for static and cyclic fatigue at high temperatures. References 1 K. Niihara, A. Nakahira, G. Sasaki, and M. Hirabayashi, “Development of Strong Al2O3/SiC Composites,” Proc. Int. Meet. Adv. Mater., Mater. Res. Soc. Jpn., 4, 124–34 (1989). 2 K. Niihara, “New Design Concept of Structural Ceramics Nanocomposites,” J. Ceram. Soc. Jpn., 99 [10] 974–82 (1991). 3 J. Zhao, L. C. Stearns, M. P. Harmer, H. M. Chan, and G. A. Miller, “Mechanical Behavior of Alumina–Silicon Nanocomposites,” J. Am. Ceram. Soc., 76 [2] 503–10 (1993). 4 P. F. Becher and G. C. Wei, “Toughening Behavior in SiC-Whisker-Reinforced Alumina,” J. Am. Ceram. Soc., 67 [12] C-267–C-269 (1984). 5 G. C. Wei and P. F. Becher, “Development of SiC-Whisker-Reinforced Ceramics,” Am. Ceram. Soc. Bull., 64 [2] 298–304 (1985). 6 J. Homeny, W. L. Vaughn, and M. K. Ferber, “Silicon Carbide Whisker/Alumina Matrix Composite: Effect of Whisker Surface Treatment on Fracture Toughness,” J. Am. Ceram. Soc., 73 [2] 394–402 (1990). 7 E. S. Fisher, M. H. Manghnani, and J. F. Wang, “Elastic Properties of Al2O3 and Si3N4 Matrix Composites with SiC Whisker Reinforcement,” J. Am. Ceram. Soc., 75 [4] 908–14 (1992). 8 H. E. Kim and A. J. Moorhead, “Oxidation Behavior and Effects of Oxidation on the Strength of SiC-Whisker-Reinforced Alumina,” J. Mater. Sci., 29, 1656–61 (1994). 9 F. F. Lange and K. C. Radford, “Healing of Surface Cracks in Polycrystalline Al2O3,” J. Am. Ceram. Soc., 53 [10–11] 420–21 (1970). 10T. K. Gupta, “Crack Healing and Strengthening of Thermally Shocked Alumina,” J. Am. Ceram. Soc., 59 [5–6] 259–62 (1976). 11A. M. Thompson, H. M. Chan, M. P. Harmer, and R. F. Cook, “Crack Healing and Stress Relaxation in Al2O3–SiC Nanocomposites,” J. Am. Ceram. Soc., 78 [3] 567–71 (1995). 12J. E. Moffatt, W. J. Plumbridge, and R. Hermann, “High Temperature Crack Annealing Effect on Fracture Toughness of Alumina and Alumina–SiC Composite,” Br. Ceram. Trans., 95 [1] 23–29 (1996). 13I. A. Chou, H. M. Chan, and M. P. Harmer, “Effect of Annealing Environment on the Crack Healing and Mechanical Behavior of Silicon Carbide-Reinforced Alumina Nanocomposites,” J. Am. Ceram. Soc., 81 [5] 1203–208 (1998). 14B. S. Kim, K. Ando, M. C. Chu, and S. Saito, “Crack-Healing Behavior of Monolithic Alumina and Strength of Crack-Healed Member” (in Jpn.), J. Soc. Mater. Sci. Jpn., 52 [6] 667–73 (2002). 15K. Ando, M. C. Chu, F. Yao, and S. Sato, “Fatigue Strength of Crack Healed Si3N4/SiC Composite Ceramics,” Fatigue Fract. Eng. Mater. Struct., 22 [10] 897–903 (1999). 16Y. W. Kim, K. Ando, and M. C. Chu, “Crack Healing Behavior of Liquid-PhaseSintered Silicon Carbide Ceramics,” J. Am. Ceram. Soc., 86 [3] 465–70 (2003). 17M. C. Chu, S. Sato, Y. Kobayashi, and K. Ando, “Damage Healing and Strengthening Behaviour in Intelligent Mullite/SiC Ceramics,” Fatigue Fract. Eng. Mater. Struct., 18 [9] 1019–29 (1995). 18K. Ando, T. Ikeda, S. Sato, F. Yao, and Y. Kobayashi, “A Preliminary Study on Crack Healing Behaviour of Si3N4/SiC Composite Ceramics,” Fatigue Fract. Eng. Mater. Struct., 21, 119–22 (1998). 19S. K. Lee, W. Ishida, and K. Ando, “Strength Properties of Crack-Healed Silicon Carbide Ceramics” (in Jpn.), J. Soc. Mater. Sci. Jpn., 52, 674–80 (2002). 20Y. Z. Zhang, L. Edwards, and W. J. Plumbridge, “Crack Healing in a Silicon Nitride Ceramics,” J. Am. Ceram. Soc., 81, 34–37 (1998). 21K. Ando, K. Furusawa, M. C. Chu, T. Hanagata, K. Tuji, and S. Sato, “Crack-Healing Behavior under Stress of Mullite/Silicon Carbide Ceramics and the Resultant Fatigue Strength,” J. Am. Ceram. Soc., 84 [9] 2073–78 (2001). 22Y. Korous, M. C. Chu, M. Nakatani, and K. Ando, “Crack Healing Behavior of SiC Ceramics,” J. Am. Ceram. Soc., 83 [11] 2788–92 (2000). 23F. Yao, K. Ando, M. C. Chu, and S. Sato, “Crack-Healing Behavior, Hightemperature and Fatigue Strength of SiC-Reinforced Silicon Nitride Composite,” J. Mater. Sci. Lett., 12 [19] 1081–83 (2000). 24F. Yao, K. Ando, M. C. Chu, and S. Sato, “Static and Cyclic Fatigue Behaviour of Crack-Healed Si3N4/SiC Composite Ceramics,” J. Eur. Ceram. Soc., 21, 991–97 (2001). 25K. Ando, M. C. Chu, K. Tsuji, T. Hirasawa, Y. Kobayashi, and S. Sato, “Crack Healing Behavior of Mullite/SiC Composite Ceramics,” J. Eur. Ceram. Soc., 22, 1313–19 (2002). 26K. Ando, K. Houjyou, M. C. Chu, S. Takeshita, K. Takahashi, S. Sakamoto, and S. Sato, “Crack-Healing Behavior of Si3N4/SiC Ceramics Under Stress and Fatigue Strength at the Temperature of Healing (1000°C),” J. Eur. Ceram. Soc., 22, 1339–46 (2002). 27K. Ando, K. Takahashi, S. Nakayama, and S. Saito, “Crack-Healing Behavior of Si3N4/SiC Ceramics under Cyclic Stress and Resultant Fatigue Strength at the Healing Temperature,” J. Am. Ceram. Soc., 85 [9] 2268–72 (2002). 28Takahashi, B. S. Kim, M. C. Chu, S. Sato, and K. Ando, “Cracked-Healing Behavior of Si3N4/SiC Ceramics Under Stress and Static Fatigue Strength at the Healing Temperatures of 800, 900, 1000°C,” J. Eur. Ceram. Soc., 23, 1971–78 (2003). 29K. Takahashi, M. Yokouchi, S. K. Lee, and K. Ando, “Crack-Healing Behavior of Al2O3 Toughened by SiC Whiskers,” J. Am. Ceram. Soc., 86 [12] 2143–47 (2003). 30K. Jakus, T. Service, and J. E. Ritter Jr., “High-Temperature Fatigue Behavior of Polycrystalline Alumina,” J. Am. Ceram. Soc., 63 [1] 4–7 (1980). 31T. Kawakubo and K. Komeya, “Static and Cyclic Fatigue Behavior of a Sintered Silicon Nitride at Room Temperature,” J. Am. Ceram. Soc., 70 [6] 400–405 (1987). 32R. H. Dauskardt, M. R. James, J. R. Porter, and R. O. Ritchie, “Cyclic Fatigue-Crack Growth in a SiC-Whisker-Reinforced Alumina Ceramic Composite: Long- and Small-Crack Behavior,” J. Am. Ceram. Soc., 75 [4] 759–71 (1992). 33L. R. Deobald and A. S. Kobayashi, “Dynamic Fracture Characterization of Al2O3 and SiCW/Al2O3,” J. Am. Ceram. Soc., 75 [10] 2867–70 (1992). 34S. Leung, E. G. Mehrtens, G. T. Stevens, S. Bandyopadhyay, and C. C. Sorrell, “On High Temperature Mechanical and Fracture Properties of an Al2O3/SiCw Ceramic Matrix Composite,” J. Mater. Sci. Lett., 13, 817–20 (1994). 35S. Zhu, M. Mizuno, Y. Kagawa, Y. Nagano, and H. Kaya, “Static and Cyclic Fatigue in SiC-Whisker-Reinforced Silicon Nitride Composite,” Mater. Sci. Eng., A251, 113–20 (2002). 36D. S. Jacobs and I-W. Chen, “Mechanical and Environmental Factors in the Cyclic and Static Fatigue of Silicon Nitride,” J. Am. Ceram. Soc., 77 [5] 1153–61 (1994). 37C.-K. J. Lin and D. F. Socie, “Static and Cyclic Fatigue of Alumina at High Temperature,” J. Am. Ceram. Soc., 74 [7] 1511–18 (1991). 38M. Masuda, T. Sowa, M. Matsui, and I. Oda, “Fatigue of Ceramics (Part 3)—Cyclic Fatigue Behavior of Sintered Si3N4 of High Temperature,” J. Ceram. Soc. Jpn., 97, 612–18 (1989). 1264 Journal of the American Ceramic Society—Lee et al. Vol. 87, No. 7