JAm. Ceran. Scc,871794-80302004) ournal Mullite(3Al2O3 2Sio2)-Aluminum Phosphate(AlPO4), Oxide, Fibrous monolithic Composites Dong-Kyu Kim and waltraud m. Riven Department of Materials Science and Engineering, University of Illinois, Urbana, Illinois 61801 Mullite-AlPO, fibrous monolithic composites were fabricated primary phase, separated by cell boundaries of a tailored, second- by a co-extrusion technique using ethylene vinyl acetate(EVA) ary phase. The cells are not ceramic fibers, but rather polycrystal as a binder. Processing routes such as mixing formulation line ceramic domains. The cell boundary phases can be weak extrusion sequence, binder removal cycle, pressing, and sin- interphases, microcracked zones, ductile-phase filaments, or inter- tering procedures are described. An effort to make tougher phases with different physical properties. 8 Coblenz first intro- composites was conducted by modifying the microstructures of duced the concept of fibrous monolithic composites. Some fibre the composites. Different kinds of monolithic composites were mono 37 Sic/BN, SiC/graphite, Si N,/BN, AL2O/graphite, Al 0y/ monolithic systems that have already been studied are the follow- fabricated by changing the number of filaments, and the composition and thickness of interphase layers, and their Al,TiOs, Al,O3/ Al2O3-ZrO2, ZrB,/BN, HfB,/BN, TiB,/BN, microstructural and mechanical properties were character- AL,O3/Ni, AL,O3/Ni-20Cr, and Y-ZrO2/Ni, TiO,/MgSiO3, and ized. To make the interphase more porous and to facilitate AlO3/Al2O3 platelets. debonding and fiber pullout in the composite, graphite was Mullite is an attractive, high-temperature, structural material added as a fugitive"space filler"into the interphase material due to its excellent strength and creep resistance at high and then removed. A fibrous monolithic composite with a ature, good thermal stability, low thermal conductivity, as sintered interphase thickness of 5-10 um and an interphase its chemical inertness. Aluminum phosphate(AlPO4)is composition of 50 vol% graphite and 50 vol% AIPO, had a mally stable(mp -2000oC ), chemically inert, electrically neu- three-point bend strength and a work of fracture of 129+2 tral, and highly covalent. Because of these properties, AlPO4 is an MPa and 0.86+0.05 kJ/m, respectively. This corresponded to attractive candidate material for high-temperature applications. o 42% of the strength but 162% of the work of fracture when Mullite and aluminum phosphate are chemically compatible with compared with the values for a single-phase mullite. Two- each other after sintering at 1600%C for 10 h 4 layer, mixed 50% two-layer: 50% three-layer, and three-layer In this study, oxide fibrous monolithic composites were fabri fibrous monoliths were fabricated and their microstructural cated by a co-extrusion technique, using mullite as a matrix and and mechanical properties were studied. The difference in the minum phosphate as an interphase materi sintering behaviors of the two-layer and three-layer compos- o,. A study to optimize the microstructure of the two-layer, fibrous onolithic composites was conducted by controlling the compo- sition and thickness of the AlPO4 interphase, as well as the number of filaments in a given area. Fibrous monolithic composites with different microstructural architectures of two layers, mixed 50% L. Introduction two layers 50% three layers, and three layers were fabricated, and everal diff aches have been reported for toughening their microstructures and room-temperature mechanical properties chanisms es. These include transformation tough were investigated toughening, microcrack toughen- and fiber-reinforced, ceramic-matrix composites IL. Experimental Procedure (CMCs ites can be simple, powder-processed, laminated composites, or The AlPOa interphase powder was synthesized by the organic al, different kinds of laminated ceramic composites were fabri- drate(Al(NO3)3. 9H,O, Aldrich Chemical, Inc, 98%)and ammo- cated by tape casting, 7-l9 ing, electrophoretic depo nium phosphate diabasic((NHA), HPO,, Fisher Scientific)as the tion,die pressing,rolling, -co-extrusion, and sequential Al and P sources, respectively. A mixture of Elvax 210, 220, and entrifuging. "Functionally graded, laminated ceramics can be 250 ethylene vinyl acetate(EVA)copolymers(Dupont, Wilming made by controlling the composition, layer thickness, and stacking ton, DE) was used as a binder phase. The molecular weight equence, etc. As another type of ceramic material which also increased from Elvax 210. 220. to 250 with values of 44 010 shows"graceful failure, the fibrous monolithic composite was developed.9,26,27 Fibrous monoliths are sintered (or hot-pressed) lithic Table l. Ceramic Powders and polymer mixing ceramics with a distinct fibrous texture consisting Formulations for extrusion' Binder(Elvax Powder 210 220 250 Plasticizer(DP) Lubricant(SA) R. J. Kerans--contributing editor Mullite ner52244.840.8 Outer 50 34 4 7.5 AlPO4 9.624 script No. 186580. Received October 24, 2002, approved December 29 percent. Elvax ethylene Fellow, American Ceramic Society
Mullite (3Al2O32SiO2)–Aluminum Phosphate (AlPO4), Oxide, Fibrous Monolithic Composites Dong-Kyu Kim and Waltraud M. Kriven** Department of Materials Science and Engineering, University of Illinois, Urbana, Illinois 61801 Mullite–AlPO4 fibrous monolithic composites were fabricated by a co-extrusion technique using ethylene vinyl acetate (EVA) as a binder. Processing routes such as mixing formulation, extrusion sequence, binder removal cycle, pressing, and sintering procedures are described. An effort to make tougher composites was conducted by modifying the microstructures of the composites. Different kinds of monolithic composites were fabricated by changing the number of filaments, and the composition and thickness of interphase layers, and their microstructural and mechanical properties were characterized. To make the interphase more porous and to facilitate debonding and fiber pullout in the composite, graphite was added as a fugitive “space filler” into the interphase material and then removed. A fibrous monolithic composite with a sintered interphase thickness of 5–10 m and an interphase composition of 50 vol% graphite and 50 vol% AlPO4 had a three-point bend strength and a work of fracture of 129 2 MPa and 0.86 0.05 kJ/m2 , respectively. This corresponded to 42% of the strength but 162% of the work of fracture when compared with the values for a single-phase mullite. Twolayer, mixed 50% two-layer:50% three-layer, and three-layer fibrous monoliths were fabricated and their microstructural and mechanical properties were studied. The difference in the sintering behaviors of the two-layer and three-layer composites is described. I. Introduction SEVERAL different approaches have been reported for toughening mechanisms of ceramics. These include transformation toughening,1–3 crack deflection toughening,4–6 microcrack toughening,7–10 and fiber-reinforced, ceramic-matrix composites (CMCs).11–15 Alternative tough ceramics to fiber-reinforced ceramic composites can be simple, powder-processed, laminated composites, or fibrous monolithic composites. Following the research of Clegg et al., 16 different kinds of laminated ceramic composites were fabricated by tape casting,17–19 slip casting,20 electrophoretic deposition,21 die pressing,22 rolling,16,23 co-extrusion,24 and sequential centrifuging.25 “Functionally graded,” laminated ceramics can be made by controlling the composition, layer thickness, and stacking sequence, etc.17 As another type of ceramic material which also shows “graceful failure,” the fibrous monolithic composite was developed.19,26,27 Fibrous monoliths are sintered (or hot-pressed) monolithic ceramics with a distinct fibrous texture, consisting of cells of a primary phase, separated by cell boundaries of a tailored, secondary phase. The cells are not ceramic fibers, but rather polycrystalline ceramic domains. The cell boundary phases can be weak interphases, microcracked zones, ductile-phase filaments, or interphases with different physical properties.28 Coblenz26 first introduced the concept of fibrous monolithic composites. Some fibrous monolithic systems that have already been studied are the following:27–37 SiC/BN, SiC/graphite, Si3N4/BN, Al2O3/graphite, Al2O3/ Al2TiO5, Al2O3/Al2O3–ZrO2, ZrB2/BN, HfB2/BN, TiB2/BN, Al2O3/Ni, Al2O3/Ni-20Cr, and Y-ZrO2/Ni, TiO2/MgSiO3, and Al2O3/Al2O3 platelets. Mullite is an attractive, high-temperature, structural material due to its excellent strength and creep resistance at high temperature, good thermal stability, low thermal conductivity, as well as its chemical inertness.38 Aluminum phosphate (AlPO4) is thermally stable (mp 2000°C39), chemically inert, electrically neutral, and highly covalent. Because of these properties, AlPO4 is an attractive candidate material for high-temperature applications.40 Mullite and aluminum phosphate are chemically compatible with each other after sintering at 1600°C for 10 h.41 In this study, oxide fibrous monolithic composites were fabricated by a co-extrusion technique, using mullite as a matrix and aluminum phosphate as an interphase material. A study to optimize the microstructure of the two-layer, fibrous monolithic composites was conducted by controlling the composition and thickness of the AlPO4 interphase, as well as the number of filaments in a given area. Fibrous monolithic composites with different microstructural architectures of two layers, mixed 50% two layers:50% three layers, and three layers were fabricated, and their microstructures and room-temperature mechanical properties were investigated. II. Experimental Procedure Mullite powder (Kyoritsu, KM 101) was the matrix material. The AlPO4 interphase powder was synthesized by the organic, steric entrapment method,42–46 using aluminum nitrate nonahydrate (Al(NO3)39H2O, Aldrich Chemical, Inc., 98%) and ammonium phosphate diabasic ((NH4)2HPO4, Fisher Scientific) as the Al and P sources, respectively. A mixture of Elvax 210, 220, and 250 ethylene vinyl acetate (EVA) copolymers (Dupont, Wilmington, DE) was used as a binder phase. The molecular weight increased from Elvax 210, 220, to 250 with values of 44 010, R. J. Kerans—contributing editor Manuscript No. 186580. Received October 24, 2002; approved December 29, 2003. **Fellow, American Ceramic Society. Table I. Ceramic Powders and Polymer Mixing Formulations for Extrusion† Powder Binder (Elvax)‡ Plasticizer (DP)§ Lubricant (SA)¶ 210 220 250 Mullite Inner 52 2.4 4.8 40.8 – – Outer 50 34 4 2 7.5 2.5 AlPO4 40 9.6 24 14.4 9 3 † All ingredients are in volume percent. ‡ Elvax ethylene vinyl acetate copolymer (Dupont). § DP dioctyl phthalate (99%, Aldrich). ¶ SA stearic acid (95%, Aldrich). J. Am. Ceram. Soc., 87 [5] 794–803 (2004) 794 journal
May 2004 Mullite(3Al,03 2SiO2/ -Aluminum Phosphate (AlPO4, Oxide, Fibrous Monolithic Composites M M 1 mm 1 mm Fig. 1. Optical micrographs of mullite-AlPOA monofilament rods(M: matrix(mullite); 1: interphase(AlPO4)):(a)25 monofilament rod which was first extruded through a die with an orifice diameter of 4.0 mm; (b)150 monofilament rod which was first extruded through a die with an orifice diameter of 1.5 58433, and 1 12 000 g, respectively. The reason for using and outer mullite layer formulations were warm pressed into half mixture of three different eva grades as a binder was to facilitate tubular shapes, with the aid of a thickness-controllable mold the binder removal process. Dioctyl phthalate(Aldrich, 99%)and at a temperature of 150C and a pressure of 34.5 MPa. Th stearic acid(Aldrich, 95%) were used as a plasticizer and a center rod and AlPO interphase, half tubular shapes were lubricant, respectively ranged into a cylindrical mold which was heated to 90C, having Ceramic powder, binder, plasticizer, and lubricant were mixed 23 mm diameter and 150 mm length, and extruded into a die of 1.5 using a computer-controlled, high-shear mixer(Model 2100, C w. 2.0, or 4.0 mm diameter, depending on subsequent applications Brabender, NJ). The mixtures were made following the formula- This is called the"first extrusion" and the extrudate is referred to tions in Table I. In the case of mullite. to facilitate the extrusion of as a two-layer monofilament rod Because the"warm"extrusion the outer r layer so as to have lower viscosity, a lower amount of was made at 90C, the"seamless"half tubular shapes were bonded solids content of 50 vol% and a higher amount of lower-molecular- around the center rod after the first extrusion: 25.93. or 150 weight EVa binder were used; 40 vol% of powder loading was two-layer monofilament rods were arranged into a cylindrical used for mixing the AlPO4 interphase. The Brabender mixing mold which was heated to 90oC and extruded again into a die of 2 perature for the mullite inner rod and AlPO4 interphase layer mm diameter. This is called the second extrusion"and the 150%C. The mixing temperature for the mullite outer layer was extrudate is called a two-layer multifilament rod. To study the C because of its lower viscosity effects of decreasing the thickness of the porous and weak AlPO The inner mullite rods were extruded into cylindrical shapes interphase layer on the strength and work of fracture ing diameters of 11, 13.5, 16, and 20 mm, depending monolithic composites, monofilament rods with green rent subsequent configurations. The extrusion rate was 50 thicknesses of 0.33, 0. 19, and 0.073 mm were extruded n /min. The Brabender-mixed, AlPO4, interphase formulations to make fibrous monolithic composites. For the composi m mm 1 mm (a) 2. Optical micrographs of the mullite-AIPO4 multifilament rod (M: matrix(mullite); 1; interphase(AlPO4)):(a)25 multifilament rod, (b)150
58 433, and 112 000 g, respectively. The reason for using a mixture of three different EVA grades as a binder was to facilitate the binder removal process. Dioctyl phthalate (Aldrich, 99%) and stearic acid (Aldrich, 95%) were used as a plasticizer and a lubricant, respectively. Ceramic powder, binder, plasticizer, and lubricant were mixed using a computer-controlled, high-shear mixer (Model 2100, C. W. Brabender, NJ). The mixtures were made following the formulations in Table I. In the case of mullite, to facilitate the extrusion of the outer layer so as to have lower viscosity, a lower amount of solids content of 50 vol% and a higher amount of lower-molecularweight EVA binder were used; 40 vol% of powder loading was used for mixing the AlPO4 interphase. The Brabender mixing temperature for the mullite inner rod and AlPO4 interphase layer was 150°C. The mixing temperature for the mullite outer layer was 120°C because of its lower viscosity. The inner mullite rods were extruded into cylindrical shapes having diameters of 11, 13.5, 16, and 20 mm, depending on different subsequent configurations. The extrusion rate was 50 mm/min. The Brabender-mixed, AlPO4, interphase formulations and outer mullite layer formulations were warm pressed into half tubular shapes, with the aid of a thickness-controllable mold,47,48 at a temperature of 150°C and a pressure of 34.5 MPa. The center rod and AlPO4 interphase, half tubular shapes were arranged into a cylindrical mold which was heated to 90°C, having 23 mm diameter and 150 mm length, and extruded into a die of 1.5, 2.0, or 4.0 mm diameter, depending on subsequent applications. This is called the “first extrusion” and the extrudate is referred to as a two-layer monofilament rod. Because the “warm” extrusion was made at 90°C, the “seamless” half tubular shapes were bonded around the center rod after the first extrusion; 25, 93, or 150 two-layer monofilament rods were arranged into a cylindrical mold which was heated to 90°C and extruded again into a die of 2 mm diameter. This is called the “second extrusion” and the extrudate is called a two-layer multifilament rod. To study the effects of decreasing the thickness of the porous and weak AlPO4 interphase layer on the strength and work of fracture of fibrous monolithic composites, monofilament rods with green interphase thicknesses of 0.33, 0.19, and 0.073 mm were extruded and used to make fibrous monolithic composites. For the composites with a Fig. 1. Optical micrographs of mullite–AlPO4 monofilament rods (M: matrix (mullite); I: interphase (AlPO4)): (a) 25 monofilament rod which was first extruded through a die with an orifice diameter of 4.0 mm; (b) 150 monofilament rod which was first extruded through a die with an orifice diameter of 1.5 mm. Fig. 2. Optical micrographs of the mullite–AlPO4 multifilament rod (M: matrix (mullite); I: interphase (AlPO4)): (a) 25 multifilament rod, (b) 150 multifilament rod. May 2004 Mullite (3Al2O32SiO2)–Aluminum Phosphate (AlPO4), Oxide, Fibrous Monolithic Composites 795
Journal of the American Ceramic Sociery Kim and Kriven Vol. 87. No. 5 mm Fig. 3. Optical micrographs of green pellets: (a)25 filament green pellet, (b) 150 filament green pellet. constant green interphase thickness of 0.073 mm, 10, 30, and 50 The microstructures of the monofilament rod. the multifila vol% graphite powder was added to the AlPO4 interphase to make ment rod, and the green pellet were examined by optical a more porous and hence weaker interphase layer after sintering. microscopy(Model SMZ-2T, Nikon, Tokyo, Japan), Scanning To make three -layer monofilament rods. the al electron microscopy(Model S-4700, Hitachi, Osaka, Japai quence for the first extrusion was inner mullite rod-AIPO was used to analyze the microstructures of sintered and me- nterphase laver-outer mullite laver: 93 two-laver or three-layer chanically tested, fibrous monolithic composites. Flexural onofilament rods were arranged into a cylindrical mold and then strengths were measured with a screw-driven machine(Model second extruded into a die with an orifice of 2 mm diameter, to 4502, Instron Corp, Canton, MA)in a three-point bend testing numbers of two-layer and three-layer monofilament rods were calculating the area under the lono ample was determined by ake two-layer and three-layer multifilament rods. The same The work of fracture for each sample was determined by randomly mixed nged into the cylindrical mold, and re- The strength and work of fracture data for each composite were extruded into a die of 2.0 mm orifice diameter, to make mixed 50% determined after testing three to five samples. The supportin two-layer: 50% three-layer multifilament rods. The aim of the span was 30 mm, the crosshead speed was 0. I mm/min, and the mixed multifilament rods was to interdisperse, on a small, micro- sample size was 3 mm(h)x 4 mm()X 40 mm(L) structural scale, regions of high strength with regions of high toughness The multifilament rods were cut into 47 mm lengths: 55 II. Results and Discussion ultifilament rods were ged into a molding die and warm ressed at 150%C and 34.5 MPa. The binder was then removed To change the number of filaments in a given area of fibrous from the pressed pellet. The heat treatment cycle for the binder monolithic composite by changing the number of filaments in a removal from the composites without graphite was as follows: heat two-layer multifilament rod, the first extrusions were made from25°to250° C at a ramp rate ofo.05°C/min, heat from250°to through dies having 1.5., and 4.0 mm diameter orifices. figure 450%C at a ramp rate of 0. 1 C/min, heat from 450 to 650 C at a I shows the monofilament rods which were passed through the 4.0 ramp rate of 0.3C/min, maintain at 650C for 2 h, and subse mm(Fig. 1(a))and 1.5 mm(Fig. 1(b) diameter orifices, respec- lently cool down to room temperature with a ramp rate of tively. The porous and weak, AlPOa interphase layer was well then sintered at 1600%C for 10 h. The binder removal cvcle for the monofilament rods which were first extruded through the 4.0, 2.0 composite with graphite in its interphase was the same as that of and 1.5 mm diameter orifices, respectively, were arranged into a the composite without graphite except for maintaining at 550Cfor cylindrical mold of 23 mm diameter and then extruded again 2 h. After removal of binder, the pellet was CIPed at 413.7 MPa through an orifice of 2 mm diameter to make 25, 93, and 150 d then the graphite was removed from the composites using a multifilament rods, respectively. Figure 2 comprises optical m heating cycle as follows: from 25 to 550 C heat at a ramp rate of crographs of the 25(Fig. 2(a))and 150(Fig. 2(b)multifilament 88°C/min, from 550% to80° C heat at a ramp rate of o.o8°C/min rod samples. The population densities of the 25, 93, and 150 d from800°to900° C heat at a ramp rate of o.l°c/ min and ples were 7, 27, and 43 filaments/mm maintain at 900C for 2 h. The binder- and graphite-free pellet was sintered at 1600C for 10 h Table Ill. Effects of Amount of Graphite in the Green Table Il. Effects of Green Interphase Thickness on the AlPO4 Interphase on the Strength and work of Strength and work of fracture of sintered fibrous Fracture of Sintered Fibrous Monolithic Composite Monolithic Composites Amount of graphite in the green Work of interphase(vol% Strength(MPa) fracture(kJ/m") Green interphase thickness Work of fracture Strength(MPa) 162±10 0.26±0.0 10 0.45±0.02 102±100.69± 0.06 0.58±0.05 Green interphase thickness =0.073 mm
constant green interphase thickness of 0.073 mm, 10, 30, and 50 vol% graphite powder was added to the AlPO4 interphase to make a more porous and hence weaker interphase layer after sintering. To make three-layer monofilament rods, the alignment sequence for the first extrusion was inner mullite rod–AlPO4 interphase layer–outer mullite layer; 93 two-layer or three-layer monofilament rods were arranged into a cylindrical mold and then second extruded into a die with an orifice of 2 mm diameter, to make two-layer and three-layer multifilament rods. The same numbers of two-layer and three-layer monofilament rods were randomly mixed, arranged into the cylindrical mold, and reextruded into a die of 2.0 mm orifice diameter, to make mixed 50% two-layer:50% three-layer multifilament rods. The aim of the mixed multifilament rods was to interdisperse, on a small, microstructural scale, regions of high strength with regions of high toughness. The multifilament rods were cut into 47 mm lengths; 55 multifilament rods were arranged into a molding die and warm pressed at 150°C and 34.5 MPa. The binder was then removed from the pressed pellet. The heat treatment cycle for the binder removal from the composites without graphite was as follows: heat from 25° to 250°C at a ramp rate of 0.05°C/min, heat from 250° to 450°C at a ramp rate of 0.1°C/min, heat from 450° to 650°C at a ramp rate of 0.3°C/min, maintain at 650°C for 2 h, and subsequently cool down to room temperature with a ramp rate of 0.5°C/min. The binder-free body was CIPed at 413.7 MPa, and then sintered at 1600°C for 10 h. The binder removal cycle for the composite with graphite in its interphase was the same as that of the composite without graphite except for maintaining at 550°C for 2 h. After removal of binder, the pellet was CIPed at 413.7 MPa, and then the graphite was removed from the composites using a heating cycle as follows: from 25° to 550°C heat at a ramp rate of 8.8°C/min, from 550° to 800°C heat at a ramp rate of 0.08°C/min, and from 800° to 900°C heat at a ramp rate of 0.1°C/min and maintain at 900°C for 2 h. The binder- and graphite-free pellet was sintered at 1600°C for 10 h. The microstructures of the monofilament rod, the multifilament rod, and the green pellet were examined by optical microscopy (Model SMZ-2T, Nikon, Tokyo, Japan). Scanning electron microscopy (Model S-4700, Hitachi, Osaka, Japan) was used to analyze the microstructures of sintered and mechanically tested, fibrous monolithic composites. Flexural strengths were measured with a screw-driven machine (Model 4502, Instron Corp., Canton, MA) in a three-point bend testing. The work of fracture for each sample was determined by calculating the area under the load versus displacement curve. The strength and work of fracture data for each composite were determined after testing three to five samples. The supporting span was 30 mm, the crosshead speed was 0.1 mm/min, and the sample size was 3 mm (H) 4 mm (W) 40 mm (L). III. Results and Discussion To change the number of filaments in a given area of fibrous monolithic composite by changing the number of filaments in a two-layer multifilament rod, the first extrusions were made through dies having 1.5, 2.0, and 4.0 mm diameter orifices. Figure 1 shows the monofilament rods which were passed through the 4.0 mm (Fig. 1(a)) and 1.5 mm (Fig. 1(b)) diameter orifices, respectively. The porous and weak, AlPO4 interphase layer was well coated around the mullite center rod. The 25, 93, and 150 monofilament rods which were first extruded through the 4.0, 2.0, and 1.5 mm diameter orifices, respectively, were arranged into a cylindrical mold of 23 mm diameter and then extruded again through an orifice of 2 mm diameter to make 25, 93, and 150 multifilament rods, respectively. Figure 2 comprises optical micrographs of the 25 (Fig. 2(a)) and 150 (Fig. 2(b)) multifilament rod samples. The population densities of the 25, 93, and 150 multifilament rod samples were 7, 27, and 43 filaments/mm2 , Fig. 3. Optical micrographs of green pellets: (a) 25 filament green pellet, (b) 150 filament green pellet. Table II. Effects of Green Interphase Thickness on the Strength and Work of Fracture of Sintered Fibrous Monolithic Composites Green interphase thickness (mm) Strength (MPa) Work of fracture (kJ/m2 ) 0.33 76 5 0.45 0.02 0.19 41 2 0.49 0.05 0.073 162 10 0.26 0.03 Table III. Effects of Amount of Graphite in the Green AlPO4 Interphase on the Strength and Work of Fracture of Sintered Fibrous Monolithic Composites† Amount of graphite in the green interphase (vol%) Strength (MPa) Work of fracture (kJ/m2 ) 0 162 10 0.26 0.03 10 109 6 0.61 0.02 30 102 10 0.69 0.06 50 77 5 0.58 0.05 † Green interphase thickness 0.073 mm. 796 Journal of the American Ceramic Society—Kim and Kriven Vol. 87, No. 5
May 2004 Mullite(3A1,03 2SiO,)-Aluminum Phosphate(AlPO4, Oxide, Fibrous Monolithic Composites 79 了mm (b) ig. 4. SEM micrographs of the sintered, mullite-AIPO fibrous monolithic composite (The thickness of the interphase was 5-10 um after sintering. The mposition of the green interphase was 50 vol% graphite: 50 vol% AlPO4, and the graphite was removed after heat treatment. spectively. The green, bend bar samples were made by stack into the interphase to cause full densification. Intermediate 55 multifilament rods into a rectangular die and pressing at 34 ness has a suitable interphase strength for debonding at room MPa. The optical micrographs of the green rectangular samples are shown in Fig 3. It was estimated that 289 and 1774 flame aligned in the 7.5 mm x 5.5 on the application temperature of the composite multifilament gre were tested in three-point bending They had bend strengths of6±2,76±s,and4±lMPa, spectively, with works of fracture of 0.10 0.02, 0.45+ 0.02 0.1 d0.03+0.01 kJ/m, respectively To make fibrous monolithic composites having different thick nesses of AlPO4 interphase layer, monofilament rods with interphase thicknesses of 0.33, 0.19, and 0.073 mm were extruded 三跺〓」 Table Il summarizes the effects of interphase thickness on the three-point bend strength and the work of fracture of the sintered composites. Even though the fibrous monolithic composite with ar AlPO4 interphase thickness of 0.073 mm had the highest strength of 162 10 MPa, it showed brittle fracture and had the lowest work of fracture of 0.26 0.03 k/mm To increase the work of fracture of this composite by making a more porous interphase and acilitating debonding and fiber pullout, 10, 30, and 50 vol% of graphite powder were added to the green interphase. Table Ill 002 represents the results of mechanical testing of these composites The strengths of the sintered composites decreased with increasin amounts of graphite in the green interphase. However the works of fracture of the composites were increase 00004-0.0600801012 der to the interphase. The fibrous monolithic te with 30 vol% graphite in the interphase had the highest work of fracture Displacement(mm) of 0.69 +0.06 k/m- The thin interphase was fully densified by diffusion of matrix after a relatively short period. In the case Dad-displacement curves for the three different kinds of thick interphase, there was insufficient matrix powder diffus Oa fibrous monolithic composites fabricated
respectively. The green, bend bar samples were made by stacking 55 multifilament rods into a rectangular die and pressing at 34.5 MPa. The optical micrographs of the green rectangular samples are shown in Fig. 3. It was estimated that 289 and 1774 filaments were aligned in the 7.5 mm 5.5 mm area for the 25 and 150 multifilament green samples, respectively. The 25, 93, and 150 multifilament sintered bars were tested in three-point bending. They had bend strengths of 6 2, 76 5, and 4 1 MPa, respectively, with works of fracture of 0.10 0.02, 0.45 0.02, and 0.03 0.01 kJ/m2 , respectively. To make fibrous monolithic composites having different thicknesses of AlPO4 interphase layer, monofilament rods with green interphase thicknesses of 0.33, 0.19, and 0.073 mm were extruded. Table II summarizes the effects of interphase thickness on the three-point bend strength and the work of fracture of the sintered composites. Even though the fibrous monolithic composite with an AlPO4 interphase thickness of 0.073 mm had the highest strength of 162 10 MPa, it showed brittle fracture and had the lowest work of fracture of 0.26 0.03 kJ/mm2 . To increase the work of fracture of this composite by making a more porous interphase and facilitating debonding and fiber pullout, 10, 30, and 50 vol% of graphite powder were added to the green interphase. Table III represents the results of mechanical testing of these composites. The strengths of the sintered composites decreased with increasing amounts of graphite in the green interphase. However the works of fracture of the composites were increased after adding graphite powder to the interphase. The fibrous monolithic composite with 30 vol% graphite in the interphase had the highest work of fracture of 0.69 0.06 kJ/m2 . The thin interphase was fully densified by diffusion of matrix after a relatively short period. In the case of the thick interphase, there was insufficient matrix powder diffusion into the interphase to cause full densification. Intermediate thickness has a suitable interphase strength for debonding at room temperature. However, the dependence of the matrix diffusion into the interphase on interphase thickness can be changed, depending on the application temperature of the composite. Fig. 4. SEM micrographs of the sintered, mullite–AlPO4 fibrous monolithic composite. (The thickness of the interphase was 5–10 m after sintering. The composition of the green interphase was 50 vol% graphite:50 vol% AlPO4, and the graphite was removed after heat treatment.) Fig. 5. Load–displacement curves for the three different kinds of mullite–AlPO4 fibrous monolithic composites fabricated. May 2004 Mullite (3Al2O32SiO2)–Aluminum Phosphate (AlPO4), Oxide, Fibrous Monolithic Composites 797
Journal of the American Ceramic Sociery Kim and Kriven Vol. 87. No. 5 M M 1 mm 1 mm Fig. 6. Optical micrographs of the monofilament rods composed of mullite and AlPO4(M: matrix(mullite); 1: interphase(AlPO4): (a) two-layer structure, (b)three-layer structure. Further efforts to make tougher fibrous monolithic of 76+5 and 83 15 MPa, respectively, and works of fracture of 0.45 0.02 and 0.46 0.03 k/m", respectively. The interphase composition of 50 vol% graphite and 50 vo The microstructures of such composites after sintering showed brittle fracture and had a bending strength of 106+ 5 MPa for 10 h are seen in Fig. 4. The interphase thickness of that and a work of fracture of 0. 17+0.03 kJ/m composite was 5-10 um after sintering. The microstructures wer The microstructures of the composites consisted of two-layer, very uniform and homogeneous. The strength and work of fracture mixed 50% two-layer: 50% three-layer, and three-layer textures of that composite were 129 2 MPa and 0.86 0.05 kJ/m respectively. To compare values, a single mullite pellet was made Figure 6 presents optical micrographs of the two-layer(Fig. 6(a)) sintered at 1600C for 10 h, and then tested in three-point bending and three-layer, first-extruded, monofilament rods(Fig. 6(b)) The strength and work of fracture of single-phase mullite w Figure 7 presents optical micrographs of three-layer( Fig. 7(a))an mixed 50% two-layer: 50% three-layer multifilament rods(Fig 308+ 11 MPa and 0.53+ 0.01 kJ/m?, respectively. Thus, 7(b)resulting from the second extrusion. The three-layer multi compared with mullite, the fibrous monolith just described had 42% of single-phase mullite strength, and 162% of the work of filament rod contained about 93 three-layer textures in a circle of fracture of pure mullite. 2.1 mm diameter. The 50 vol% two-laver and 50 vol% three-laver To increase the overall strength of the composite, 10 and 30 monofilament rods were randomly mixed and extruded a second vol% mullite powders were added to the aluminum phosphate time to make a mixed 50% two-layer: 50% three-layer multifila interphase. The green interphase thickness of the composite then 50% three-layer multifile was 0.33 mm. Figure 5 compares the load versus displacement ment rod possessed an interlocking texture of the mullite matrix curves for the three different kinds of composites Composites with and APOa interphase. Figure 8 displays optical micrographs of pure AlPO4 and 10 vol% mullite added to the interphase compo- three-layer(Fig. 8(a)) and mixed 50% two-layer: 50% three-layer ition showed apparent nonbrittle fracture, with bending strengths (Fig 8(b) green bodies 1 mm 1 mm (a) Fig. 7. Optical microgrphs of multifilament rods: (a) three-layer structure, (b)mixed 50% two-layer: 50% three-layer structure
Further efforts to make tougher fibrous monolithic composites included making a composite with a thinner interphase and a green interphase composition of 50 vol% graphite and 50 vol% AlPO4. The microstructures of such composites after sintering at 1600°C for 10 h are seen in Fig. 4. The interphase thickness of that composite was 5–10 m after sintering. The microstructures were very uniform and homogeneous. The strength and work of fracture of that composite were 129 2 MPa and 0.86 0.05 kJ/m2 , respectively. To compare values, a single mullite pellet was made, sintered at 1600°C for 10 h, and then tested in three-point bending. The strength and work of fracture of single-phase mullite were 308 11 MPa and 0.53 0.01 kJ/m2 , respectively. Thus, compared with mullite, the fibrous monolith just described had 42% of single-phase mullite strength, and 162% of the work of fracture of pure mullite. To increase the overall strength of the composite, 10 and 30 vol% mullite powders were added to the aluminum phosphate interphase. The green interphase thickness of the composite then was 0.33 mm. Figure 5 compares the load versus displacement curves for the three different kinds of composites. Composites with pure AlPO4 and 10 vol% mullite added to the interphase composition showed apparent nonbrittle fracture, with bending strengths of 76 5 and 83 15 MPa, respectively, and works of fracture of 0.45 0.02 and 0.46 0.03 kJ/m2 , respectively. The composite with 30-vol%-mullite-added interphase composition showed brittle fracture and had a bending strength of 106 5 MPa and a work of fracture of 0.17 0.03 kJ/m2 . The microstructures of the composites consisted of two-layer, mixed 50% two-layer:50% three-layer, and three-layer textures. Figure 6 presents optical micrographs of the two-layer (Fig. 6(a)) and three-layer, first-extruded, monofilament rods (Fig. 6(b)). Figure 7 presents optical micrographs of three-layer (Fig. 7(a)) and mixed 50% two-layer:50% three-layer multifilament rods (Fig. 7(b)) resulting from the second extrusion. The three-layer multifilament rod contained about 93 three-layer textures in a circle of 2.1 mm diameter. The 50 vol% two-layer and 50 vol% three-layer monofilament rods were randomly mixed and extruded a second time to make a mixed 50% two-layer:50% three-layer multifilament rod. The mixed 50% two-layer:50% three-layer multifilament rod possessed an interlocking texture of the mullite matrix and AlPO4 interphase. Figure 8 displays optical micrographs of three-layer (Fig. 8(a)) and mixed 50% two-layer:50% three-layer (Fig. 8(b)) green bodies. Fig. 6. Optical micrographs of the monofilament rods composed of mullite and AlPO4 (M: matrix (mullite); I: interphase (AlPO4)): (a) two-layer structure, (b) three-layer structure. Fig. 7. Optical microgrphs of multifilament rods: (a) three-layer structure, (b) mixed 50% two-layer:50% three-layer structure. 798 Journal of the American Ceramic Society—Kim and Kriven Vol. 87, No. 5
May 2004 Mullite(3Al,03 2SiO2/ -Aluminum Phosphate (AlPO4, Oxide, Fibrous Monolithic Composites 1 mm ma (b) Fig. 8. Optical micrographs of the green pellets: (a) three-layer structure, (b)mixed 50% two-layer: 50% three-layer structure 50 microns 9. SEM micrographs of the sintered, two-layer, mullite-AlPO4 fibrous monolithic composite (M: matrix(mullite), 1: interphase(AIPO), arrows ate the circumferential cracks formed around mullite center rods 300 microns人 50 microns 10. SEM micrographs of the sintered, three-layer, mullite-AlPOa fibrous monolithic composite(M: matrix(mullite); I: interphase(AlPO4)
Fig. 8. Optical micrographs of the green pellets: (a) three-layer structure, (b) mixed 50% two-layer:50% three-layer structure. Fig. 9. SEM micrographs of the sintered, two-layer, mullite–AlPO4 fibrous monolithic composite (M: matrix (mullite); I: interphase (AlPO4); arrows indicate the circumferential cracks formed around mullite center rods). Fig. 10. SEM micrographs of the sintered, three-layer, mullite–AlPO4 fibrous monolithic composite (M: matrix (mullite); I: interphase (AlPO4)). May 2004 Mullite (3Al2O32SiO2)–Aluminum Phosphate (AlPO4), Oxide, Fibrous Monolithic Composites 799
800 Journal of the American Ceramic Sociery Kim and Kriven Vol. 87. No. 5 The SEM micrographs of the sintered two-layer composite are presented in Fig 9. The dense mullite matrix, as well as the porous nd weak AlPO interphase, sintered to a uniform microstructur The higher-magnification SEM micrograph shows the microstruc ture of the porous AlPO4 interphase layer. The arrows in Fig. %(a) indicate microcracks formed along the mullite-AlPOa interfaces after sintering. Figures 10(a) and(b) are SEM micrographs of the three-layer mullite-AIPO4 fibrous monolithic composite. The three-layer structure of the mullite inner rod-AlPO4 interphase layer-mullite outer layer can clearly be seen. There is no formation of interface microcracks in the three-layer composite The difference in the sintering behaviors of the two-layer and three-layer fibrous monolithic composites is schematically ex- plained in Fig. 11. In the two-layer fibrous monolithic composite (Fig. Il(a), the isolated, highly sinterable, inner mullite matrix rods underwent sintering shrinkage, but the interconnected, poorly sinterable interphase layers did not densify significantly. Because Fig. 12. SEM micrograph of the sintered, mixed 50% two-layer: 50% of this sinterability difference, circumferential shrinkage cracks three-layer, mullite-AIPOA fibrous monolithic composite. were formed around the mullite inner rods, as seen in Fig 9(a), and the two-layer composite had little sintering shrinkage. In the case of the three-layer structure(Fig. 11(b)), the isolated, mullite inner rods shrunk, but the AlPOa interphase layers did not shrink as composites. These materials showed apparent nonbrittle, interme- much. However, the interconnected, mullite outer layer shrunk considerably and produced compressive stresses in the AlPO nterphase layers. Thus, the overall sample size was reduced composite are seen in Fig. 14. The side view shows that the crack during sintering and the circumferential cracks did not form followed a tortuous path due to deflection. The cross-sectional around the mullite inner rods as can be seen in Fig. 10. The view confirms that the crack propagates along the weak AlPO4 SEM micrograph of the sintered, mixed 50% two-layer: 50% nterphase. The SEM micrographs of Fig. 15 display the fracture three-layer fibrous monolithic composite is presented in Fig. surfaces of the two-layer fibrous monolithic composite, and give 12. The mullite matrix and AlPO4 interphase layer formed an evidence for somewhat extensive fiber pullout and hence a rough terlocking microstructure. fracture surface. This kind of extensive fiber pullout, apparent The three-point bending strengths for the two-layer, mixed 50% nonbrittle fracture, and low strength behavior in the two-layer corresponding works of fracture of 0.45 +0.02, 0.30+0.05, and 9(a)and as discussed using a xpected from the circumferential shrinkage cracks around the mullite center rods as depicted in Fig 0.25 0.01 kJ/m, respectively. Figure 13 shows load versus three-layer fibrous monolithic composite is seen in Fig. 16. The two-layer: 50% three-layer, and three-layer fibrous monolithic crack propagates not only along the weak APOS surfaces of the displacement curves corresponding to the two-layer, mixed 50% across the strong mullite matrix phase. Fracture three-layer fibrous monolithic composites are seen in Fig. 17. The three-layer structure was maintained throughout the composite, but there was very little fiber pullout, at room temperature, leading to a relatively smooth fracture surface. These poor fiber pullout, brittle fracture, and higher strength behaviors of the three-layer fibrous monolithic composite could be predicted from the com- pressional stresses acting on the interphase layer during sintering. as can be seen from Fig. 10 and as illustrated in Fig. 11(b) OOo O0O 0.15 (a)Two-layer fibrous monolithic composite (Mullite- AIPO4) 50% 2-layer and 50%3-layer 3·laye 005 000000 (b) Three-layer fibrous monolithic composite(Mullite- AIPO4-mullite) E::::: Matrix(Mullite )2: Interphase(AIPO):Crack Displacement (mm) Fig. 13. Load-displacement curves for the two-layer, mixed 50% two- Fig. 11. Schematic of the sintering behavior of the two-layer and layer: 50% three-layer, and three-layer mullite-AlPO4 fibrous monolithic three-layer fibrous monolithic composite
The SEM micrographs of the sintered two-layer composite are presented in Fig. 9. The dense mullite matrix, as well as the porous and weak AlPO4 interphase, sintered to a uniform microstructure. The higher-magnification SEM micrograph shows the microstructure of the porous AlPO4 interphase layer. The arrows in Fig. 9(a) indicate microcracks formed along the mullite–AlPO4 interfaces after sintering. Figures 10(a) and (b) are SEM micrographs of the three-layer mullite–AlPO4 fibrous monolithic composite. The three-layer structure of the mullite inner rod–AlPO4 interphase layer–mullite outer layer can clearly be seen. There is no formation of interface microcracks in the three-layer composite. The difference in the sintering behaviors of the two-layer and three-layer fibrous monolithic composites is schematically explained in Fig. 11. In the two-layer fibrous monolithic composite (Fig. 11(a)), the isolated, highly sinterable, inner mullite matrix rods underwent sintering shrinkage, but the interconnected, poorly sinterable interphase layers did not densify significantly. Because of this sinterability difference, circumferential shrinkage cracks were formed around the mullite inner rods, as seen in Fig. 9(a), and the two-layer composite had little sintering shrinkage. In the case of the three-layer structure (Fig. 11(b)), the isolated, mullite inner rods shrunk, but the AlPO4 interphase layers did not shrink as much. However, the interconnected, mullite outer layer shrunk considerably and produced compressive stresses in the AlPO4 interphase layers. Thus, the overall sample size was reduced during sintering and the circumferential cracks did not form around the mullite inner rods as can be seen in Fig. 10. The SEM micrograph of the sintered, mixed 50% two-layer:50% three-layer fibrous monolithic composite is presented in Fig. 12. The mullite matrix and AlPO4 interphase layer formed an interlocking microstructure. The three-point bending strengths for the two-layer, mixed 50% two-layer:50% three-layer, and three-layer structures were measured as 76 5, 123 12, and 176 7 MPa, respectively, having corresponding works of fracture of 0.45 0.02, 0.30 0.05, and 0.25 0.01 kJ/m2 , respectively. Figure 13 shows load versus displacement curves corresponding to the two-layer, mixed 50% two-layer:50% three-layer, and three-layer fibrous monolithic composites. These materials showed apparent nonbrittle, intermediate, and brittle fracture behavior, respectively. The side and cross- sectional views of a fractured two-layer fibrous monolithic composite are seen in Fig. 14. The side view shows that the crack followed a tortuous path due to deflection. The cross-sectional view confirms that the crack propagates along the weak AlPO4 interphase. The SEM micrographs of Fig. 15 display the fracture surfaces of the two-layer fibrous monolithic composite, and give evidence for somewhat extensive fiber pullout and hence a rough fracture surface. This kind of extensive fiber pullout, apparent nonbrittle fracture, and low strength behavior in the two-layer fibrous monolithic composite is expected from the circumferential shrinkage cracks around the mullite center rods as depicted in Fig. 9(a) and as discussed using Fig. 11(a). Crack propagation in the three-layer fibrous monolithic composite is seen in Fig. 16. The crack propagates not only along the weak AlPO4 interphase, but also across the strong mullite matrix phase. Fracture surfaces of the three-layer fibrous monolithic composites are seen in Fig. 17. The three-layer structure was maintained throughout the composite, but there was very little fiber pullout, at room temperature, leading to a relatively smooth fracture surface. These poor fiber pullout, brittle fracture, and higher strength behaviors of the three-layer fibrous monolithic composite could be predicted from the compressional stresses acting on the interphase layer during sintering, as can be seen from Fig. 10 and as illustrated in Fig. 11(b). Fig. 11. Schematic of the sintering behavior of the two-layer and three-layer fibrous monolithic composite. Fig. 12. SEM micrograph of the sintered, mixed 50% two-layer:50% three-layer, mullite–AlPO4 fibrous monolithic composite. Fig. 13. Load–displacement curves for the two-layer, mixed 50% twolayer:50% three-layer, and three-layer mullite–AlPO4 fibrous monolithic composite. 800 Journal of the American Ceramic Society—Kim and Kriven Vol. 87, No. 5
May 2004 Mullite(3A1,03 2SiO,)-Aluminum Phosphate(AlPO4, Oxide, Fibrous Monolithic Composites M epmm 200me Fig. 14. SEM micrographs of the fractured two-layer fibrous monolithic composite (M: matrix(mullite): I: interphase (AlPO4):(a) side view, (b) 500 microns (a) (b) Fig. 15. SEM micrographs of the fracture surface of the two-layer fibrous monolithic composite. Fracture surfaces of the mixed 50% two-layer: 50% three-layer toughening at elevated temperatures. Further work needs to be fibrous monolithic composites are presented in Fig 18. The shape done on such composites to explore their high-temperature me- of pulled-out fibers was irregular, and the fracture surface was chanical behavior While the room-temperature mechanical properties of the three- layer, oxide, fibrous monolithic microstructure produced were not cantly tough, they may still exhibit significant pullout and IV. Summary Mullite-AlPO4, oxide fibrous monolithic composit different microstructures, interphase thicknesses, and tions were made by a co-extrusion technique. Of the mullite-AlPO4 fibrous monolithic composites made with 25, 93 and 150 multifilament rods, the 93 multifilament composite exhibited apparent nonbrittle fracture and had the highest three- bending strength of 76 +5 MPa Of the sintered composites having different interphase thicknesses, the composite with 0.07 mm green interphase thickness had the highest three-point bend strength of 162 * 10 MPa, but underwent brittle fracture a work of fracture of 0.26 +0.03 kJ/m. By adding 30 graphite to the fibrous monolithic composite with a green hase thickness of 0.073 mm. the work of fracture increased to .69+0.06 kJ/m. A fibrous monolithic composite with sintered nterphase thickness of 5-10 um and interphase composition 300 micron ontaining 50 vol% graphite showed pseudoductile fracture behav- ior and had a bend strength of 129 2 MPa and a work of fracture of 0.86+0.05 kJ/m2. The composites with pure AIPOa and 16. SEM micrograph showing crack propagation in the three-layer and 30-vol%-mullite-added interphase compositions were fabri cated. As the amount of mullite in the interphase increased, the
Fracture surfaces of the mixed 50% two-layer:50% three-layer fibrous monolithic composites are presented in Fig. 18. The shape of pulled-out fibers was irregular, and the fracture surface was rough. While the room-temperature mechanical properties of the threelayer, oxide, fibrous monolithic microstructure produced were not significantly tough, they may still exhibit significant pullout and toughening at elevated temperatures. Further work needs to be done on such composites to explore their high-temperature mechanical behavior. IV. Summary Mullite–AlPO4, oxide fibrous monolithic composites having different microstructures, interphase thicknesses, and compositions were made by a co-extrusion technique. Of the sintered mullite–AlPO4 fibrous monolithic composites made with 25, 93, and 150 multifilament rods, the 93 multifilament composite exhibited apparent nonbrittle fracture and had the highest threepoint bending strength of 76 5 MPa. Of the sintered composites having different interphase thicknesses, the composite with 0.073 mm green interphase thickness had the highest three-point bend strength of 162 10 MPa, but underwent brittle fracture, having a work of fracture of 0.26 0.03 kJ/m2 . By adding 30 vol% graphite to the fibrous monolithic composite with a green interphase thickness of 0.073 mm, the work of fracture increased to 0.69 0.06 kJ/m2 . A fibrous monolithic composite with sintered interphase thickness of 5–10 m and interphase composition containing 50 vol% graphite showed pseudoductile fracture behavior, and had a bend strength of 129 2 MPa and a work of fracture of 0.86 0.05 kJ/m2 . The composites with pure AlPO4 and 10- and 30-vol%-mullite-added interphase compositions were fabricated. As the amount of mullite in the interphase increased, the Fig. 14. SEM micrographs of the fractured two-layer fibrous monolithic composite (M: matrix (mullite); I: interphase (AlPO4)): (a) side view, (b) cross-sectional view. Fig. 15. SEM micrographs of the fracture surface of the two-layer fibrous monolithic composite. Fig. 16. SEM micrograph showing crack propagation in the three-layer composite. May 2004 Mullite (3Al2O32SiO2)–Aluminum Phosphate (AlPO4), Oxide, Fibrous Monolithic Composites 801
802 Journal of the American Ceramic Sociery Kim and Kriven Vol. 87. No. 5 00 microns (b) Fig. 17. SEM microgrphs of the fracture surface of the three-layer fibrous monolithic composite 500 microns Fig. 18. SEM micrographs of the fracture surface of the mixed 50% two-layer: 50% three-layer fibrous monolithic composite. three-point bending strength of the composite was increased K. T. Faber and A. G. Evans, "Intragranular Crack However, the work of fracture of the composites did not show any Silicon Carbide, J. Am. Ceram. Soc., 66, C-94-C-96(1983). clear dependence on the composition of the composes d thre Evans, "Crack Deflection Processes-l. Theory,Acta Metall2,314]565-7601983) K. T. Faber, A.G. Evans, and M. D. Dory, "A Statistical layer mullite-AlPOa fibrous monolithic composites were fabri- cated. They exhibited apparent nonbrittle, intermediate, and brittle Mechanics of Ceramics, Vol. 6. Edited by R. C. Bradt, A. G D. P H fracture behaviors, respectively. The two-layer composite had the Hasselman, and FF. Lange. Plenum Press, New York, 1983 lowest three-point bending strength of 76 5 MPa and the highest N. Claussen, J, Steep, and R. F. Pabst, "Effects of Induced Microcracking on the work of fracture of 0.45+ 0.02 kJ/m. The two-layer mullite- A. G. E, gness of Ceramics,"Am Ceram. Soc. Bull, 56 [6]559-62(1977) ing of Ceramics by Circumferential AlPO4 fibrous monolithic composite showed evidence of exten- Microcracking, ". Am. Ceram. Soc., 64 [7] 394-98198 sive pullout and had a corresponding rough fracture surface. The .A. G. Evans, "Toughening Mechanisms in Zirconia Alloys' 193-212in reasons for this behavior were attributed to the formation of Advances in Ceramics, Vol. 12, Science and Technology of Zirconia /I Edited aussen, M. Ruhle, and A H. Heuer. American Ceramic Society, Columbus, OH, circumferential debonding cracks around the inner mullite rods sulting from differential sintering shrinkages between the porous M. Ruhle, N. Claussen, and A. H. Heuer, "Transformation and Microcrack AlPO interphase and the isolated inner mullite rods. The low level ing as Complementary Processes in ZrO2-Toughened Al2O3, " J Am Ceram. of pullout and brittle fracture of the three-layer fibrous monolithic so269115970196 of Fiber-Reinforced Ceramic Matrix composite were attributed to the formation of compressive stresses Cor in the AlPOa interphase layer, resulting from the high sintering Zok, Review: The Physics and Mechanics of shrinkage of the interconnected outer mullite layer. The fracture Fiber-Reinforced Brittle Matrix Composites, ". Mater. Sci., 29, 3857-96( 1994) urface of the mixed 50% two-layer: 50% three-layer fibrous Fiber-Reinforced Ceramics, ". Mech. Phys. Solids, 34, 167-89(1986). monolithic composite showed irregularly shaped pullout. nouless and A. G. Evans, " Effects of Pullout on the Mechanical perties of Ceramic-Matrix Composites, Acta Metall, 36, 517-22(1988) A. Curtin,"Theory of Mechanical Properties of Ceramic-Matrix 16W.J. Clegg. K. Kendall on, and J R. C. Garvie, R. H. J. Hannink, T. Pascoe."Ceramic Steel? Nature Simple Way to Make Tough Ceramics, Nature(London), 347[41455 17P. Boch, T. Chartier and M. H. Heuer, "Mechanisms of Toughening Partially Stabilized Composites,J. Am. Ceram Soc. Zirconia(PSZ),J.Am. Ceran. Soc., 60[3-4]183-84(1977) zhes, and D. S. Wilkinson, "Processing of H. Heuer, N. Claussen, W. M. Kriven, and M. Ruhle, "Stability of Tetragonal Tape Cast Laminates Prepared fror Alumina/Zirconia Powders, J.Anm Ceran ZrO2 Particles in Ceramic Matrices, J. Am. Ceram Soc., 65 [12]642-50(1982). Soc,778]2145-53(1994)
three-point bending strength of the composite was increased. However, the work of fracture of the composites did not show any clear dependence on the composition of the composites. Two-layer, mixed 50% two-layer:50% three-layer, and threelayer mullite–AlPO4 fibrous monolithic composites were fabricated. They exhibited apparent nonbrittle, intermediate, and brittle fracture behaviors, respectively. The two-layer composite had the lowest three-point bending strength of 76 5 MPa and the highest work of fracture of 0.45 0.02 kJ/m2 . The two-layer mullite– AlPO4 fibrous monolithic composite showed evidence of extensive pullout and had a corresponding rough fracture surface. The reasons for this behavior were attributed to the formation of circumferential debonding cracks around the inner mullite rods, resulting from differential sintering shrinkages between the porous AlPO4 interphase and the isolated inner mullite rods. The low level of pullout and brittle fracture of the three-layer fibrous monolithic composite were attributed to the formation of compressive stresses in the AlPO4 interphase layer, resulting from the high sintering shrinkage of the interconnected outer mullite layer. The fracture surface of the mixed 50% two-layer:50% three-layer fibrous monolithic composite showed irregularly shaped pullout. References 1 R. C. Garvie, R. H. J. Hannink, and R. T. Pascoe, “Ceramic Steel?,” Nature (London), 258, 703–704 (1975). 2 D. L. Porter and A. H. Heuer, “Mechanisms of Toughening Partially Stabilized Zirconia (PSZ),” J. Am. Ceram. Soc., 60 [3–4] 183–84 (1977). 3 A. H. Heuer, N. Claussen, W. M. Kriven, and M. Ru¨hle, “Stability of Tetragonal ZrO2 Particles in Ceramic Matrices,” J. Am. Ceram. Soc., 65 [12] 642–50 (1982). 4 K. T. Faber and A. G. Evans, “Intragranular Crack-Deflection Toughening in Silicon Carbide,” J. Am. Ceram. Soc., 66, C-94–C-96 (1983). 5 K. T. Faber and A. G. Evans, “Crack Deflection Processes–I. Theory,” Acta Metall., 31 [4] 565–76 (1983). 6 K. T. Faber, A. G. Evans, and M. D. Dory, “A Statistical Analysis of Crack Deflection as a Toughening Mechanism in Ceramic Materials”; pp. 77–91 in Fracture Mechanics of Ceramics, Vol. 6. Edited by R. C. Bradt, A. G. Evans, D. P. H. Hasselman, and F. F. Lange. Plenum Press, New York, 1983. 7 N. Claussen, J. Steep, and R. F. Pabst, “Effects of Induced Microcracking on the Fracture Toughness of Ceramics,” Am. Ceram. Soc. Bull., 56 [6] 559–62 (1977). 8 A. G. Evans and K. T. Faber, “Toughening of Ceramics by Circumferential Microcracking,” J. Am. Ceram. Soc., 64 [7] 394–98 (1981). 9 A. G. Evans, “Toughening Mechanisms in Zirconia Alloys”; pp. 193–212 in Advances in Ceramics, Vol. 12, Science and Technology of Zirconia II. Edited by N. Claussen, M. Ru¨hle, and A. H. Heuer. American Ceramic Society, Columbus, OH, 1984. 10M. Ru¨hle, N. Claussen, and A. H. Heuer, “Transformation and Microcrack Toughening as Complementary Processes in ZrO2-Toughened Al2O3,” J. Am. Ceram. Soc., 69 [3] 195–97 (1986). 11A. G. Evans, “The Mechanical Performance of Fiber-Reinforced Ceramic Matrix Composites,” Mater. Sci. Eng., A107, 227–39 (1989). 12A. G. Evans and F. W. Zok, “Review: The Physics and Mechanics of Fiber-Reinforced Brittle Matrix Composites,” J. Mater. Sci., 29, 3857–96 (1994). 13B. Budiansky, J. W. Hutchinson, and A. G. Evans, “Matrix Fracture in Fiber-Reinforced Ceramics,” J. Mech. Phys. Solids, 34, 167–89 (1986). 14M. D. Thouless and A. G. Evans, “Effects of Pullout on the Mechanical Properties of Ceramic-Matrix Composites,” Acta Metall., 36, 517–22 (1988). 15W. A. Curtin, “Theory of Mechanical Properties of Ceramic-Matrix Composities,” J. Am. Ceram. Soc., 74 [11] 2837–45 (1991). 16W. J. Clegg, K. Kendall, N. M. Alford, T. W. Button, and J. D. Birchall, “A Simple Way to Make Tough Ceramics,” Nature (London), 347 [4] 455–57 (1990). 17P. Boch, T. Chartier, and M. Huttepain, “Tape Casting of Al2O3/ZrO2 Laminated Composites,” J. Am. Ceram. Soc., 69 [8] C-191–C-192 (1986). 18K. P. Plucknett, C. H. Caceres, C. Hughes, and D. S. Wilkinson, “Processing of Tape Cast Laminates Prepared from Fine Alumina/Zirconia Powders,” J. Am. Ceram. Soc., 77 [8] 2145–53 (1994). Fig. 17. SEM microgrphs of the fracture surface of the three-layer fibrous monolithic composite. Fig. 18. SEM micrographs of the fracture surface of the mixed 50% two-layer:50% three-layer fibrous monolithic composite. 802 Journal of the American Ceramic Society—Kim and Kriven Vol. 87, No. 5
May 2004 Mullite(3A1-03 2Si02-Aluminum Phosphate(AlPO4, Oxide, Fibrous Monolithic Composites 803 S J. Lee and w. M. Kriven,"Toughed Oxide Composites Based on Porous, 3SG. Hilmas, A. Brady, U. Abdali, G. Zywicki, and J. W. Halloran,"Fibrous sbS. C Mirek, "Phase Transformation Weakening in Fibrous Ceramic Composites ic Microlaminates by Electrophoretic Deposition, "J. Am. Ceram Soc., 75[10] 2907-909(1992) an Investigation of the Enstatite(MgSiO3MTitania(TiO2)System", MS, Thesis H. Wang and x. Hu, "Surface Properties of Ceramic Laminates Fabricated by Die niversity of Illinois, Urbana, IL, 1993 S J. Lee and w. M. Kriven, "Toughened Oxide Composites Based on Porous, Pressing," J. Am. Ceram Soc., 79(2]553-56(1996) Fabrication and Failure of Laminar Ceramic Composites, Alumina Platelet Interphases, J. Am. Ceram Soc., 84(4]767-74(2001). Acta metall.,4013085-9(199 T. Nakamura, and T 2T. Shannon and S Blackburn, "The Production of Alumina/Zirconia Laminated of Mullite Ceramics", pp, 605-12 in Ceramic Transaction Composites by Co-extrusion, "Ceram, Eng. Sci. Proc., 16[5-6] 1115-20(1995) Vol. 6, Mullite and Mullite Matrix Composites. Edited by S. Somrya, R. F. Davis, D Morgan and D B. Marshall, "Ceramic Composites of Monazite and J. A. Pask. American Ceramic Society, Westerville, OH, 1990 3D. E C. Corbidge, Phosphorus: An Outline of Its Chemistry, Biochemistry ceramic and Method for its Production, U.S. Pat. No. 4772 524, September 20, 1988 4J. V. Bothe Jr and P W. Brown,"Low-Temperature Formation of Aluminum D. Kovar, B. H. King, R. w. Trice, and J. w. Halloran,"Fibrous Monolithic Orthophosphate,JAm Ceram Soc., 76[2]362-68(1993 Ceramics,J. Am. Ceram Soc., 80[10]2471-87(1997). 4D. K. Kim and W. M. Kriven,"Mullite(3Al- 0, 2SiO,-Aluminum Phosphate Baskaran, S. D. Nunn, D. Popovic, and J. w. Halloran,"Fibrous Monolithic O4)Laminated Composite Fabricated by Tape Casting, J.Am. Ceran. Soc. microstructure, and Indentation Behavior, "J. Am. Ceram. [111 1962-64(2003) Soc,76{92209-16(1993) -M. A. Gulguin and W. M. Kriven, "A Simple Solution-Polymerization Route for S. Baskaran and J. w. Halloran, "Fibrous Monolithic Ceramics: Il, Flexural iation of Porder Synthesis ns, ol. 62, Sciemce Oxide Powder Synthesis", Pp. 57-66 in Ceramic Transacti Strength and Fracture Behavior of the Silicon Carbide/Graphite System, "J.Am. Cera.Soc,7692217-24(1993) SD. Popovic, S. Baskaran, G. Zywicki, C. Arens, and J. W. Hallora Nitride and Silocon Carbide Fibrous Monolithic Ceramics", Pp. 173-86 4M. A Gulgun, M. H. Nguyen, and W.M. Transactions, Vol. 42, Silicon Based Structural Ceramics. Edited by B. V 4M. H Nguyen, S. J. Lee, and W. M. Riven, Properties and Oxidation Behavior of the Silicon Carbide/Boron Nitride System, Polymeric Steric Entrapment Precursor Route, " J. Mater. Res, 14 [8] 3417-26 JAm. Cera. Soc.,77(51249-55(1994) 999) G. E. Hilmas, G. A. Brady, and J. w. Halloran, "SiC and Si3N4 Fibrous 4SW. M. Kriven, S. J. Lee, M, A. Gulgun, M. H. Nguyen, and D. K. Kim, Synthesis of Oxide Powders via Polymeric Steric Entrapment": pp. 99-110 in p. 609-14 in Ceramic Transactions, Vol, 51, Ceramic Processing ceramic Transactions, Vol. 108, InnoTative Processing and Synthesis of Ce Science and Technology. Edited by H, Hausner, G. L Messing, and S. Hirano. Glasses, and Composites Il. Edited by J.P. Singh, N. P. Bansal, and K. Niihara. Ceramic Society. Westerville. OH 2000. ' S. Baskaran, S. D. Nunn, and J. W. Halloran, "Fibrous Monolithic Ceramics: guye,""Processes for Preparing Mechanical Properties and Oxidation Behavior of the Alumina/Nickel Syster Mixed-Oxide Powders, " U.S. Pat. No 6482 387, November 19, 2002 J.Am. Ceram soc,77{51256-62(1994) 47D. K. Kim and W. M. Kriven, "Oxide Fibrous Monoliths of Mullite-AIPOa and 3G. A. Danko, D. Popovic, K. Stuffle, B. H. King, J. W. Halloran, and J. w Alumina/Y AG-Alumina Platelets, Mater. Sci. Eng, A, in press. Holmes, "Commercial Development of Fibrous Monolithic Ceramics, "Ceram. Eng 4D. K. Kim, "Design of Sintered, Tough, Oxide Laminate and Fibrous Monolithic Sc.Proc,165]673-80(1995) Composites", Ph D. Thesis. University of Illinois, Urbana, IL, 2001
19S. J. Lee and W. M. Kriven, “Toughed Oxide Composites Based on Porous, Alumina Platelet Interphases,” J. Am. Ceram. Soc., 84 [4] 767–74 (2001). 20H. Takebe and K. Morinaga, “Fabrication and Mechanical Properties of Lamellar Al2O3 Ceramics,” J. Ceram. Soc. Jpn., 96 [12] 1149–56 (1988). 21P. Sarkar, X. Haung, and P. S. Nicholson, “Structural Ceramic Microlaminates by Electrophoretic Deposition,” J. Am. Ceram. Soc., 75 [10] 2907–909 (1992). 22H. Wang and X. Hu, “Surface Properties of Ceramic Laminates Fabricated by Die Pressing,” J. Am. Ceram. Soc., 79 [2] 553–56 (1996). 23W. J. Clegg, “The Fabrication and Failure of Laminar Ceramic Composites,” Acta Metall., 40 [11] 3085–93 (1992). 24T. Shannon and S. Blackburn, “The Production of Alumina/Zirconia Laminated Composites by Co-extrusion,” Ceram. Eng. Sci. Proc., 16 [5–6] 1115–20 (1995). 25P. E. D. Morgan and D. B. Marshall, “Ceramic Composites of Monazite and Alumina,” J. Am. Ceram. Soc., 78 [6] 1553–63 (1995). 26W. S. Coblenz, “Fibrous Monolithic Ceramic and Method for its Production,” U.S. Pat. No. 4 772 524, September 20, 1988. 27D. Kovar, B. H. King, R. W. Trice, and J. W. Halloran, “Fibrous Monolithic Ceramics,” J. Am. Ceram. Soc., 80 [10] 2471–87 (1997). 28S. Baskaran, S. D. Nunn, D. Popovic, and J. W. Halloran, “Fibrous Monolithic Ceramics: I, Fabrication, Microstructure, and Indentation Behavior,” J. Am. Ceram. Soc., 76 [9] 2209–16 (1993). 29S. Baskaran and J. W. Halloran, “Fibrous Monolithic Ceramics: II, Flexural Strength and Fracture Behavior of the Silicon Carbide/Graphite System,” J. Am. Ceram. Soc., 76 [9] 2217–24 (1993). 30D. Popovic, S. Baskaran, G. Zywicki, C. Arens, and J. W. Halloran, “Silicon Nitride and Silocon Carbide Fibrous Monolithic Ceramics”; pp. 173–86 in Ceramic Transactions, Vol. 42, Silicon Based Structural Ceramics. Edited by B. W. Sheldon and S. C. Danforth. American Ceramic Society, Westerville, OH, 1994. 31S. Baskaran and J. W. Halloran, “Fibrous Monolithic Ceramics: III, Mechanical Properties and Oxidation Behavior of the Silicon Carbide/Boron Nitride System,” J. Am. Ceram. Soc., 77 [5] 1249–55 (1994). 32G. E. Hilmas, G. A. Brady, and J. W. Halloran, “SiC and Si3N4 Fibrous Monoliths: Non-Brittle Fracture from Powder Processed Ceramics Produced by Coextrusion”; pp. 609–14 in Ceramic Transactions, Vol. 51, Ceramic Processing Science and Technology. Edited by H. Hausner, G. L. Messing, and S. Hirano. American Ceramic Society, Westerville, OH, 1994. 33S. Baskaran, S. D. Nunn, and J. W. Halloran, “Fibrous Monolithic Ceramics: IV, Mechanical Properties and Oxidation Behavior of the Alumina/Nickel System,” J. Am. Ceram. Soc., 77 [5] 1256–62 (1994). 34G. A. Danko, D. Popovic, K. Stuffle, B. H. King, J. W. Halloran, and J. W. Holmes, “Commercial Development of Fibrous Monolithic Ceramics,” Ceram. Eng. Sci. Proc., 16 [5] 673–80 (1995). 35G. Hilmas, A. Brady, U. Abdali, G. Zywicki, and J. W. Halloran, “Fibrous Monoliths: Non-brittle Fracture from Powder-Processed Ceramics,” Mater. Sci. Eng., A195, 263–68 (1995). 36S. C. Mirek, “Phase Transformation Weakening in Fibrous Ceramic Composites an Investigation of the Enstatite (MgSiO3)/Titania (TiO2) System”; M.S. Thesis. University of Illinois, Urbana, IL, 1993. 37S. J. Lee and W. M. Kriven, “Toughened Oxide Composites Based on Porous, Alumina Platelet Interphases,” J. Am. Ceram. Soc., 84 [4] 767–74 (2001). 38H. Ohnishi, K. Maeda, T. Nakamura, and T. Kawanami, “High Temperature Mechanical Properties of Mullite Ceramics”; pp. 605–12 in Ceramic Transaction, Vol. 6, Mullite and Mullite Matrix Composites. Edited by S. Somiya, R. F. Davis, J. A. Pask. American Ceramic Society, Westerville, OH, 1990. 39D. E. C. Corbidge, Phosphorus: An Outline of Its Chemistry, Biochemistry and Technology; pp. 141. Elsevier, New York, 1985. 40J. V. Bothe Jr. and P. W. Brown, “Low-Temperature Formation of Aluminum Orthophosphate,” J. Am. Ceram. Soc., 76 [2] 362–68 (1993). 41D. K. Kim and W. M. Kriven, “Mullite (3Al2O32SiO2)–Aluminum Phosphate (AlPO4) Laminated Composite Fabricated by Tape Casting,” J. Am. Ceram. Soc., 86 [11] 1962–64 (2003). 42M. A. Gu¨lgu¨n and W. M. Kriven, “A Simple Solution-Polymerization Route for Oxide Powder Synthesis”; pp. 57–66 in Ceramic Transactions, Vol. 62, Science, Technology, and Commercialization of Powder Synthesis and Shape Forming Processes. Edited by J. J. Kingsley, C. H. Schilling, and J. H. Adair. American Ceramic Society, Westerville, OH, 1996. 43M. A. Gu¨lgu¨n, M. H. Nguyen, and W. M. Kriven, “Polymerized Organic– Inorganic Synthesis of Mixed Oxides,” J. Am. Ceram. Soc., 82 [3] 556–60 (1999). 44M. H. Nguyen, S. J. Lee, and W. M. Kriven, “Synthesis of Oxide Powders via a Polymeric Steric Entrapment Precursor Route,” J. Mater. Res., 14 [8] 3417–26 (1999). 45W. M. Kriven, S. J. Lee, M. A. Gu¨lgu¨n, M. H. Nguyen, and D. K. Kim, “Synthesis of Oxide Powders via Polymeric Steric Entrapment”; pp. 99–110 in Ceramic Transactions, Vol. 108, Innovative Processing and Synthesis of Ceramics, Glasses, and Composites III. Edited by J. P. Singh, N. P. Bansal, and K. Niihara. American Ceramic Society, Westerville, OH, 2000. 46M. A. Gu¨lgu¨n, W. M. Kriven, and M. H. Nguyen, “Processes for Preparing Mixed-Oxide Powders,” U.S. Pat. No. 6 482 387, November 19, 2002. 47D. K. Kim and W. M. Kriven, “Oxide Fibrous Monoliths of Mullite–AlPO4 and Alumina/YAG–Alumina Platelets,” Mater. Sci. Eng., A, in press. 48D. K. Kim, “Design of Sintered, Tough, Oxide Laminate and Fibrous Monolithic Composites”; Ph.D. Thesis. University of Illinois, Urbana, IL, 2001. May 2004 Mullite (3Al2O32SiO2)–Aluminum Phosphate (AlPO4), Oxide, Fibrous Monolithic Composites 803