2194 Journal of the American Ceramic Society--Liversage et al Vol. 90. No. 7 32-1 Fig 8. The diffraction pattern for ST100 showing the peaks corre- sponding to the khamrabaevite structure and apparent oxide phase Fig 10. Thermal sion coefficients as a function of the Tic The numbers in the legend refer to the applicable JCPDS numbers fraction showing a clear linear trend within the expected experimental accuracy nclusions would have lead to discrepancies as extensive as IV. Conclusions actually observed between the present data and previously ublished data. It is unclear at this stage as to the origin of the use of sintering aids in the present laminated Sic-TiC sys- differences between the elastic property values reported here and in conjunction with an applied pressure during sintering, hose observed by Cho et al. and Sand et al., both of whom was found to result in well-densified materials and was found to determined their values in an LPS-ST system, using similar ad ot result in the formation of any additional phases. With the ditives and additive fractions onfirmed absence of additional phase formation between the Thermal expansion coefficients for different TiC-fractions, as SiC and TiC components, one is assured that the system is well determined from thermal dilation measurements between 200% suited for the fabrication of a functionally graded multilayer and 750C, have been presented in Fig. 10. The dilation curve material. This is because the properties of the composite will were found to be strongly linear over this temperature range and ake on values that are dependent on the properties of the con herefore the thermal expansion coefficients were determined stituent phases in a semi-predictable way, and will not be infit directly from the slope of a linear regression fit over the entire enced by the formation of additional ternary phases. In this temperature range of dilation data. Also shown in Fig. 10 egard it is a simple matter to tailor any property variations linear regression of the thermal expansion data, the plac within a functionally graded material, particularly the thermal ement hich suggests an approximately linear trend in the thermal ex oefficient, which was found to a reasonable approx TiC fraction. In imation as being linea dependent on the TiC fraction. One of contrast with the elastic property data, the thermal expansion the shortcomings of the technique used to produce both the data was found to be very much in agreement with values quot layers and the resultant laminated materials is associated with ed for similar SiC-TiC syst 10. the development of appreciable amounts of carbonaceous in- Once again, however, the clusions. It was speculated that these inclusions were a result of occurrence of an anisotropically distorted microstructure and an additional carbonaceous phase may ncomplete removal of the organic additives used in the fabri ve con any slight differences seen between the values herein and those ation of the ceramic tapes. This is rted by the observed quoted by others absence of such inclusions in sintered Sic and TiC materia produced in an identical manner but without making use of or- ganic-binder-based slurries before sintering. Consequently, it ry to out an optimization exercise on the ● Elastic modulus D Poisson ratio binder burnout process before the transfer of the technology into the production of graded multilayer materials References M. B. Bever and P. F. Duwez,"Gradients in Composite Materials. "Mater. Sci. Eng tionally Graded Materials, Mater. Sci. Eng BG.C. Wei and P. F. Becher, "Improvements in Mechanical Properties in Sic by 0,1 .S.-M Dong D.L. Jiang, S.-H Tan, and J.-K Guo. "Mechanical SiC/TiC Composites by Hot Isostatic Pressing, "J. Mater. Sci. L 知mHmM工Mms Fig 9. Youngs moduli and the Poisson ratio dependence on the TiC Sic and of SiC-TiC Composites, " Am. Ceram. Soc. Bull, 65(21 3-these inclusions would have lead to discrepancies as extensive as those actually observed between the present data and previously published data. It is unclear at this stage as to the origin of the differences between the elastic property values reported here and those observed by Cho et al. 10 and Sand et al.,12 both of whom determined their values in an LPS–ST system, using similar ad￾ditives and additive fractions. Thermal expansion coefficients for different TiC-fractions, as determined from thermal dilation measurements between 2001 and 7501C, have been presented in Fig. 10. The dilation curves were found to be strongly linear over this temperature range and therefore the thermal expansion coefficients were determined directly from the slope of a linear regression fit over the entire temperature range of dilation data. Also shown in Fig. 10 is a linear regression of the thermal expansion data, the placement of which suggests an approximately linear trend in the thermal ex￾pansion coefficient with respect to a changing TiC fraction. In contrast with the elastic property data, the thermal expansion data was found to be very much in agreement with values quot￾ed for similar SiC–TiC systems.10,12 Once again, however, the occurrence of an anisotropically distorted microstructure and an additional carbonaceous phase may have contributed to any slight differences seen between the values herein and those quoted by others. IV. Conclusions The use of sintering aids in the present laminated SiC–TiC sys￾tem, in conjunction with an applied pressure during sintering, was found to result in well-densified materials, and was found to not result in the formation of any additional phases. With the confirmed absence of additional phase formation between the SiC and TiC components, one is assured that the system is well suited for the fabrication of a functionally graded multilayer material. This is because the properties of the composite will take on values that are dependent on the properties of the con￾stituent phases in a semi-predictable way, and will not be influ￾enced by the formation of additional ternary phases. In this regard it is a simple matter to tailor any property variations within a functionally graded material, particularly the thermal expansion coefficient, which was found to a reasonable approx￾imation as being linearly dependent on the TiC fraction. One of the shortcomings of the technique used to produce both the layers and the resultant laminated materials is associated with the development of appreciable amounts of carbonaceous in￾clusions. It was speculated that these inclusions were a result of incomplete removal of the organic additives used in the fabri￾cation of the ceramic tapes. This is supported by the observed absence of such inclusions in sintered SiC and TiC material produced in an identical manner but without making use of or￾ganic-binder-based slurries before sintering. Consequently, it may be necessary to carry out an optimization exercise on the binder burnout process before the transfer of the technology into the production of graded multilayer materials. References 1 M. B. Bever and P. F. Duwez, ‘‘Gradients in Composite Materials,’’ Mater. Sci. Eng., 10, 1–8 (1972). 2 B. Kieback, A. Neubrand, and H. Riedel, ‘‘Processing Techniques for Func￾tionally Graded Materials,’’ Mater. Sci. Eng., A362, 81–106 (2003). 3 G. C. Wei and P. F. Becher, ‘‘Improvements in Mechanical Properties in SiC by the Addition of TiC Particles,’’ J. Am. Ceram. Soc., 67 [8] 571–4 (1984). 4 H. Endo, M. Ueki, and H. Kubo, ‘‘Microstructure and Mechanical Properties of Hot-Pressed SiC–TiC Composites,’’ J. Mater. Sci., 26, 3769–74 (1991). 5 S.-M Dong, D.-L. Jiang, S.-H Tan, and J.-K Guo, ‘‘Mechanical Properties of SiC/TiC Composites by Hot Isostatic Pressing,’’ J. Mater. Sci. Lett., 15, 394–6 (1996). 6 D.-L. Jiang, J.-H. Wang, Y.-L. Li, and L.-T. Ma, ‘‘Studies on the Strength￾ening of Silicon Carbide-Based Multiphase Ceramics I: The SiC–TiC System,’’ Mater. Sci. Eng., A109, 401–6 (1989). 7 M. A. Janney, ‘‘Microstructure Development and Mechanical Properties of SiC and of SiC–TiC Composites,’’ Am. Ceram. Soc. Bull., 65 [2] 357–62 (1986). Fig. 8. The diffraction pattern for ST100 showing the peaks corre￾sponding to the khamrabaevite structure and apparent oxide phases. The numbers in the legend refer to the applicable JCPDS numbers. ν Fig. 9. Young’s moduli and the Poisson ratio dependence on the TiC fraction. ° Fig. 10. Thermal expansion coefficients as a function of the TiC fraction showing a clear linear trend within the expected experimental accuracy. 2194 Journal of the American Ceramic Society—Liversage et al. Vol. 90, No. 7