P. Pettersson, M. Johnsson /Journal of the European Ceramic Society 23(2003)309-313 (SEM, JEOL 820). The whisker yield is estimated to be Table I 80 vol. and the remaining 20 voL. is particles of the Density, hardness, and fracture toughness same compound; the whiskers have a length of 30-40 Reinforcing Volume Density H um and a diameter in the range 1-3 um. The matrix phase GPa) (MPa m/2) material was-Al2O3(AKP-30, Sumitomo Chemical Composites containing 5-40 vol. Ti(C, N)whiskers Ti(C Nw were prepared. In order to suppress alumina grain Ti(C, N)w growth, 0.25 wt. Mgo was added(by AlO3 dry Ti(C, Nw 15 weight). A slurry of Al2O3, Mg(NO3)2 6H,O, and polyacrylic acid dispersant(Dispex A40, Allied Col- Ti(C, N)w Ti(C, N)w 09999999 lords, USA)was mixed by ball milling with Al,O cylpebs as milling media in deionised water for Ti(C N)w 998 22.4 approximately 18 h. The reinforcing phase was then added to the slurry and the milling was continued for another 5 h. The slurry was then instantly frozen in liquid nitrogen and subsequently freeze-dried Table 2 Before sintering the composite mixtures were heat Thermal shock parameters: sample thickness, indentation load, and initial crack lengt reated at 600C( h)in Ar-5% H atmosphere to remove residues of Dispex A40 and to decompose Reinforcing Amount Sample Load Initial crack Mg(NO3)2- 6H,0 to Mgo before hot pressing phase (vol %) thickness (M (um) Green bodies with a diameter of o=12 mm were prepared and sintered in a hot-press furnace(Thermal TiC Nw 5 4.07 Ti(C, Nw Ti(C, Nw 2. 2. Characterisation Ti(C, Nw Ti(C, Nw The density of all sintered samples was measured by TI(C, Nw 4.17 use of Archimedes'principle. Before physical char- Ti(C, Nw 4.04 acterisation the specimens were carefully polished by Ti(C N)w standard diamond polishing techniques down to a dia mo nd particle size of The microstructure of the composites, in sections both (612 GPa). Youngs modulus for the different compo- parallel and perpendicular to the pressure directions, sites was estimated by assuming a linear relation was investigated with a scanning electron microscope between the values for Al2O3(380 GPa) and the rein- (SEM, JEOL 820). To obtain the best contrast, the forcing phase. The R-curve behaviour for two of the micrographs were recorded in back-scattered electron whisker composites was determined with loads in the mode(bse) range 35-98 N The hardness (H) and fracture toughness (Klc)at room temperature were evaluated by the Vickers inden 23. Thermal shock measurements tation technique at a load of 49 N for all compositions see table l five indents were made in a row at the mid Vickers indents were introduced into disc-shaped dle(to minimise near-surface effects)of each sample. The samples(0=12 mm, h=4 mm) with parallel surfaces, fracture toughness was calculated by the indentation one of them polished. Each indent generates four method according to Anstis et al., 6 see Eq. (I) cracks. and four indents were introduced on each sam ple, giving a total of 16 cracks per sample. The crack c3/2 () length is defined as the distance from the centre of the indent to the crack tip. To facilitate comparison between different samples, the initial crack length() A is a constant for Vickers-produced radial cracks(a was held at approximately 100 um, meaning that the value of 0.016 has been used), E is Youngs modulus, H indentation load was varied, see Table 2. The cracks is the hardness, P is the load and c is the crack length. were measured in an optical microscope (Olympus For the calculations a value for Youngs modulus of the PMG3). Each one was monitored individually, and the whiskers was estimated for a whisker composition of total crack length after thermal shock (T) was measured iCo.25 No75 by assuming a linear relation between the and the percentage crack growth(Ac)was calculated Youngs modulus values for TiC (451 GPa)and TiN using Eq (2). If one or two cracks deviated from the(SEM, JEOL 820). The whisker yield is estimated to be 80 vol.% and the remaining 20 vol.% is particles of the same compound; the whiskers have a length of 30–40 mm and a diameter in the range 1–3 mm. The matrix material was -Al2O3 (AKP-30, Sumitomo Chemical, Inc.). Composites containing 5–40 vol.% Ti(C,N) whiskers were prepared. In order to suppress alumina grain growth, 0.25 wt.% MgO was added (by Al2O3 dry weight).5 A slurry of Al2O3, Mg(NO3)2 .6H2O, and polyacrylic acid dispersant (Dispex A40, Allied Col￾loids, USA) was mixed by ball milling with Al2O3 cylpebs as milling media in deionised water for approximately 18 h. The reinforcing phase was then added to the slurry and the milling was continued for another 5 h. The slurry was then instantly frozen in liquid nitrogen and subsequently freeze-dried. Before sintering the composite mixtures were heat treated at 600
C (1 h) in Ar–5% H2 atmosphere to remove residues of Dispex A40 and to decompose Mg(NO3)2 .6H2O to MgO before hot pressing. Green bodies with a diameter of Ø=12mm were prepared and sintered in a hot-press furnace (Thermal Technology Inc.) at 1700
C, 28 MPa, for 90 min. 2.2. Characterisation techniques The density of all sintered samples was measured by use of Archimedes’ principle. Before physical char￾acterisation the specimens were carefully polished by standard diamond polishing techniques down to a dia￾mond particle size of 1 mm. The microstructure of the composites, in sections both parallel and perpendicular to the pressure directions, was investigated with a scanning electron microscope (SEM, JEOL 820). To obtain the best contrast, the micrographs were recorded in back-scattered electron mode (BSE). The hardness (H) and fracture toughness (KIc) at room temperature were evaluated by the Vickers inden￾tation technique at a load of 49 N for all compositions, see Table 1. Five indents were made in a row at the mid￾dle (to minimise near-surface effects) of each sample. The fracture toughness was calculated by the indentation method according to Anstis et al.,6 see Eq. (1). K1C ¼ A E H
1=2 P c3=2
ð1Þ A is a constant for Vickers-produced radial cracks (a value of 0.016 has been used7 ), E is Young’s modulus, H is the hardness, P is the load and c is the crack length. For the calculations a value for Young’s modulus of the whiskers was estimated for a whisker composition of TiC0.25N0.75 by assuming a linear relation between the Young’s modulus values for TiC (451 GPa) and TiN (612GPa). Young’s modulus for the different compo￾sites was estimated by assuming a linear relation between the values for Al2O3 (380 GPa) and the rein￾forcing phase. The R-curve behaviour for two of the whisker composites was determined with loads in the range 35–98 N. 2.3. Thermal shock measurements Vickers indents were introduced into disc-shaped samples (Ø=12mm, h=4 mm) with parallel surfaces, one of them polished. Each indent generates four cracks, and four indents were introduced on each sam￾ple, giving a total of 16 cracks per sample. The crack length is defined as the distance from the centre of the indent to the crack tip. To facilitate comparison between different samples, the initial crack length (l ) was held at approximately 100 mm, meaning that the indentation load was varied, see Table 2. The cracks were measured in an optical microscope (Olympus PMG3). Each one was monitored individually, and the total crack length after thermal shock (lT) was measured and the percentage crack growth (c) was calculated, using Eq. (2). If one or two cracks deviated from the Table 1 Density, hardness, and fracture toughness Reinforcing phase Volume fraction (%) Density (%) H (GPa) KIc (MPa m1=2 ) – 0 100.0 17.6 2.6 Ti(C,N)W 5 99.5 17.9 2.9 Ti(C,N)W 10 99.6 18.6 3.4 Ti(C,N)W 15 99.7 19.24.2 Ti(C,N)W 20 99.9 20.3 4.5 Ti(C,N)W 25 99.6 22.6 4.9 Ti(C,N)W 30 99.7 24.2 5.0 Ti(C,N)W 35 99.6 23.6 5.0 Ti(C,N)W 40 99.8 22.4 4.7 Table 2 Thermal shock parameters: sample thickness, indentation load, and initial crack length Reinforcing phase Amount (vol.%) Sample thickness (mm) Load (N) Initial crack length (mm) – 0 3.90 40 110 Ti(C,N)W 5 4.07 58 128 Ti(C,N)W 10 3.87 58 121 Ti(C,N)W 15 4.15 58 92 Ti(C,N)W 20 4.05 58 101 Ti(C,N)W 25 4.15 58 104 Ti(C,N)W 30 4.17 58 95 Ti(C,N)W 35 4.11 58 97 Ti(C,N)W 40 4.04 58 107 310 P. Pettersson, M. Johnsson / Journal of the European Ceramic Society 23 (2003) 309–313