《复合材料 Composites》课程教学资源（学习资料）第五章 陶瓷基复合材料_19fibrous molithic-23 Microstructure control and mechanical properties of fibrous Al2O3/ZrO2 composites fabricated by extrusion process

Availableonlineatwww.sciencedirect.com SCE區 DIRECT G ta materialia ELSEVIER Scripta Materialia 52(2005)725-729 www.actamat-journals.com Microstructure control and mechanical properties of fibrous Al2O3 /ZrO2 composites fabricated by extrusion process Taek-Soo Kim , Takashi Goto, Byong-Taek Lee Adcanced Materials R&D Center, Korea Institute of Industrial Technology, 994-32 Techno-park Songdo, Dongchhun-Dong, Incheon 406-130, Korea b Institute for Material, Research, Tohoku Unicersity, 2-1-1, Katahira, Aoba-ku, Sendai, Japan School of Advanced Materials Engineering, Kongju National University, 182 Shinkwan-dong, Kongju, Chungnam 314-701, Korea Received 17 August 2004: received in revised form 10 November 2004: accepted 15 December 2004 Abstract Fibrous Al2O3/ZrO2 composites were newly fabricated using an extrusion process, and the relationship between the microstruc ure and mechanical properties was investigated. The density, hardness, bending strength and fracture toughness increased as the number of extrusion passes increased, and the highest values attained were 98.5%, 12 GPa, 564 MPa and 6.3 MPa m, respectively o 2004 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved Keywords: Al2Or-ZrO2 composites: Microstructure control; Extrusions; Mechanical properties; Fracture surface 1. Introduction and shape of continuously porous ceramic bodies [10] The microstructure control of ceramic composites was Al2O3/ZrO2 composites are widely used in many done by repeated extrusion of ceramic powders after industrial areas due to their excellent properties such mixing with an organic binder. There is, however, no as high strength, excellent corrosion resistance, good work reported on the mechanical properties of fibrous biocompatibility and sound wear resistance [1-4].A Al,O3/ZrO, composites fabricated by the multi-extru- drawback of the materials is their intrinsic low fracture sion process as a function of controlled microstructure toughness, which prevents their further application in In this work, the extrusion process was used to fabr those parts which require a bigger volume and more reli cate the Al,O3/ZrO, composites as well as to control the ability. Recently, it has been reported that the formation microstructure from the millimeter to nanometer scale of fibrous structures is a promising alternative in the In order to investigate a relationship between the con- modification of fracture toughness [5-7]; researchers re- trolled microstructure and mechanical properties, the ported that the fibrous structure formed in Si3 N,Bn density, hardness, bending strength and fracture tough- composites enhanced the fracture toughness corre- ness were measured as a function of the number of ponding to the fiber pulling-out fracture patterns. Re- extrusion passes cently, Kim et al. reported that the extrusion process was very useful in controlling the microstructure of Al2O3/ZrOz composites [8, 9], as well as the pore size 2. Experimental Corresponding author. Tel: +82 32 8500 409: fax: +82 32 8500 In order to fabricate the Al,O,m-ZrO2)/50 vol% t-zrO2 composites using an extrusion process, commer E-mail address: skim(@kitech re kr (T.-S. Kim) cial t-ZrO2 and Al2O3(m-zrO2) powders were mixed 1359-6462/S- see front matter 2004 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved doi: 10.1016/j-scriptamat. 2004.12.022

Microstructure control and mechanical properties of fibrous Al2O3/ZrO2 composites fabricated by extrusion process Taek-Soo Kim a,*, Takashi Goto b , Byong-Taek Lee c a Advanced Materials R&D Center, Korea Institute of Industrial Technology, 994-32Techno-park Songdo, Dongchun-Dong, Incheon 406-130, Korea b Institute for Material, Research, Tohoku University, 2-1-1, Katahira, Aoba-ku, Sendai, Japan c School of Advanced Materials Engineering, Kongju National University, 182Shinkwan-dong, Kongju, Chungnam 314-701, Korea Received 17 August 2004; received in revised form 10 November 2004; accepted 15 December 2004 Abstract Fibrous Al2O3/ZrO2 composites were newly fabricated using an extrusion process, and the relationship between the microstruc￾ture and mechanical properties was investigated. The density, hardness, bending strength and fracture toughness increased as the number of extrusion passes increased, and the highest values attained were 98.5%, 12 GPa, 564 MPa and 6.3 MPa m1/2, respectively.
2004 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Al2O3–ZrO2 composites; Microstructure control; Extrusions; Mechanical properties; Fracture surface 1. Introduction Al2O3/ZrO2 composites are widely used in many industrial areas due to their excellent properties such as high strength, excellent corrosion resistance, good biocompatibility and sound wear resistance [1–4]. A drawback of the materials is their intrinsic low fracture toughness, which prevents their further application in those parts which require a bigger volume and more reli￾ability. Recently, it has been reported that the formation of fibrous structures is a promising alternative in the modification of fracture toughness [5–7]; researchers re￾ported that the fibrous structure formed in Si3N4/BN composites enhanced the fracture toughness corre￾sponding to the fiber pulling-out fracture patterns. Re￾cently, Kim et al. reported that the extrusion process was very useful in controlling the microstructure of Al2O3/ZrO2 composites [8,9], as well as the pore size and shape of continuously porous ceramic bodies [10]. The microstructure control of ceramic composites was done by repeated extrusion of ceramic powders after mixing with an organic binder. There is, however, no work reported on the mechanical properties of fibrous Al2O3/ZrO2 composites fabricated by the multi-extru￾sion process as a function of controlled microstructure. In this work, the extrusion process was used to fabri￾cate the Al2O3/ZrO2 composites as well as to control the microstructure from the millimeter to nanometer scale. In order to investigate a relationship between the con￾trolled microstructure and mechanical properties, the density, hardness, bending strength and fracture tough￾ness were measured as a function of the number of extrusion passes. 2. Experimental In order to fabricate the Al2O3–(m-ZrO2)/50 vol.% t-ZrO2 composites using an extrusion process, commer￾cial t-ZrO2 and Al2O3–(m-ZrO2) powders were mixed 1359-6462/$ - see front matter
2004 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.scriptamat.2004.12.022 * Corresponding author. Tel.: +82 32 8500 409; fax: +82 32 8500 390. E-mail address: tskim@kitech.re.kr (T.-S. Kim). www.actamat-journals.com Scripta Materialia 52 (2005) 725–729

T-S. Kim et al. Scripta Materialia 52(2005)725-729 with a binder in a shear mixer( C w. Brabender Instru- using Archimedes principle. Hardness of the sintered ments, Inc, PL2000 Plasti-Corder, USA), in which body was measured using a Vickers hardness tester Al2O3 25 vol %(m-ZrO2) was prepared using a ball mill-(HV-112, Akashi, Japan)under a pressure of 1.96N ing process; m-ZrO2 and t-ZrO2 indicate monoclinic and for 10 s The fracture toughness, KIc was obtained using tetragonal ZrO2, respectively. The m-ZrO2 was added to a combination of the indentation method and the evans suppress the grain growth of Al_O3 during sintering. The equation at a load of 10 kg as follows, than 300 nm for A10,(AKP-50, Sumimoto, Japan) Kic=(E/)(/ce' and 80 nm for ZrO2 powder( Toso, Japan). The binder where A is a material-independent constant (=0.016), E material was composed of ethylene vinyl acetate is the elastic modulus of the materials being indented, H (EVA)in granules(Elvax 250, Dupont). For lubrication the hardness, P the applied load and c the crack half during blending, stearic acid( CH3(CH2)16COOH, Dae- length jung Chemicals Metals Co, Ltd)was added. The vol to measure the bending strength, four-po ume per cent of the mixture was 50/50 for the Al2O3 bending tests were carried out using a universal testing powder/polymer and 40/60 for the ZrO2 powder/poly- machine (UnitechM, R&B, Korea), in which the mer. In order to blend each powder with polymer, the round-type of filament as extruded was used for the test mixing head and chamber of the blender were first specimen. The distance between the upper rolls was heated to 130C using an oil-type heating source, fol- 30 mm and between the bottom rolls 10 mm lowed by addition of the polymer. After rotating the head for about 120 s each ceramic powder was added slowly 3. Results and discussion Mixtures of Al2O3-ZrO2/polymer and ZrO/polymer vere extruded into filaments of 3. 5 mm in diameter with Fig. 1(a)(d) shows BSE-SEM micrographs of a sin an area reduction ratio of 70: 1. The first passed Al2O3- ered Al2O3(m-ZrO2)/ZrO2 composite body fabricated te ZrO2/polymer filament and ZrO/polymer filament were by the multi-extrusion process, in which Fig. 1(aH(d) reloaded into the extrusion die with an identical volume. show the second-, third-, fourth- and fifth- pass fila The extrusion was repeated until the fifth-pass filament ments, respectively. The composite presents a homoge- was obtained. The extrusion temperature was 120C neous distribution of AlO3(m-ZrO2) and ZrO2 and the rate was 15 mm/min. Binder burning-out of phases. The Zro2 phase appears to be brighter than the filaments was carried out at 700C under flowing Al2O3-(m-ZrO2) phases. In order to arrange both the fil- nitrogen atmosphere, and sintered at 1450C aments in a certain pattern as shown in Fig. 1, the fila Microstructure and fracture surface of the filaments ments of composition Al2O3(m-ZrO2)and ZrO2 were vere examined using backscattered electrons(BSE)in initially prepared using the extrusion process. The size a scanning electron microscope (SEM) (JSM 6331 F, of both the filaments was 3. 5 mm in diameter. The same Japan). The density of the sintered body was measured number of each extruded filament were put in the con 含 Fig 1. BSE-SEM micrographs of sintered AlO]-(m-ZrO2)Zro2 composite body fabricated by fibrous monolithic process:(akd) were obtained by the second, third, fourth and fifth extrusions, respectively

with a binder in a shear mixer (C.W. Brabender Instru￾ments, Inc., PL2000 Plasti-Corder, USA), in which Al2O3 25 vol.% (m-ZrO2) was prepared using a ball mill￾ing process; m-ZrO2 and t-ZrO2 indicate monoclinic and tetragonal ZrO2, respectively. The m-ZrO2 was added to suppress the grain growth of Al2O3 during sintering. The average particle size of the starting materials was less than 300 nm for Al2O3 (AKP-50, Sumimoto, Japan) and 80 nm for ZrO2 powder (Toso, Japan). The binder material was composed of ethylene vinyl acetate (EVA) in granules (Elvax 250, Dupont). For lubrication during blending, stearic acid (CH3(CH2)16COOH, Dae￾jung Chemicals & Metals Co., Ltd) was added. The vol￾ume per cent of the mixture was 50/50 for the Al2O3 powder/polymer and 40/60 for the ZrO2 powder/poly￾mer. In order to blend each powder with polymer, the mixing head and chamber of the blender were first heated to 130 C using an oil-type heating source, fol￾lowed by addition of the polymer. After rotating the head for about 120 s, each ceramic powder was added slowly. Mixtures of Al2O3–ZrO2/polymer and ZrO2/polymer were extruded into filaments of 3.5 mm in diameter with an area reduction ratio of 70:1. The first passed Al2O3– ZrO2/polymer filament and ZrO2/polymer filament were reloaded into the extrusion die with an identical volume. The extrusion was repeated until the fifth-pass filament was obtained. The extrusion temperature was 120 C and the rate was 15 mm/min. Binder burning-out of the filaments was carried out at 700 C under flowing nitrogen atmosphere, and sintered at 1450 C. Microstructure and fracture surface of the filaments were examined using backscattered electrons (BSE) in a scanning electron microscope (SEM) (JSM 6331 F, Japan). The density of the sintered body was measured using Archimedes principle. Hardness of the sintered body was measured using a Vickers hardness tester (HV-112, Akashi, Japan) under a pressure of 1.96 N for 10 s. The fracture toughness, K1C was obtained using a combination of the indentation method and the Evans equation at a load of 10 kg as follows, K1C ¼ AðE=HÞ 1=2 ðP=c3=2 Þ where A is a material-independent constant (=0.016), E is the elastic modulus of the materials being indented, H the hardness, P the applied load and c the crack half￾length. In order to measure the bending strength, four-point bending tests were carried out using a universal testing machine (UnitechTM, R&B, Korea), in which the round-type of filament as extruded was used for the test specimen. The distance between the upper rolls was 30 mm and between the bottom rolls 10 mm. 3. Results and discussion Fig. 1(a)–(d) shows BSE-SEM micrographs of a sin￾tered Al2O3–(m-ZrO2)/ZrO2 composite body fabricated by the multi-extrusion process, in which Fig. 1(a)–(d) show the second-, third-, fourth- and fifth- pass fila￾ments, respectively. The composite presents a homoge￾neous distribution of Al2O3–(m-ZrO2) and ZrO2 phases. The ZrO2 phase appears to be brighter than Al2O3–(m-ZrO2) phases. In order to arrange both the fil￾aments in a certain pattern as shown in Fig. 1, the fila￾ments of composition Al2O3–(m-ZrO2) and ZrO2 were initially prepared using the extrusion process. The size of both the filaments was 3.5 mm in diameter. The same number of each extruded filament were put in the con￾Fig. 1. BSE-SEM micrographs of sintered Al2O3–(m-ZrO2)/ZrO2 composite body fabricated by fibrous monolithic process: (a)–(d) were obtained by the second, third, fourth and fifth extrusions, respectively. 726 T.-S. Kim et al. / Scripta Materialia 52 (2005) 725–729

T-S. Kim et al. Scripta Materialia 52(2005)725-729 tainer of 30 mm in diameter, and then extruded through a die of 3.5 mm in diameter each filament was com- ■ Relative density Bending strength formly as designated. Each pha ase became thin ner by the repeated extrusion. The grain size became fine as the number of extrusion passes increased, in which the thicknesses of both the Al,O(m-zrO2) and Zro yers were about 375 50 um, 5 um and 0.7 um for the second-, third-, fourth- and the fifth-pass filament It is anticipated that microstructural change hap- pened during the extrusion using a relationship between Pass of extrusion the mean diameter(Dn)of the filaments with the pass of extrusion; the relationship can be described as follows, Fig. 2. Relative density and bending strength of the Al2Or-(m-ZrO2) ZrO2 sintered composite with the number of extrusion passes. Dn=D-1/R/ strength is attributed to the microstructural refinement where D,_I is initial diameter of the container, and R is (shown in Fig. 1)as well as the densification an extrusion ratio. The equation indicates that the diam Fig. 3 shows fracture surfaces of fibrous Al,O3- eter of both the extruded bar and the fiber varies only (m-ZrO2)Zro2 composites examined by Bse (a)and depending on the reduction ratio(R). The resultant val- secondary electron(b and c) modes on SEM. The sec- ues(D) for the first- to fifth-pass filaments are 3.5 ond-pass filament fractured well and maintains well 418 um, 45 um, 6 um and 0. 7 um, respectively. It is seen the as extruded microstructure, in which the Al2O3- that the data agrees well, indicating that the equation is (m-ZrO2) particles are darker than Zro2. It is seen that a convenient way for the materials designer to anticipate fracture of the second filament progresses in a transgra he microstructural variations for a ceramic composite nular way with a partial intergranular way due to a formation of different fracture patterns between Al2O The rather heterogeneous microstructure in the fifth- (m-ZrO2) and ZrO, phases. This is identified from pass filament is attributed to factors such as a relatively he magnified photos of ZrO2(Fig. 3b)and Al2O3- large size mismatch between the as received powders (m-ZrO2)(Fig. 3c), where fracture with fine dimples (AlO3 300 nm and Zro2 80 nm in diameter), discret observed from the Al2O3-(m-ZrO2) parts. The com ancy in the plastic flow rate between Al2O,m-ZrO2)/ bined fracture modes in the fibrous composites could polymer and ZrO2/polymer layers during extrusion possibly modify the intrinsic low fracture toughness of and the large flexibility of the polymer compared to ceramic materials as reported elsewhere [5-9]. Not only the ceramic [10 that, the homogeneous and fibrous microstructure The relationship between the microstructure and the formed in the extruded composites could improve the materials properties is as shown in Fig. 2, in which rel- other mechanical properties, such as fracture toughness ative density and bending strength of fibrous Al2O3- Fig. 4 shows a fracture pattern taken from the third (m-zrO2)/ZrO2 composites are plotted against the num passed composite. The fractured specimen presents a ber of extrusion passes. Both values are simultaneously microstructure identical to the one as extruded(Fig increased as the number of extrusion passes increased, 4a), and builds up a sound interface between Al_- where the density varies from 95.8% for the second-pass (m-ZrO2) and ZrO2 phases(Fig 4b ). It is seen that the filaments to 98.5% for the fifth-pass filaments, while the transgranular fracture with the dimples formed in the respective bending strength changes from 160to hird-pass filament, but the portion of intergranular 564 MPa. The increase in density might be due to the re- mode determined from the dimples is less than that peated extrusion, whereas the increase in bending for the second one. ZrO2(Fig. 4c)presents a still more Fig. 3. BSE-SEM fractography of the second passed composite examined at(a) low magnification and high magnification of (b) ZrO2 and (c)Al2O3-

tainer of 30 mm in diameter, and then extruded through a die of 3.5 mm in diameter. Each filament was com￾bined uniformly as designated. Each phase became thin￾ner by the repeated extrusion. The grain size became fine as the number of extrusion passes increased, in which the thicknesses of both the Al2O3–(m-ZrO2) and ZrO2 layers were about 375 lm, 50 lm, 5 lm and 0.7 lm for the second-, third-, fourth- and the fifth-pass filaments, respectively. It is anticipated that microstructural change hap￾pened during the extrusion using a relationship between the mean diameter (Dn) of the filaments with the pass of extrusion; the relationship can be described as follows, Dn ¼ Dn1=R1=2 ð1Þ where Dn1 is initial diameter of the container, and R is an extrusion ratio. The equation indicates that the diam￾eter of both the extruded bar and the fiber varies only depending on the reduction ratio (R). The resultant val￾ues (Dn) for the first- to fifth-pass filaments are 3.5 mm, 418 lm, 45 lm, 6 lm and 0.7 lm, respectively. It is seen that the data agrees well, indicating that the equation is a convenient way for the materials designer to anticipate the microstructural variations for a ceramic composite bar. The rather heterogeneous microstructure in the fifth￾pass filament is attributed to factors such as a relatively large size mismatch between the as received powders (Al2O3 300 nm and ZrO2 80 nm in diameter), discrep￾ancy in the plastic flow rate between Al2O3–(m-ZrO2)/ polymer and ZrO2/polymer layers during extrusion and the large flexibility of the polymer compared to the ceramic [10]. The relationship between the microstructure and the materials properties is as shown in Fig. 2, in which rel￾ative density and bending strength of fibrous Al2O3– (m-ZrO2)/ZrO2 composites are plotted against the num￾ber of extrusion passes. Both values are simultaneously increased as the number of extrusion passes increased, where the density varies from 95.8% for the second-pass filaments to 98.5% for the fifth-pass filaments, while the respective bending strength changes from 160 to 564 MPa. The increase in density might be due to the re￾peated extrusion, whereas the increase in bending strength is attributed to the microstructural refinement (shown in Fig. 1) as well as the densification. Fig. 3 shows fracture surfaces of fibrous Al2O3– (m-ZrO2)/ZrO2 composites examined by BSE (a) and secondary electron (b and c) modes on SEM. The sec￾ond-pass filament fractured well and maintains well the as extruded microstructure, in which the Al2O3– (m-ZrO2) particles are darker than ZrO2. It is seen that fracture of the second filament progresses in a transgra￾nular way with a partial intergranular way due to a formation of different fracture patterns between Al2O3– (m-ZrO2) and ZrO2 phases. This is identified from the magnified photos of ZrO2 (Fig. 3b) and Al2O3– (m-ZrO2) (Fig. 3c), where fracture with fine dimples is observed from the Al2O3–(m-ZrO2) parts. The com￾bined fracture modes in the fibrous composites could possibly modify the intrinsic low fracture toughness of ceramic materials as reported elsewhere [5–9]. Not only that, the homogeneous and fibrous microstructure formed in the extruded composites could improve the other mechanical properties, such as fracture toughness. Fig. 4 shows a fracture pattern taken from the third￾passed composite. The fractured specimen presents a microstructure identical to the one as extruded (Fig. 4a), and builds up a sound interface between Al2O3– (m-ZrO2) and ZrO2 phases (Fig. 4b). It is seen that the transgranular fracture with the dimples formed in the third-pass filament, but the portion of intergranular mode determined from the dimples is less than that for the second one. ZrO2 (Fig. 4c) presents a still more Fig. 2. Relative density and bending strength of the Al2O3–(m-ZrO2)/ ZrO2 sintered composite with the number of extrusion passes. Fig. 3. BSE-SEM fractography of the second passed composite examined at (a) low magnification and high magnification of (b) ZrO2 and (c)Al2O3– (m-ZrO2). T.-S. Kim et al. / Scripta Materialia 52 (2005) 725–729 727

T.S. Kim et al Scripta Materialia 52(2005)725-729 t-zro Q Fig 4. BSE-SEM fractography of the third passed composite, in which(a) shows an overall microstructure while(b) shows an interface between(c) ZrOz phase and (d)Al2O3-(m-ZrO2) phase brittle pattern than that of Al_O3(m-ZrO2)(Fig. 4d), toughness, showing the highest value of approximately although the difference between the phases has almost 5.5 MPa m"at the fourth-pass and fifth-pass filaments disappeared. The similarity in the fracture pattern be- The improvement in both the properties with increasing tween the phases becomes more clear in the fourth and number of passes is due to the grain size refinement of fifth-pass filaments(not shown). Thus, the mechanical the homogeneous microstructure. The increase obtained properties of fibrous Al2O3-(m-ZrO2) and ZrO2 com- from the second to the fourth passes corresponded to a posites fabricated by the extrusion process seem to be combination of the fracture modes as well as the grain dependent on the degree of microstructural refinement refinement, while only the operation of grain refining be- and densification as the number of extrusion passes in- tween the fourth-pass and the fifth-pass filaments con- crease,while the addition of the fiber pullout effect is tributes to the increase in the hardness. The formation only effective in the initial steps of extrusion as in the of the partially heterogeneous phase in the fifth-pass fil- case of the second and third-pass filaments ament as shown in Fig. I(d) would not improve the frac- Fig. 5 shows hardness and fracture toughness of fi- ture toughness further brous Al2O3-(m-zrO2)/ZrO2 composites measured with the number of extrusion passes. The initial hardness of about 11. 1 GPa from the second-pass filament is in- 4. Conclusion creased, respectively, to 11.3, 11. 8 and 12. 1 GPa as the number of extrusion passes increased up to the fifth Homogeneously fibrous Al2O3(m-ZrO2 )/ZrO2 com- pass. A similar pattern was also presented in the fracture posites were fabricated, and the variation of microstruc properties were investiga function of the number of extrusion passes. The thick nesses of both phases were reduced to 375um, 50um Hardness 5um and 0.7um as the number of extrusion passes in creased up to the fifth pass. Bending strength and hard- ness also increased with the extrusion passes, due to the microstructural refinement and the densification the maximum values in the density, hardness and bending strength were, respectively, 98.5%, 1236 Hv and 564 MPa, which were all obtained from the fifth-pass fil- aments. The fracture toughness was enhanced until the fourth pass, while not between the fourth and fifth-pass filaments, corresponding to the similar fracture patterns between Al_O3(m-ZrO2) and ZrO, and the partial for- Fig. 5. Hardness and fracture toughness of the A1 O,-(m-Zro2VZrO2 mation of heterogeneous phases in the fifth-pass fila mposite with the number of extrusion passes ment. In order to maintain the different fracture mode

brittle pattern than that of Al2O3–(m-ZrO2) (Fig. 4d), although the difference between the phases has almost disappeared. The similarity in the fracture pattern be￾tween the phases becomes more clear in the fourth and fifth-pass filaments (not shown). Thus, the mechanical properties of fibrous Al2O3–(m-ZrO2) and ZrO2 com￾posites fabricated by the extrusion process seem to be dependent on the degree of microstructural refinement and densification as the number of extrusion passes in￾crease, while the addition of the fiber pullout effect is only effective in the initial steps of extrusion as in the case of the second and third-pass filaments. Fig. 5 shows hardness and fracture toughness of fi- brous Al2O3–(m-ZrO2)/ZrO2 composites measured with the number of extrusion passes. The initial hardness of about 11.1 GPa from the second-pass filament is in￾creased, respectively, to 11.3, 11.8 and 12.1 GPa as the number of extrusion passes increased up to the fifth￾pass. A similar pattern was also presented in the fracture toughness, showing the highest value of approximately 5.5 MPa m1/2 at the fourth-pass and fifth-pass filaments. The improvement in both the properties with increasing number of passes is due to the grain size refinement of the homogeneous microstructure. The increase obtained from the second to the fourth passes corresponded to a combination of the fracture modes as well as the grain refinement, while only the operation of grain refining be￾tween the fourth-pass and the fifth-pass filaments con￾tributes to the increase in the hardness. The formation of the partially heterogeneous phase in the fifth-pass fil￾ament as shown in Fig. 1(d) would not improve the frac￾ture toughness further. 4. Conclusion Homogeneously fibrous Al2O3–(m-ZrO2)/ZrO2 com￾posites were fabricated, and the variation of microstruc￾ture and mechanical properties were investigated as a function of the number of extrusion passes. The thick￾nesses of both phases were reduced to 375lm, 50lm, 5lm and 0.7lm as the number of extrusion passes in￾creased up to the fifth pass. Bending strength and hard￾ness also increased with the extrusion passes, due to the microstructural refinement and the densification. The maximum values in the density, hardness and bending strength were, respectively, 98.5%, 1236 Hv and 564 MPa, which were all obtained from the fifth-pass fil￾aments. The fracture toughness was enhanced until the fourth pass, while not between the fourth and fifth-pass filaments, corresponding to the similar fracture patterns between Al2O3–(m-ZrO2) and ZrO2 and the partial for￾mation of heterogeneous phases in the fifth-pass fila￾ment. In order to maintain the different fracture mode Fig. 4. BSE-SEM fractography of the third passed composite, in which (a) shows an overall microstructure while (b) shows an interface between (c) ZrO2 phase and (d) Al2O3–(m-ZrO2) phase. Fig. 5. Hardness and fracture toughness of the Al2O3–(m-ZrO2)/ZrO2 composite with the number of extrusion passes. 728 T.-S. Kim et al. / Scripta Materialia 52 (2005) 725–729

T.S. Kim et al. Scripta Materialia 52(2005)725-729 p to the higher passes, the combination of more strong References and weak phases is thought to be needed. Higher frac- ture toughness was obtained from the longitudinal [1 Farmer SC, Sayir A. Eng Fract Mech 2002: 69: 1015. plane, corresponding to the fibrous alignment of the [ Li w, Gao L Biomater 2003: 24: 937 constituent phases 3 Tarlazzi A, Roncari E, Pinasco P, Guicciardi S, Melandri C. We 4 Jordan DR Ophthal Clin North Am 2000: 13(4): 587 5 Betz U, Sturm A, Offor JF, Wagner W, wiedenmann A, Hahn H. Mater Sci Eng A 2000: 281: 68. Acknowledgment [6 Coretta KC, Gutierrez- Mora F, Picciolo JJ, Routbort JL Mater This work was supported by the 2002 NRL research hemann W. Scr Mater 2001: 45: 435 sKim 1 Gh. Goto t Lee bt mater Transer 2004: 45: 431 program of Korean Ministry of Science and (9 Kim Ts DH,Goto T,Lee BT. Rev Adv Mat Sci 2004: 6:34 T echnology [0] Kim TS, Kang IC, Goto T, Lee BT Mater Trans 2003: 44: 1851

up to the higher passes, the combination of more strong and weak phases is thought to be needed. Higher frac￾ture toughness was obtained from the longitudinal plane, corresponding to the fibrous alignment of the constituent phases. Acknowledgment This work was supported by the 2002 NRL research program of Korean Ministry of Science and Technology. References [1] Farmer SC, Sayir A. Eng Fract Mech 2002;69:1015. [2] Li W, Gao L. Biomater 2003;24:937. [3] Tarlazzi A, Roncari E, Pinasco P, Guicciardi S, Melandri C. Wear 2003;244:29. [4] Jordan DR. Ophthal Clin North Am 2000;13(4):587. [5] Betz U, Sturm A, Lofflor JF, Wagner W, Wiedenmann A, Hahn H. Mater Sci Eng A 2000;281:68. [6] Goretta KC, Gutierrez-Mora F, Picciolo JJ, Routbort JL. Mater Sci Eng A 2003;341(1–2):158. [7] Yang XC, Riehemann W. Scr Mater 2001;45:435. [8] Kim TS, Kim GH, Goto T, Lee BT. Mater Transer 2004;45:431. [9] Kim TS, Jang DH, Goto T, Lee BT. Rev Adv Mat Sci 2004;6:34. [10] Kim TS, Kang IC, Goto T, Lee BT. Mater Trans 2003;44:1851. T.-S. Kim et al. / Scripta Materialia 52 (2005) 725–729 729