MIATERIALS SENE S ENGEERING ELSEVIER Materials Science and Engineering A345 (2003)28-35 Mechanical properties of several advanced Tyranno-SA fiber reinforced CVI-SiC matrix composites Wen Yang a, b *, Tetsuji Noda Hiroshi Araki, Jinnan Yu, Akira Kohyama d Kohyama laboratory, Institute of Adranced Energy, Kyoto University, Kyoto 611-0011, Japan Nano-Material Laboratory, Nano-Fabrication Research Group, 5th Sub-Group, National Institute for Materials Science, CREST, 1-2-1 Sengen Tsukuba, Ibaraki 305-0047, Japan China Institute of Atomic Energy, 102413 Beijing, PR China CREST, Japan Science and Technology Corporation, 4-1-8 Kawaguchi, Saitama 332-0012, Japan Received 30 November 2001: received in revised form 7 March 2002 Co A recently developed SiC fiber, Tyranno-SA (2D plain-woven ) was used as the reinforcement in several SiC/SiC composites. The re fabricated by chemical vapor infiltration(CVi)process. The mechanical properties and fracture behaviors were investigated using three-point bending test. The Tyranno-SA fiber possesses rough fiber surface with pure SiC surface chemistry, which may result in strong fiber/matrix bonding and fiber sliding resistance. Various pyrolytic carbon(PyC) and SiC/PyC interlayer coatings were applied in the composites to modify the mechanical properties of the interface. The interlayers were deposited by isothermal CVI process. The test results revealed a close Pyc layer dependence of the strength of the composites. The ultimate flexural strength(UFS)increased with the increasing of the Pyc layer thickness up to 100 nm, and then, kept at similar level till 200 nm. The Tyranno-SA/SiC composites exhibited relatively high proportional limit stresses due mainly to the large Youngs modulus of the fiber. Fiber pullouts were observed at the fracture surfaces of all the interlayered composites. Attractive promising was exhibited on further improvement of the mechanical properties of the composites through further improvement of the interfacial properties and the matrix densification process C 2002 Elsevier Science B V. All rights reserved Keywords: Tyranno-SA fiber; Chemical vapor infiltration(CVn) process; PyC and SiC/PyC interlayer coatings; Flexural properties 1. Introduction For SiC/SiC composites, the property of fiber/matrix interface is one of the key factors [7-ll] that determine There is a strong and increasing interest in the r&d the materials performance, primarily because that of continuous SiC fiber reinforced Sic matrix compo- damage tolerance results from the deviation of matrix sites(SiC/SiC composites)for a variety of high-tem- cracks into the interfaces. This phenomina can be perature, high-stress applications in aerospace, hot controlled through deposition of a thin coating layer engine and energy conversion systems [1-4]. They are on the fibers. Carbon remains the most frequently used also quite attractive candidates for blanket first wall interlayer. Recently, an alternating multiple interlayers structures in nuclear fusion power systems due mainly to (pyrolytic carbon(PyC)-SiC)n, have been developed for their inherent low induced radioactivation. radiation improved oxidation and radiation resistance [12-14 resistance and chemical stability at elevated tempera- However, the understandings already established are tures [1, 5, 6 mostly for the old-generation SiC-based fibers such Nicalon-CG. Recently an advanced fiber, Tyranno-SA which is predominantly B-Sic crystals with a near stoichiometric C/Si atomic ratio, has been developed Corresponding author. Tel: +81-298-59-2842: Fax: +81-298-59 Ube Industry Ltd Japan[15]. Several selected proper ties of the Tyranno-SA fiber are listed in Table 1. This E-mail address: yang wen(@nims.go. jp (W. Yang). fiber exhibits excellent mechanical properties, coupled 0921-5093/02S ter c 2002 Elsevier Science B.V. All rights reserved. PI:S092150
Mechanical properties of several advanced Tyranno-SA fiberreinforced CVI-SiC matrix composites Wen Yang a,b,, Tetsuji Noda b , Hiroshi Araki b , Jinnan Yu c , Akira Kohyama d a Kohyama laboratory, Institute of Advanced Energy, Kyoto University, Kyoto 611-0011, Japan b Nano-Material Laboratory, Nano-Fabrication Research Group, 5th Sub-Group, National Institute for Materials Science, CREST, 1-2-1 Sengen, Tsukuba, Ibaraki 305-0047, Japan c China Institute of Atomic Energy, 102413 Beijing, PR China d CREST, Japan Science and Technology Corporation, 4-1-8 Kawaguchi, Saitama 332-0012, Japan Received 30 November 2001; received in revised form 7 March 2002 Abstract A recently developed SiC fiber, Tyranno-SA (2D plain-woven), was used as the reinforcement in several SiC/SiC composites. The composites were fabricated by chemical vapor infiltration (CVI) process. The mechanical properties and fracture behaviors were investigated using three-point bending test. The Tyranno-SA fiber possesses rough fiber surface with pure SiC surface chemistry, which may result in strong fiber/matrix bonding and fiber sliding resistance. Various pyrolytic carbon (PyC) and SiC/PyC interlayer coatings were applied in the composites to modify the mechanical properties of the interface. The interlayers were deposited by isothermal CVI process. The test results revealed a close PyC layer dependence of the strength of the composites. The ultimate flexural strength (UFS) increased with the increasing of the PyC layer thickness up to 100 nm, and then, kept at similar level till 200 nm. The Tyranno-SA/SiC composites exhibited relatively high proportional limit stresses due mainly to the large Young’s modulus of the fiber. Fiber pullouts were observed at the fracture surfaces of all the interlayered composites. Attractive promising was exhibited on further improvement of the mechanical properties of the composites through further improvement of the interfacial properties and the matrix densification process. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Tyranno-SA fiber; Chemical vapor infiltration (CVI) process; PyC and SiC/PyC interlayer coatings; Flexural properties 1. Introduction There is a strong and increasing interest in the R&D of continuous SiC fiber reinforced SiC matrix composites (SiC/SiC composites) for a variety of high-temperature, high-stress applications in aerospace, hot engine and energy conversion systems [1/4]. They are also quite attractive candidates for blanket first wall structures in nuclear fusion power systems due mainly to their inherent low induced radioactivation, radiation resistance and chemical stability at elevated temperatures [1,5,6]. For SiC/SiC composites, the property of fiber/matrix interface is one of the key factors [7/11] that determine the materials performance, primarily because that damage tolerance results from the deviation of matrix cracks into the interfaces. This phenomina can be controlled through deposition of a thin coating layer on the fibers. Carbon remains the most frequently used interlayer. Recently, an alternating multiple interlayers, (pyrolytic carbon (PyC)/SiC)n , have been developed for improved oxidation and radiation resistance [12/14]. However, the understandings already established are mostly for the old-generation SiC-based fibers such as Nicalon-CG. Recently an advanced fiber, Tyranno-SA, which is predominantly b-SiC crystals with a near stoichiometric C/Si atomic ratio, has been developed (Ube Industry Ltd. Japan) [15]. Several selected properties of the Tyranno-SA fiber are listed in Table 1. This fiber exhibits excellent mechanical properties, coupled Corresponding author. Tel.: /81-298-59-2842; Fax: /81-298-59- 2801 E-mail address: yang.wen@nims.go.jp (W. Yang). Materials Science and Engineering A345 (2003) 28/35 www.elsevier.com/locate/msea 0921-5093/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 9 2 1 - 5 0 9 3 ( 0 2 ) 0 0 4 6 8 - 9
w. Yang et al Materials Science and Engineering 4345(2003)28-35 Table I Several selected properties of the Tyranno-SA fiber C/Si Diameter(um) Structure Tensile strength (MPa) Tensile modulus( Gpa) Elongation(%) Thermal conductivity (W m-IK-) 1.087.5 Crystal 2510 646(RT) The fiber and properties were provided by the Ube Industry Ltd. Japan. with much improved thermal conductivity and thermal be found elsewhere [18]. The densification process stability. Good radiation resistance is also expected due generally required x 15 h to its stoichiometric chemistry and high-crystalline structure [16]. Furthermore, the fabrication cost is 2.2. Mechanical tests only about one third of that for the Hi-Nicalon Type- S fiber [1]. The excellent performance and low fabrica Three-point bending tests(with a support span of 18 tion cost make the Tyranno-SA fiber very attractive. nm) were performed at room temperature. Bending particularly for nuclear fusion application. However, 1 specimens were cut parallel to one of the fiber bundle remains unclear about the mechanical performances of directions of the fabric cloth using a diamond cutter and SiC/SiC composites reinforced with the Tyranno-sa both the tensile and compression surfaces were carefully fiber. The Tyranno-sa fiber possesses quite different ground using diamond slurry to eliminate the effects of surface characteristics such as near stoichiometric sic surface CVD-SiC layers which were formed at the end of surface chemistry with rough surface compared with the the CVI process [18]. Three tests were conducted for Nicalon-CG and Hi-Nicalon fibers [17]. Pure Sic each composite. The dimension of the specimen is 30X surface chemistry and rough fiber surface may have 4.0Wx1.5 mm. The crosshead speed was 0.0083 significant affections on the interfacial bonding and mm s-I. The load-displacement data was recorded fiber sliding, and therefore, on the mechanical properties of the composites. This is also an important issue to be 2.3 Microstructural characterization understood In this study, several CVI-SiC/SiC composites rein forced with the Tyranno-SA fiber were fabricated with The morphologies, thickness, and uniformities of the interlayers were examined with scanning electron mict varIous PyC and SiC/PyC interlayers. The flexural scope(SEM) using JOEL JIM-6700F. The interlayer properties and fracture behaviors were studied using thickness was measured with an estimated resolution of three-point bending tests. The main objectives are to get 10 nm understanding of the mechanical properties of The fracture surfaces were observed with seM various composites. The simple bending tests were used interfacial debonding and fiber pullout to study the effects of the rough fiber surface character istics and the various interlayers 3.1. Compo 2. 1. Composite processing Totally six composites were fabricated. The interlayer structures and thickness and composite densities/poros- Fibrous preforms were fabricated by stacking ities are listed in table 2. the fiber volume fractions are layers of 2D plain-woven Tyranno-SA fiber cloths in 42-44%. No intentional interlayer was applied in 0/90. The preforms were compressed to keep a fiber composite T-NL. Single PyC interlayer was deposited volume fraction of x43% using a set of graphite in composites T-C50, T-C100, and T-C200. In order fixtures. The normal size of the preforms was 40 mm investigate the effect of Sic layer in Tyranno-SA/SIC in diameter and 2.0 mm in thickness. The preforms were composites, two composites, T-SiC/C80 and T-SiC/ pre-coated with single PyC layers of different thickness C150, were deposited with SiC/Pyc bi-interlayers. The or SiC/Pyc bi-layers using an isothermal chemical vapor interlayer thickness and space homogeneity are as in infiltration(CVI) process through the thermal decom- Table 2. Fig. I shows the SEM images of the interlayers position of methane and CH3Sicl3(MTS), respectively. of composites T-C200 and T-SiC/C150. A uniform 200 MTS was carried by hydrogen. nIm- In compo The pre-coated preforms were finally densified with C200 while in composite T-SiC/C150, a SiC layer of 150 Sic matrix by an isothermal- forced flow CVi proces nm thickness was deposited on the fibers prior to the 1273 K and 14.7 kPa. Detailed fabrication process can deposition of 150 nm-thick PyC layer. SEM interlayer
with much improved thermal conductivity and thermal stability. Good radiation resistance is also expected due to its stoichiometric chemistry and high-crystalline structure [16]. Furthermore, the fabrication cost is only about one third of that for the Hi-Nicalon TypeS fiber [1]. The excellent performance and low fabrication cost make the Tyranno-SA fiber very attractive, particularly for nuclear fusion application. However, it remains unclear about the mechanical performances of SiC/SiC composites reinforced with the Tyranno-SA fiber. The Tyranno-SA fiber possesses quite different surface characteristics such as near stoichiometric SiC surface chemistry with rough surface compared with the Nicalon-CG and Hi-NicalonTM fibers [17]. Pure SiC surface chemistry and rough fiber surface may have significant affections on the interfacial bonding and fiber sliding, and therefore, on the mechanical properties of the composites. This is also an important issue to be understood. In this study, several CVI-SiC/SiC composites reinforced with the Tyranno-SA fiber were fabricated with various PyC and SiC/PyC interlayers. The flexural properties and fracture behaviors were studied using three-point bending tests. The main objectives are to get an understanding of the mechanical properties of the various composites. The simple bending tests were used to study the effects of the rough fiber surface characteristics and the various interlayers. 2. Experimental 2.1. Composite processing Fibrous preforms were fabricated by stacking 11 layers of 2D plain-woven Tyranno-SA fiber cloths in 0/908. The preforms were compressed to keep a fiber volume fraction of /43% using a set of graphite fixtures. The normal size of the preforms was 40 mm in diameter and 2.0 mm in thickness. The preforms were pre-coated with single PyC layers of different thickness or SiC/PyC bi-layers using an isothermal chemical vapor infiltration (CVI) process through the thermal decomposition of methane and CH3SiCl3 (MTS), respectively. MTS was carried by hydrogen. The pre-coated preforms were finally densified with SiC matrix by an isothermal-forced flow CVI process at 1273 K and 14.7 kPa. Detailed fabrication process can be found elsewhere [18]. The densification process generally required /15 h. 2.2. Mechanical tests Three-point bending tests (with a support span of 18 mm) were performed at room temperature. Bending specimens were cut parallel to one of the fiber bundle directions of the fabric cloth using a diamond cutter and both the tensile and compression surfaces were carefully ground using diamond slurry to eliminate the effects of surface CVD-SiC layers which were formed at the end of the CVI process [18]. Three tests were conducted for each composite. The dimension of the specimen is 30L/ 4.0W//1.5T mm3 . The crosshead speed was 0.0083 mm s1 . The load/displacement data was recorded. 2.3. Microstructural characterization The morphologies, thickness, and uniformities of the interlayers were examined with scanning electron microscope (SEM) using JOEL JIM-6700F. The interlayer thickness was measured with an estimated resolution of /10 nm. The fracture surfaces were observed with SEM interfacial debonding and fiber pullouts. 3. Results 3.1. Composites and interlayer structures Totally six composites were fabricated. The interlayer structures and thickness and composite densities/porosities are listed in Table 2. The fiber volume fractions are 42/44%. No intentional interlayer was applied in composite T-NL. Single PyC interlayer was deposited in composites T-C50, T-C100, and T-C200. In order to investigate the effect of SiC layer in Tyranno-SA/SiC composites, two composites, T-SiC/C80 and T-SiC/ C150, were deposited with SiC/PyC bi-interlayers. The interlayer thickness and space homogeneity are as in Table 2. Fig. 1 shows the SEM images of the interlayers of composites T-C200 and T-SiC/C150. A uniform 200 nm-thick PyC interlayer was deposited in composite TC200 while in composite T-SiC/C150, a SiC layer of 150 nm thickness was deposited on the fibers prior to the deposition of 150 nm-thick PyC layer. SEM interlayer Table 1 Several selected properties of the Tyranno-SA fiber C/Si Diameter (mm) Structure Tensile strength (MPa) Tensile modulus (Gpa) Elongation (%) Thermal conductivity (W m1 K1 ) 1.08 7.5 Crystal 2510 409 0.7 64.6 (RT) The fiber and properties were provided by the Ube Industry Ltd. Japan. W. Yang et al. / Materials Science and Engineering A345 (2003) 28 /35 29
w. Yang et al. Materials Science and Engineering 4345(2003)28-35 Table 2 The fabricated Tyranno-SA/SiC composites pecten L D Interlayer structure and thickness(nm) Fiber volume fraction (% Density(Mg m) Porosity (% T-NL 2.78(0.01) T-C50 F/CSO(/M TC200 423343 2.61(0.03) F/SiCSO23)/CSOIS)/M 2.58(0.04) 16.4 SiC/C150 F/SC15025C15028M 2.370.05) -a Included in the parenthesis are the standard deviations. The standard deviations of the thickness of interlayers were obtained using a same method as described in Ref. [18] examinations of other interlayered composites also bonding and crack deflection had occurred at SiC/PyC confirmed successful deposition of the interlayers with interface or at the matrix/PyC interface for the former quite well through-thickness uniformity, as indicated while at the fiber surface for the latter before the failure Table 2. Composites T-C50 and T-SiC/C150 showed of the composites. Therefore, it is likely that a first Sic lower densities compared with the others, resulting in layer on the fiber is able to control the interfacial the porosities over 20% debonding to occur within the SiC/PyC interlayers or at the matrix/PyC interface, rather than at the fib 3.2. Fracture behaviors surface. For composites T-C50 and T-SiC/C150, sig- nificant inter-fabric layers delaminating occurred during the bending tests, as indicated by the large inter-fabric Transverse cracks initiated at the tensile surfaces of e s ens from composite T-NL and propagated layers pores in Fig 4(composite T-C50) almost vertically to the compression surfaces. SEM Typical load-displacement curves of the composites fracture surface examination revealed a smooth and are shown in Fig. 5. Composite T-NL exhibited low load flat fracture surface. with no evidence of fiber/matrix fiber/matrix maximum and displayed brittle failure mode, with no signs of toughening. This is in consistent with the flat debonding and fiber pullout. Multiple deflection or fracture surface. Improved toughness was indicated in transverse cracks occurred in the PyC or SiC/PyC layered composites. Fig. 2 shows the transverse crack the curves for the interlayered composites. Both the load propagation behaviors in composite T-C100. Cracks maximums and displacement at load maximums were nitiated at both tensile and compression surfaces. The increased. Initially, the load increased linearly with the main crack initiated at tensile surface and was multi- increasing of displacement, reflecting the elastic re- deflected by the 0/90 fiber bundles. Fiber pullouts were ponse of the composites. Deviation of the curves to observed at the fracture surface as shown in Fig 3(a and the linearity occurred at certain loads, followed by a b). Similar fracture behaviors were observed with nonlinear domain of deformation until the load max composites T-C200 and T-SiC/C80. Fig. 3(c and imum, due mainly to the matrix cracking, interfacial show the fracture surface and fiber pullouts of compo- debonding and fibe ng, and individual fiber fail site T-SiC/C80. It is seen from Fig 3(b and d) that the ures. Finally, the composites failed owing to the failures interlayer(s)was pulled out together with the debonded of the fibers. Some differences of the load maximums are fibers for composite T-SiC/C80, while no PyC layer demonstrated by the various composites Composite T flake was found attaching on the pulled out fibers for C100 yielded the highest load maximum at a displace- composite T-C100. This indicates that interfacial de- ment up to -0.18 mm (a)T-C200 Matrix Fig. 1. SEM images of the interlayers in composites: (a)T-C200: and (b)T-SiC/C150
examinations of other interlayered composites also confirmed successful deposition of the interlayers with quite well through-thickness uniformity, as indicated in Table 2. Composites T-C50 and T-SiC/C150 showed lower densities compared with the others, resulting in the porosities over 20%. 3.2. Fracture behaviors Transverse cracks initiated at the tensile surfaces of the specimens from composite T-NL and propagated almost vertically to the compression surfaces. SEM fracture surface examination revealed a smooth and flat fracture surface, with no evidence of fiber/matrix debonding and fiber pullout. Multiple deflection of transverse cracks occurred in the PyC or SiC/PyC layered composites. Fig. 2 shows the transverse crack propagation behaviors in composite T-C100. Cracks initiated at both tensile and compression surfaces. The main crack initiated at tensile surface and was multideflected by the 0/908 fiber bundles. Fiber pullouts were observed at the fracture surface as shown in Fig. 3(a and b). Similar fracture behaviors were observed with composites T-C200 and T-SiC/C80. Fig. 3(c and d) show the fracture surface and fiber pullouts of composite T-SiC/C80. It is seen from Fig. 3(b and d) that the interlayer(s) was pulled out together with the debonded fibers for composite T-SiC/C80, while no PyC layer flake was found attaching on the pulled out fibers for composite T-C100. This indicates that interfacial debonding and crack deflection had occurred at SiC/PyC interface or at the matrix/PyC interface for the former while at the fiber surface for the latter before the failure of the composites. Therefore, it is likely that a first SiC layer on the fiber is able to control the interfacial debonding to occur within the SiC/PyC interlayers or at the matrix/PyC interface, rather than at the fiber surface. For composites T-C50 and T-SiC/C150, significant inter-fabric layers delaminating occurred during the bending tests, as indicated by the large inter-fabric layers pores in Fig. 4 (composite T-C50). Typical load/displacement curves of the composites are shown in Fig. 5. Composite T-NL exhibited low load maximum and displayed brittle failure mode, with no signs of toughening. This is in consistent with the flat fracture surface. Improved toughness was indicated in the curves for the interlayered composites. Both the load maximums and displacement at load maximums were increased. Initially, the load increased linearly with the increasing of displacement, reflecting the elastic response of the composites. Deviation of the curves to the linearity occurred at certain loads, followed by a nonlinear domain of deformation until the load maximum, due mainly to the matrix cracking, interfacial debonding and fiber sliding, and individual fiber failures. Finally, the composites failed owing to the failures of the fibers. Some differences of the load maximums are demonstrated by the various composites. Composite TC100 yielded the highest load maximum at a displacement up to /0.18 mm. Table 2 The fabricated Tyranno-SA/SiC composites Specimen I.D. Interlayer structure and thickness (nm)a Fiber volume fraction (%) Density (Mg m3 ) a Porosity (%) T-NL F/M 44 2.78(0.01) 10.6 T-C50 F/C50(8)/M 42 2.41(0.03) 20.4 T-C100 F/C100(19)/M 43 2.63(0.04) 15.1 T-C200 F/C200(34)/M 43 2.61(0.03) 14.6 T-SiC/C80 F/SiC150(23)/C80(15)/M 44 2.58(0.04) 16.4 T-SiC/C150 F/SiC150(25)/C150(28)/M 43 2.37(0.05) 21.8 a Included in the parenthesis are the standard deviations. The standard deviations of the thickness of interlayers were obtained using a same method as described in Ref. [18]. Fig. 1. SEM images of the interlayers in composites: (a) T-C200; and (b) T-SiC/C150. 30 W. Yang et al. / Materials Science and Engineering A345 (2003) 28 /35
w. Yang et al I Materials Science and Engineering 4345(2003)28-35 50 pm Interlay 50g C ion side 5 Fig 3. SEM images of fracture surfaces and fiber pullouts of composites T-C100(a and b)and T-SiC/C80(c and d)
Fig. 2. SEM images showing transverse crack propagations through the cross section of composite T-C100. Fig. 3. SEM images of fracture surfaces and fiber pullouts of composites T-C100 (a and b) and T-SiC/C80 (c and d). W. Yang et al. / Materials Science and Engineering A345 (2003) 28 /35 31
w. Yang et al. Materials Science and Engineering 4345(2003)28-35 Inter-fabric layers delaminating the same CVI process with similar fiber volume fraction, T-C50 composite density and interlayers are considered to be the main reasons causing the differences of the e lexar properties. The highest PLS and UFS, 426 and 606 MPa, respectively, were obtained comi C100, which was coated with 100 nm PyC interlayer This composite possesses the highest density of the interlayered composit I mm Tensile surface 4. Discussion Fig. 4. SEM image of cross section of osite T-C50, showing inter-fabric layer debonding. 4. 1. Effects of the fiber The Tyranno-SA/SiC composites exhibited high PLS but relatively small displacement at load maximums Fig. 5 and Table 3). High PLS was a direct benefit from the advanced Tyranno-SA fiber. The tensile modulus the Tyranno-SA fiber is 410 Gpa (Table 1),versus ----T-NL lower value of 200 and x 280 GPa for the nicalon and Hi-Nicalon fibers, respectively [18]. Stronger fiber causes less load share by the matrix in a SiC/SiC composite and therefore, results in a larger PLS [13] Table 3 shows quite large ISS, from 195 to 414 MPa, for the interlayered composites. Large IsS generally makes 0.000050.100.150.20 250300.35 it more difficult for the interfacial debonding and fiber D isplacement/mm liding to occur. This will reduce the abilities for the interfaces and the fibers to deviate and bridge the Fig. 5. Representative load-displacement curves of the Tyranno-SA/ already initiated matrix cracks upon bending, and Sic composites. hence, cause quicker failure of the composites at relatively small displacement. Large ISS is a result of 3.3. Flexural properties he pure SiC surface chemistry and rough surface of the Tyranno-SA fiber [17]. Therefore, weaker interfacial The flexural properties that were extracted from the bonding(thicker PyC interlayer or more suitable alter- load-displacement curves according to ASTMC 1341- nating compliant interlayers)may be necessary for the 97[19] are summarized in Table 3. The proportional Tyranno-SA/SiC composites to achieve further im- limit stress(PLS) was the stress corresponding to 0.019 proved fracture tolerance offset strain. The ultimate flexural strength(UFS)was calculated by the simple elastic beam theory [20]. The 4.2. Effects of density and interlayer interfacial shear strengths(ISS)were obtained from Ref Table 3 exhibits high PLS but relatively small structures dependence of the flexuralper and interlayer displacements at load maximums for the Tyranno-SA/ highest PLS and UFS were obtained with composite T Sic composites. Since the composites were fabricated by C100, which had the highest densities of 2.63 Mg m ecimen I.D. Flexural modulus(GPa) PLS (MPa) UFS(MPa) Displacement at load maximum(mm) ISS (MPa) T.NL 0.07(0.02) >633 T-C50 135(13) 0.13(0.05) 1(140) TC100 57(1 426(2) 0.18(00 42(97) TC200 33918) 549(58) 0.15(0.01) 195(51 T-SiC/C80 148(2) 339(45) 4l4(117) T-SiC/C150 0.14(0.02)
3.3. Flexural properties The flexural properties that were extracted from the load/displacement curves according to ASTM C 1341- 97 [19] are summarized in Table 3. The proportional limit stress (PLS) was the stress corresponding to 0.01% offset strain. The ultimate flexural strength (UFS) was calculated by the simple elastic beam theory [20]. The interfacial shear strengths (ISS) were obtained from Ref. [17]. Table 3 exhibits high PLS but relatively small displacements at load maximums for the Tyranno-SA/ SiC composites. Since the composites were fabricated by the same CVI process with similar fiber volume fraction, composite density and interlayers are considered to be the main reasons causing the differences of the flexural properties. The highest PLS and UFS, 426 and 606 MPa, respectively, were obtained with composite TC100, which was coated with 100 nm PyC interlayer. This composite possesses the highest density of the interlayered composites. 4. Discussion 4.1. Effects of the fiber The Tyranno-SA/SiC composites exhibited high PLS but relatively small displacement at load maximums (Fig. 5 and Table 3). High PLS was a direct benefit from the advanced Tyranno-SA fiber. The tensile modulus of the Tyranno-SA fiber is /410 Gpa (Table 1), versus lower value of /200 and /280 GPa for the Nicalon and Hi-Nicalon fibers, respectively [18]. Stronger fiber causes less load share by the matrix in a SiC/SiC composite and therefore, results in a larger PLS [13]. Table 3 shows quite large ISS, from 195 to 414 MPa, for the interlayered composites. Large ISS generally makes it more difficult for the interfacial debonding and fiber sliding to occur. This will reduce the abilities for the interfaces and the fibers to deviate and bridge the already initiated matrix cracks upon bending, and hence, cause quicker failure of the composites at relatively small displacement. Large ISS is a result of the pure SiC surface chemistry and rough surface of the Tyranno-SA fiber [17]. Therefore, weaker interfacial bonding (thicker PyC interlayer or more suitable alternating compliant interlayers) may be necessary for the Tyranno-SA/SiC composites to achieve further improved fracture tolerance. 4.2. Effects of density and interlayer Table 3 shows a composite densities and interlayer structures dependence of the flexural performance. The highest PLS and UFS were obtained with composite TC100, which had the highest densities of 2.63 Mg m3 Fig. 4. SEM image of cross section of composite T-C50, showing inter-fabric layer debonding. Fig. 5. Representative load-displacement curves of the Tyranno-SA/ SiC composites. Table 3 Flexural properties and ISS of the Tyranno-SA/SiC composites Specimen I.D. Flexural modulus (GPa) PLS (MPa) UFS (MPa) Displacement at load maximum (mm) ISS (MPa) T-NL 156(13) 275(48) 281(60) 0.07(0.02) 633 T-C50 135(13) 257(5) 410(92) 0.13(0.05) 331(140) T-C100 157(11) 426(32) 606(28) 0.18(0.02) 342(97) T-C200 140(10) 339(18) 549(58) 0.15(0.01) 195(51) T-SiC/C80 148(2) 339(45) 440(1) 0.12(0.03) 414(117) T-SiC/C150 141(2) 370(43) 495(85) 0.14(0.02) 284(96) 32 W. Yang et al. / Materials Science and Engineering A345 (2003) 28 /35
w. Yang et al. Materials Science and Engineering 4345(2003)28-35 among the interlayered composites. Composite T-C50 yielded much lower PLS and UFS, 257 and 410 MPa, respectively, which are about 170 and 200 MPa less than those of T-C100. The density of this composite is 2.41 Mg m. The low density resulted from insufficient CVI-matrix densification and causes a weak inter-fabl layers bonding. Therefore, the inter-laminar debonding occurred(Fig. 4) due to the inter-fabric layers shear stress upon bending load, and caused the shear failure behavior of the composite with low ultimate strengt Similar situation occurred with composite T-Sic/C150 1014· Tyranno-SA/siC▲Hi- Nicalon/siC of which the density is 2.37 Mg m. The flexural modulus also demonstrated density-dependence as shown in Fig. 6. High flexural modulus was obtained Py C layer thickness/nm by the composites with dense matrix (T-NL and T Fig. 7. PyC layer thickness dependence of the UFS 9. As mentioned before, SiC/SiC composites are very or thicker(T-C200) PyC layer. A 150 nm SiC layer was attractive candidates for applications under severe environments. The current key issue is the further deposited on the fibers prior to the deposition of the 150 improvement of the mechanical properties. Develop nm PyC layer in this composite. Another bi-layered ment of appropriate interlayer(s)with desired materials, tendency in Fig. 7. However, the effects of the Sic layer interlayer structures and layer thickness is a main on the composite strength remains unclear because that solution. For the present Tyranno-SA/SIC composites, although Fig. 7 relates the UFS to the PyC layer it was found [17] that the interfacial properties were thickness. it is, in fact, a blend effect of the interlayer, significantly affected by the thickness of the PyC layers density, and fracture mechanism, etc. The low density Fig. 7 relates the UFS of the Tyranno-SA/SiC compo- and inter-laminar shear failure of T-SiC/C150 are sites to the thickness of the PyC layers. The results of 2D considered to be the main reasons causing the low plain-woven CVI-Hi-Nicalon/SiC composites [ll] are ultimate strength. Further studies are necessary to get a also presented for comparison. The UFS of the Tyr- clear understanding of the effects the SiC layer. Any anno-SA/SiC composites increases from 280 to 600 how, Fig. 7 shows a clear trend of the effects of the PyC MPa with increasing the Pyc layer to 100 nm, then layer thickness on the strength of the composites keeps at a similar level up to 200 nm considering the li-Nicalon/SiC composites, the Tyranno-SA/SiC com- 4.3. Effects of lss posites exhibit a slightly less sensitivity to the Pyc layer thickness For the Hi-Nicalon/SiC composites, the high The PLs of Sic/SiC composites is significantly est strength is achieved by the composite with 150 nm affected by the ISS. Inghels and Lamon [21] developed gle PyC layer while the Tyranno-SA/SiC composite a theoretical mode to predict the strength of unidirec- with same thickness of PyC (T-Sic/C150) layer yielded a tional Sic/Sic composites upon flexural loading. The lower strength than those with either thinner(T-C100) PLS was given by 2,mElfa Ec+ Er eVE 4E BEC erve where ,m is the surface energy of CVD-SiC, which was 25 Jm--[21]. T; is the interfacial shear Em, Ef, Vm, and Ve are the Youngs modulus and volume fractions of the matrix and the fiber. Ec is the composite modulus determined from the law of mixture [21] re is the diameter of the fiber The mode was developed based on a ID model cell Composite density/Mg/m3 while the reinforcements in the present composites are 2D plain-woven fabrics. Here for simplicity [22, 23], the Fig.6. Density dependence of the flexural modulus of the composites. 0 fiber bundles are regarded as the reinforcement while
among the interlayered composites. Composite T-C50 yielded much lower PLS and UFS, 257 and 410 MPa, respectively, which are about 170 and 200 MPa less than those of T-C100. The density of this composite is 2.41 Mg m3 . The low density resulted from insufficient CVI-matrix densification and causes a weak inter-fabric layers bonding. Therefore, the inter-laminar debonding occurred (Fig. 4) due to the inter-fabric layers shear stress upon bending load, and caused the shear failure behavior of the composite with low ultimate strength. Similar situation occurred with composite T-SiC/C150 of which the density is 2.37 Mg m3 . The flexural modulus also demonstrated density-dependence as shown in Fig. 6. High flexural modulus was obtained by the composites with dense matrix (T-NL and TC100). As mentioned before, SiC/SiC composites are very attractive candidates for applications under severe environments. The current key issue is the further improvement of the mechanical properties. Development of appropriate interlayer(s) with desired materials, interlayer structures and layer thickness is a main solution. For the present Tyranno-SA/SiC composites, it was found [17] that the interfacial properties were significantly affected by the thickness of the PyC layers. Fig. 7 relates the UFS of the Tyranno-SA/SiC composites to the thickness of the PyC layers. The results of 2D plain-woven CVI-Hi-Nicalon/SiC composites [11] are also presented for comparison. The UFS of the Tyranno-SA/SiC composites increases from /280 to /600 MPa with increasing the PyC layer to 100 nm, then keeps at a similar level up to 200 nm considering the relatively large error bars. Compared with that of the Hi-Nicalon/SiC composites, the Tyranno-SA/SiC composites exhibit a slightly less sensitivity to the PyC layer thickness. For the Hi-Nicalon/SiC composites, the highest strength is achieved by the composite with 150 nm single PyC layer while the Tyranno-SA/SiC composite with same thickness of PyC (T-SiC/C150) layer yielded a lower strength than those with either thinner (T-C100) or thicker (T-C200) PyC layer. A 150 nm SiC layer was deposited on the fibers prior to the deposition of the 150 nm PyC layer in this composite. Another bi-layered composite, T-SiC/C80, also yielded a strength below the tendency in Fig. 7. However, the effects of the SiC layer on the composite strength remains unclear because that although Fig. 7 relates the UFS to the PyC layer thickness, it is, in fact, a blend effect of the interlayer, density, and fracture mechanism, etc. The low density and inter-laminar shear failure of T-SiC/C150 are considered to be the main reasons causing the low ultimate strength. Further studies are necessary to get a clear understanding of the effects the SiC layer. Anyhow, Fig. 7 shows a clear trend of the effects of the PyC layer thickness on the strength of the composites. 4.3. Effects of ISS The PLS of SiC/SiC composites is significantly affected by the ISS. Inghels and Lamon [21] developed a theoretical mode to predict the strength of unidirectional SiC/SiC composites upon flexural loading. The PLS was given by, sPLSEc 12gmEfV2 f ti EcE2 mVmrf 1=3Ec EfVf 2EfVf 4EfVf 3Ec EfVf 1=3 ; (1) where gm is the surface energy of CVD-SiC, which was given as 25 J m2 [21]. ti is the interfacial shear strength. Em, Ef, Vm, and Vf are the Young’s modulus and volume fractions of the matrix and the fiber. Ec is the composite modulus determined from the law of mixture [21]. rf is the diameter of the fiber. The mode was developed based on a 1D model cell while the reinforcements in the present composites are 2D plain-woven fabrics. Here for simplicity [22,23], the Fig. 6. Density dependence of the flexural modulus of the composites. 08 fiber bundles are regarded as the reinforcement while Fig. 7. PyC layer thickness dependence of the UFS. W. Yang et al. / Materials Science and Engineering A345 (2003) 28 /35 33
w. Yang et al. Materials Science and Engineering 4345(2003)28-35 the 90 bundles are regarded as matrix. Thus the volume that further marked improvement of the strength of the fraction of the 'matrixis the volume fractions of the real Tyranno-SA/SiC composites is possible through the CVI-SiC matrix and the 90 direction fibers given br optimization of the CVi process for dense and pure V"=1-V-V crystalline matrix deposition and appropriate woven type of the reinforcements where Vp is the volume fraction of pores in the Another main assumption of the model [21] is that composites,as given in Table 2. Vf is the volume multiple matrix cracking occur with constant distance fraction of the 0o fiber bundles(half of ve between two consecutive cracks when the matrix crack Fig. 8 shows the calculated PLS, as well as the initiated, followed by interfacial debonding and fiber experiment results, via the ISS. Compared with the sliding without further damage of the matrix until the experiment results, the mode over estimated the PLs failure of the fibers. This is not always true in practical with a simple increase with the ISS. As, mentioned materials, especially for composites with strong inter- before, the mode was developed based on unidirectional facial bonding, because when the interfacial bonding is composite cell, assuming defect-free fibers and matrix. strong, interfacial debonding or matrix crack being In above calculation, the 2d plain-woven fabrics were deflected at the interfaces or bridged by the fibers are simply regarded as unidirectional reinforcement fibers difficult to occur. Instead, significant stress concentra- (O bundles)andmatrix(90 bundles). Clearly, crack is tion will take place at the crack tip. Extremely, when the easier to initiate and propagate in the 90 bundle fiber/matrix bonding is so strong that before interfacial matrix. Therefore, the load-bearing ability of the debonding and matrix crack deflections occur, a single matrix under such simple treatment was over esti- main matrix crack(s)forms and propagates through the thickness of the composite, resulting a brittle failure CVI-matrix due to the mechanical and thermal residual (without fiber pullout) behavior with lower PLS, as stresses resulting from the fabrication process. In addi- exhibited by composites T-NL. This is likely to be the are, more or less, curved. This phenomenon, wich practical composites, the reinforcement fibers reason that the mode predicted a simple increase with the iss while the experimental results displayed a not considered in the mode, is enhanced in a 2D plain contrary tendency when the ISs is over certain value woven or more complex woven fabric preforms, result- This indicates that there is a threshold of ISS beyond ing in a decrease of the load-sharing ability of the fibers. which the mode becomes invalid. For the present These factors are believed to be the main reasons Tyranno-SA/SIC composites, this threshold seems to causing the over estimation of the Pls by the mode be 340 MPa at which the highest Pls was achieved Assuming that the effects of above factors on the PLs among all the composites can be expressed by simply applying a constant coeffi- cient,K, in Eq (1). Then a value of 0.373 for the K ielded a much better agreement between the mode 5. Conclusion estimation and experimental results, as shown in Fig 8 (modified estimations were slightly shifted towards Several Tyranno-SA fiber reinforced SiC/SiC compo- smaller ISS for clarity). The original mode(K-1)gives sites with various PyC and SiC/PyC interlayers we a simple theoretical prediction of the strength of a fat fabricated and the mechanical properties and fracture composite from the properties of the constitutive(the behaviors were studied using three-point bending tests fiber, the matrix, etc. ) By fitting the model calculation The conclusions are into the experiment results(K=0.373), it is indicative The isothermal CVI process was confirmed to be able to deposit PyC and Sic/C interlayers on small diameter Tyranno-SA fibers in SiC/SiC composites with sufficient Calculation (K= 1) thickness and uniformity control alculation (K=0.373) The flexural strengths of the composites demonstrated close dependence on the PyC interlayer. The ultimate strength increased with the increasing of the PyC layer thickness to 100 nm, and then kept at a similar level till t300 200nm. The materials demonstrated high PLSs but relatively small displacements at load maximums. High PLS is a Interfacial shear strength /MPa direct benefit from the large tensile modulus of the fiber while relatively small displacements at load maximums Fig. 8. Calculated and experimental PLS via ISS of the composites are likely due mainly to the strong interfacial bonding
the 908 bundles are regarded as matrix. Thus the volume fraction of the ‘matrix’ is the volume fractions of the real CVI-SiC matrix and the 908 direction fibers given by: V? m1VpV? f; where Vp is the volume fraction of pores in the composites, as given in Table 2. V? f is the volume fraction of the 08 fiber bundles (half of Vf). Fig. 8 shows the calculated PLS, as well as the experiment results, via the ISS. Compared with the experiment results, the mode over estimated the PLS with a simple increase with the ISS. As, mentioned before, the mode was developed based on unidirectional composite cell, assuming defect-free fibers and matrix. In above calculation, the 2D plain-woven fabrics were simply regarded as unidirectional reinforcement fibers (08 bundles) and ‘matrix’ (908 bundles). Clearly, crack is easier to initiate and propagate in the 908 bundle ‘matrix’. Therefore, the load-bearing ability of the ‘matrix’ under such simple treatment was over estimated, let along the possible pre-exist cracks in the real CVI-matrix due to the mechanical and thermal residual stresses resulting from the fabrication process. In addition, in practical composites, the reinforcement fibers are, more or less, curved. This phenomenon, which is not considered in the mode, is enhanced in a 2D plainwoven or more complex woven fabric preforms, resulting in a decrease of the load-sharing ability of the fibers. These factors are believed to be the main reasons causing the over estimation of the PLS by the mode. Assuming that the effects of above factors on the PLS can be expressed by simply applying a constant coefficient, K, in Eq. (1). Then a value of 0.373 for the K yielded a much better agreement between the model estimation and experimental results, as shown in Fig. 8 (modified estimations were slightly shifted towards smaller ISS for clarity). The original mode (K/1) gives a simple theoretical prediction of the strength of a composite from the properties of the constitutive (the fiber, the matrix, etc.). By fitting the model calculation into the experiment results (K/0.373), it is indicative that further marked improvement of the strength of the Tyranno-SA/SiC composites is possible through the optimization of the CVI process for dense and pure crystalline matrix deposition and appropriate woven type of the reinforcements. Another main assumption of the model [21] is that multiple matrix cracking occur with constant distance between two consecutive cracks when the matrix crack initiated, followed by interfacial debonding and fiber sliding without further damage of the matrix until the failure of the fibers. This is not always true in practical materials, especially for composites with strong interfacial bonding, because when the interfacial bonding is strong, interfacial debonding or matrix crack being deflected at the interfaces or bridged by the fibers are difficult to occur. Instead, significant stress concentration will take place at the crack tip. Extremely, when the fiber/matrix bonding is so strong that before interfacial debonding and matrix crack deflections occur, a single main matrix crack(s) forms and propagates through the thickness of the composite, resulting a brittle failure (without fiber pullout) behavior with lower PLS, as exhibited by composites T-NL. This is likely to be the reason that the mode predicted a simple increase with the ISS while the experimental results displayed a contrary tendency when the ISS is over certain value. This indicates that there is a threshold of ISS beyond which the mode becomes invalid. For the present Tyranno-SA/SiC composites, this threshold seems to be /340 MPa at which the highest PLS was achieved among all the composites. 5. Conclusion Several Tyranno-SA fiber reinforced SiC/SiC composites with various PyC and SiC/PyC interlayers were fabricated and the mechanical properties and fracture behaviors were studied using three-point bending tests. The conclusions are: The isothermal CVI process was confirmed to be able to deposit PyC and SiC/C interlayers on small diameter Tyranno-SA fibers in SiC/SiC composites with sufficient thickness and uniformity control. The flexural strengths of the composites demonstrated a close dependence on the PyC interlayer. The ultimate strength increased with the increasing of the PyC layer thickness to 100 nm, and then kept at a similar level till 200 nm. The materials demonstrated high PLSs but relatively small displacements at load maximums. High PLS is a direct benefit from the large tensile modulus of the fiber while relatively small displacements at load maximums Fig. 8. Calculated and experimental PLS via ISS of the composites. are likely due mainly to the strong interfacial bonding. 34 W. Yang et al. / Materials Science and Engineering A345 (2003) 28 /35
w. Yang et al Materials Science and Engineering 4345(2003)28-35 The influences of the interfacial shear stress on the 8T. K. Jacobsen, P. Brondsted, J. Am. Ceram Soc. 84(5)(2001) Pls were discussed based on both a theoretical mode calculations and the experiment results. The highest PLS 9]RJ Kerans, R.S. Hay, N.J. Pagano, T.A. Parthasarathy, Ceram. as achieved by the composites with interfacial shear (10R. Naslain, in: A.G.Evans, R. Naslain(Eds),Ceramic Transac stress of x 340 MPa. Attractive promising was exhib- ons High-Temperature Ceramic-Matrix Composites Il: Manu ited on further improvement of the strength of the facturing and Materials Development, vol 58, America Ceramic composites through further improvement of the inter Society, Westerville, OH, 1995, P. 23 facial properties and matrix densification process. [1 w. Yang, H. Araki, T Noda, J.Y. Park, Y Katoh, T. Hinoki, J. Yu, A. Kohyama, submitted to J Am Ceram Soc C. Droillard, J. Lamon, J. Am. Ceram Soc. 79(4)(1996)849. S. Bertrand. P. Forio. R. Pailler J. Lamon. J. Am. Ceram. Soc. 82 Acknowledgements (9)(19992465 [14 T M. Besmann, E.R. Kupp, E. Lara-Curzio, K L More, J. Am Ceram.Soc.83(12)(2000)3014 This work was supported by the CREST, Japan [15] T Ishikawa, Y. Kohtoku, K. Kumagawa, T. Yamamura, T Science and Technology Corporation asawa, Nature391(6669)(1998)773. [16 G.E. Youngblood, R H. Jones, A Kohyama, LL. Snead, J Nucl. Mater.258-263(1998)1551 [17 w. Yang, A. Kohyama, T Noda, Y. Katoh, H. Araki, J. Yu, J References Nucl. Mater, in press [18 w. Yang, Development of CvI Process and Property Evaluation [A.Kohyama, M. Seki, K. Abe, T. Muroga, H. Matsui,S of CVI-Sic/SiC Composites, Doctoral thesis, Institute of Ad- Jitsukawa. S. Matsuda, J. Nucl. Mater. 283-287(2000)20 []AG. Evans, J Am Ceram Soc. 73(2)(1990) [19 ASTMC 1341-97, Standard Test Method for Flexural Properties 3]D. Brewer, Mater. Sci. Eng. A261(1999)284. of Continuous Fiber-reinforced Advanced Ceramic Composites [4 T Noda, H. Araki, F. Abe, M. Okada, J Nucl. Mater. 191-194 2000.pp.509-526 (1992)539 [ E. Inghels, J. Lamon, J. Mater. Sci. 26(1991)5403. [5 LL. Snead, O.J. Schwarz, J Nucl. Mater. 219(1995)3. 21 E. Inghels, J. Lamon, J Mater. Sci. 26(1991)5411 [6R H. Jones, C H. Henager, Jr, G.w. Hollenberg. J Nucl. Mater. [22A.G. Evans, J.-M. Domergue, E. Vagaggini, J. Am. Ceram Soc. 191-194(1992)75 77(6)01994)1425 [7 T.M. Besmann, D P. Stinton, E.R. Kupp, S Shanmugham, P K. (23 J.M. Domergue, F.E. Heredia, A.G. Evans, J. Am. Ceram Soc. Liaw, J. Mater Res. Soc. Symp. Proc. 458(1997)1 79(1)(1996161
The influences of the interfacial shear stress on the PLS were discussed based on both a theoretical mode calculations and the experiment results. The highest PLS was achieved by the composites with interfacial shear stress of /340 MPa. Attractive promising was exhibited on further improvement of the strength of the composites through further improvement of the interfacial properties and matrix densification process. Acknowledgements This work was supported by the CREST, Japan Science and Technology Corporation. References [1] A. Kohyama, M. Seki, K. Abe, T. Muroga, H. Matsui, S. Jitsukawa, S. Matsuda, J. Nucl. Mater. 283/287 (2000) 20. [2] A.G. Evans, J. Am. Ceram. Soc. 73 (2) (1990) 187. [3] D. Brewer, Mater. Sci. Eng. A261 (1999) 284. [4] T. Noda, H. Araki, F. Abe, M. Okada, J. Nucl. Mater. 191/194 (1992) 539. [5] L.L. Snead, O.J. Schwarz, J. Nucl. Mater. 219 (1995) 3. [6] R.H. Jones, C.H. Henager, Jr., G.W. Hollenberg, J. Nucl. Mater. 191/194 (1992) 75. [7] T.M. Besmann, D.P. Stinton, E.R. Kupp, S. Shanmugham, P.K. Liaw, J. Mater. Res. Soc. Symp. Proc. 458 (1997) 147. [8] T.K. Jacobsen, P. Brondsted, J. Am. Ceram. Soc. 84 (5) (2001) 1043. [9] R.J. Kerans, R.S. Hay, N.J. Pagano, T.A. Parthasarathy, Ceram. Bull. 68 (2) (1989) 429. [10] R. Naslain, in: A.G. Evans, R. Naslain (Eds.), Ceramic Transactions High-Temperature Ceramic-Matrix Composites II: Manufacturing and Materials Development, vol. 58, America Ceramic Society, Westerville, OH, 1995, p. 23. [11] W. Yang, H. Araki, T. Noda, J.Y. Park, Y. Katoh, T. Hinoki, J. Yu, A. Kohyama, submitted to J. Am. Ceram. Soc. [12] C. Droillard, J. Lamon, J. Am. Ceram. Soc. 79 (4) (1996) 849. [13] S. Bertrand, P. Forio, R. Pailler, J. Lamon, J. Am. Ceram. Soc. 82 (9) (1999) 2465. [14] T.M. Besmann, E.R. Kupp, E. Lara-Curzio, K.L. More, J. Am. Ceram. Soc. 83 (12) (2000) 3014. [15] T. Ishikawa, Y. Kohtoku, K. Kumagawa, T. Yamamura, T. Nagasawa, Nature 391 (6669) (1998) 773. [16] G.E. Youngblood, R.H. Jones, A. Kohyama, L.L. Snead, J. Nucl. Mater. 258/263 (1998) 1551. [17] W. Yang, A. Kohyama, T. Noda, Y. Katoh, H. Araki, J. Yu, J. Nucl. Mater, in press. [18] W. Yang, Development of CVI Process and Property Evaluation of CVI-SiC/SiC Composites, Doctoral thesis, Institute of Advanced Energy, Kyoto University, 2002. [19] ASTMC 1341-97, Standard Test Method for Flexural Properties of Continuous Fiber-reinforced Advanced Ceramic Composites, 2000, pp. 509/526. [20] E. Inghels, J. Lamon, J. Mater. Sci. 26 (1991) 5403. [21] E. Inghels, J. Lamon, J. Mater. Sci. 26 (1991) 5411. [22] A.G. Evans, J.-M. Domergue, E. Vagaggini, J. Am. Ceram. Soc. 77 (6) (1994) 1425. [23] J.M. Domergue, F.E. Heredia, A.G. Evans, J. Am. Ceram. Soc. 79 (11) (1996) 161. W. Yang et al. / Materials Science and Engineering A345 (2003) 28 /35 35