K.K. Chawla et al. Journal of the European Ceramic Society 20(2000)551-559 strain A/r is about 0.00011. Thermal strain AoAT in 4. Conclusions Saphikon/monazite interface is equal to 0.0008 if AT=1000.C. Comparatively, the thermal strain is From the results and discussions presented in this about 8 times greater than the roughness-induced strain. present work, we can make the following conclusions Thus, the effect of interfacial roughness of monazite/ Saphikon fiber would not be significant. Specifically 1. The incorporation of monazite interphase coating when compared with that of monazite/polycrystalline by sol-gel dip coating method is an effective way alumina, the roughness-induced radial clamping strain of creating weak interfacial bonds between mon- at the monazite/Saphikon interface is very small azide an umina 2. Two-phase layered liquid dipping method for 3.2. Thermal stress analysis coating monofilament Saphikon fibers with mon- azite sol was effective We used a three-layer model to calculate the thermal 3. The presence of monazite as an interphase was stresses for the five- and 10-dip coated composites, uccessful in providing weak interface to both sin- respectively. A 0.01 fiber volume fraction we used a gle-crystal alumina and polycrystalline alumina. 1000C temperature change and 0.5 and 1 um coating 4. The roughness-induced clamping was much thickness for five- and 10-dip, respectively. The distribu greater at the rough monazite/polycrystalline alu tion of thermal stresses in Saphikon fiber/Lapo/alu mina interface as compared with the smooth single mina, matrix composite for two coating thicknesses is crystal alumina/monazite interface. Therefore, the Figs. 10 and 11. the difference between the Saphikon fiber/monazite interface was relatively two figures is insignificant. One can see that all the stress weak and interfacial debonding and fiber pullout components are constant in the central fiber. The o, is were easier to initiate at this interface than at the constant in the coating as well as the matrix. The e is polycrystalline alumina/monazite interface. It did discontinuous at the interface the o maintains continuity not matter in which phase the crack originated at the interface, and goes to zero at the free surface 5. A sinusoidal variation in the five-dip shear stress- This analysis shows radial gripping increased the displacement curve due to the sinusoidal asperities strength of the fiber/monazite and monazite/matrix on fiber surface was observed interfaces. However, the magnitude of the thermal 6. The energy expended in interfacial debonding radial stress is low and gripping due to thermal stress is ading to fiber pullout caused an increase not significant. The total radial stress is the sum of the toughness or work of fracture of such composite hermal stress and the stress due to the interfacial materials roughness. The roughness-induced radial stress is much 7. Thermal stress analysis showed radial gripping, greater at the monazite/polycrystalline alumina inter- which increased the strength of the interfaces face than at the single-crystal alumina/monazite inter However, the magnitude was not significant face, although they are both mechanically bonded. This because of the small thickness of coating probably is the reason why debonding occurred more often at the fiber/monazite interface instead of the monazite/matrix interface References The present work has demonstrated the effective interface engineering approach by sol-gel dip coating L. Prewo, K. M. and Batt, J.A., The oxidative stability of carbon method. There are, however, some unanswered ques fiber reinforced glass-matrix composites. J. Mater. Sci., 1988, 23 tions. a detailed study on the coating thickness effects 523-527 on the debonding behavior is necessary. More uniform 2. Chawla, K. K, Ceramic Matrix Composite Champion and Hall, coating thickness should be obtained. Measurements of interface shear stress as a function of thickness should 3. Chawla, K.K., Composite Materials, 2nd ed. Springer Verlag, New York. 1998 be done in a systematic manner. We also recognize that 4. Chawla. K.K. Ferber. M. K. Xu. Z.R. and venkatesh. R. very low fiber volume fraction was used in present work Interface engineering in alumina/glass composites. Materials Sci- (0.001)due to the high cost of the single-crystal alumina ence and Engineering. 1993.A 162. 35-44. fiber and small amount of monazite precursor sol. We 5. Chawla, K. K. Schneider, H. Schmuicker. M. and Xu. Z.R. in suggest increasing fiber volume fraction and decreasing Processing and Design Issues in High Temperature Materials, TMS, Warrendale, PA, 1997, pp 235-245 fiber diameter to create statistically better measurements 6. Morgan, P. E. D and Marshall, D. B, Functional interfaces for One practical way is to use NextelTM 610 oxide alumina oxide/oxide composites. Mater. Sci. Eng, 1993, A162, 15-25. fibers instead, which are less expensive. Also, because of 7. Morgan, P. E D. and Marshall, D. B. Ceramic composites of the ambiguous estimates of interfacial energy for LaPO4 monazite and alumina. J. m Ceram. Soc.. 1995 78. 1553-1563. 8. Morgan, P.E. D, Marshall, D. B. and Housley, R. M, High 12O3 interface and for LaPO4/ single crystal AlO3, future temperature stability of monazite-alumina composites. Mater work to measure these values would be useful Sci.Eng.1995,A195,215-22strain A/r is about 0.00011. Thermal strain
a
T in Saphikon/monazite interface is equal to 0.0008 if
T=1000C. Comparatively, the thermal strain is about 8 times greater than the roughness-induced strain. Thus, the e€ect of interfacial roughness of monazite/ Saphikon ®ber would not be signi®cant. Speci®cally, when compared with that of monazite/polycrystalline alumina, the roughness-induced radial clamping strain at the monazite/Saphikon interface is very small. 3.2. Thermal stress analysis We used a three-layer model2 to calculate the thermal stresses for the ®ve- and 10-dip coated composites, respectively. A 0.01 ®ber volume fraction we used a 1000C temperature change and 0.5 and 1 mm coating thickness for ®ve- and 10-dip, respectively. The distribu￾tion of thermal stresses in Saphikon ®ber/LaPO4/alu￾mina, matrix composite for two coating thicknesses is shown in Figs. 10 and 11. The di€erence between the two ®gures is insigni®cant. One can see that all the stress components are constant in the central ®ber. The z is constant in the coating as well as the matrix. The y is discontinuous at the interface. The r maintains continuity at the interface, and goes to zero at the free surface. This analysis shows radial gripping increased the strength of the ®ber/monazite and monazite/matrix interfaces. However, the magnitude of the thermal radial stress is low and gripping due to thermal stress is not signi®cant. The total radial stress is the sum of the thermal stress and the stress due to the interfacial roughness.2 The roughness-induced radial stress is much greater at the monazite/polycrystalline alumina inter￾face than at the single-crystal alumina/monazite inter￾face, although they are both mechanically bonded. This probably is the reason why debonding occurred more often at the ®ber/monazite interface instead of the monazite/matrix interface. The present work has demonstrated the e€ective interface engineering approach by sol±gel dip coating method. There are, however, some unanswered ques￾tions. A detailed study on the coating thickness e€ects on the debonding behavior is necessary. More uniform coating thickness should be obtained. Measurements of interface shear stress as a function of thickness should be done in a systematic manner. We also recognize that very low ®ber volume fraction was used in present work (0.001) due to the high cost of the single-crystal alumina ®ber and small amount of monazite precursor sol. We suggest increasing ®ber volume fraction and decreasing ®ber diameter to create statistically better measurements. One practical way is to use NextelTM 610 oxide alumina ®bers instead, which are less expensive. Also, because of the ambiguous estimates of interfacial energy for LaPO4/ Al2O3 interface and for LaPO4/single crystal Al2O3, future work to measure these values would be useful. 4. Conclusions From the results and discussions presented in this present work, we can make the following conclusions: 1. The incorporation of monazite interphase coating by sol±gel dip coating method is an e€ective way of creating weak interfacial bonds between mon￾azite and alumina. 2. Two-phase layered liquid dipping method for coating mono®lament Saphikon ®bers with mon￾azite sol was e€ective. 3. The presence of monazite as an interphase was successful in providing weak interface to both sin￾gle-crystal alumina and polycrystalline alumina. 4. The roughness-induced clamping was much greater at the rough monazite/polyerystalline alu￾mina interface as compared with the smooth single crystal alumina/monazite interface. Therefore, the Saphikon ®ber/monazite interface was relatively weak and interfacial debonding and ®ber pullout were easier to initiate at this interface than at the polycrystalline alumina/monazite interface. It did not matter in which phase the crack originated. 5. A sinusoidal variation in the ®ve-dip shear stress± displacement curve due to the sinusoidal asperities on ®ber surface was observed. 6. The energy expended in interfacial debonding leading to ®ber pullout caused an increase in toughness or work of fracture of such composite materials. 7. Thermal stress analysis showed radial gripping, which increased the strength of the interfaces. However, the magnitude was not signi®cant because of the small thickness of coating. References 1. Prewo, K. M. and Batt, J. A., The oxidative stability of carbon ®ber reinforced glass-matrix composites. J. Mater. Sci., 1988, 23, 523±527. 2. Chawla, K. K., Ceramic Matrix Composite. Champion and Hall, London, 1993. 3. Chawla, K. K., Composite Materials, 2nd ed. Springer Verlag, New York, 1998. 4. Chawla, K. K., Ferber, M. K., Xu, Z. R. and Venkatesh, R., Interface engineering in alumina/glass composites. Materials Sci￾ence and Engineering, 1993, A 162, 35±44. 5. Chawla, K. K., Schneider, H., SchmuÈcker, M. and Xu, Z. R. in Processing and Design Issues in High Temperature Materials, TMS, Warrendale, PA, 1997, pp. 235±245. 6. Morgan, P. E. D. and Marshall, D. B., Functional interfaces for oxide/oxide composites. Mater. Sci. Eng., 1993, A162, 15±25. 7. Morgan, P. E. D. and Marshall, D. B., Ceramic composites of monazite and alumina. J. Am. Ceram. Soc., 1995, 78, 1553±1563. 8. Morgan, P. E. D., Marshall, D. B. and Housley, R. M., High￾temperature stability of monazite±alumina composites. Mater. Sci. Eng., 1995, A195, 215±222. 558 K.K. Chawla et al. / Journal of the European Ceramic Society 20 (2000) 551±559