October 1997 Fibrous Monolithic ceramics 2485 1000 BN/SiN Fibrous Monolith 120014(N behave similarly to conventional layered materials. It is there- Shear failure is favored when the interfacial fracture resistance fore interesting to compare and contrast their properties is low or when the span-to-depth ratio of the specimen is large (A Strength: The strength of fibrous monoliths and cor When the interfacial fracture resistance is higher or when the ventional layered materials made from the same base materials span-to-depth ratio is lower, failure occurs when the tensile and with the same compositions are comparable, 11 35.51 It is stress exceeds a critical value. The strength is determined by expected that the strength of layered materials would be the orientation and strength of the individual cells with respect slightly higher in flexure because more of the stiffer, load- to the loading axis. When cells are oriented parallel or nearly earing phase is farther from the neutral axis. However, for arallel to the loading axis, the strength is determined primarily pecimens containing many layers, this effect is negligible. A by the strength of the SiN, cells. If, however, cells are oriente more important influence on strength comes from the fact that perpendicular or nearly perpendicular to the loading axis, fail- the strength of layered materials is determined by the largest e is determined by the strength of the BN cell boundaries flaw anywhere on the surface of the specimen. In contrast, for The energy absorption capacity of fibrous monoliths is a fibrous monoliths, a single, large flaw on the tensile surface result of two energy dissipation mechanisms: cracking and fric- causes failure of only a single cell. The peak load is achieved tional sliding. Both of these mechanisms are more effective when the failure of individual cells accumulates to the point when extensive delamination occurs prior to fracture of the where a layer of cells no longer can bear the applied load, and individual cells. Long delamination distances are achieved only the failure process is similar to the failure of a bundle of fila- when delamination cracks remain in the Bn interphase and do ments. Thus, compared to layered materials, the strength of not kink into the surrounding Si3 N4 cells. It has been shown fibrous monoliths should be less sensitive to defects that are present on the specimen surface. Experimental evidence tance is high and/or large flaws are present in the Si3, cells firms that defects, such as indentations, have little effect e A model has been verified that successfully predicts the values trength of fibrous monolithic ceramics, whereas they have a of the flaw size and interfacial fracture resistanc arge effect on the strength of conventional, two-dimensional laminates. 5 sliding plays an important role in dissipating energy during the (B) Energy Absorption: Both layered materials and fi- fracture process. Our calculations and measurement brous monoliths rely on the creation of interfacial crack area that approximately one half of the energy dissipation capacity and frictional sliding to dissipate energy. However, because of of fibrous monoliths results from frictional slidin the cellular nature of the architecture there is significantly The insight we have gained from uniaxially aligned fibrous more interfacial crack area per unit volume in fibrous monoliths has allowed us to design fibrous monoliths with a hs compared to layered materials. Thus, more crack variety of multiaxial architectures. This has led to the ability to created in fibrous monoliths during fracture, and, once cra urs, there are more sliding interfaces to dissipate energy Thus, it is expected that the energy absorption capability of roperties and the load-deflection response of these material fibrous monoliths should exceed that of layered materials. Ex- have been presented and verified. Typical properties for ar perimental results confirm that the work-of-fracture is 30% architecture that exhibits in-plane elastic isotropy are a strengt 0% higher for fibrous monoliths compared to layered mate- of 285 MPa and a work-of-fracture of 4600 J/m2 rials made from the same materials Acknowledgments: The authors thank Advanced Ceramics Research, Tucson, AZ, for providing green material. We also thank Brady manufacturing some of the specimens used in this study In many respects, the fracture process in fibrous monoliths is similar to that of laminates. A modified laminate theory, used to predict the elastic response of fibrous monoliths, can be used to calculate the stress within any layer of cells when specimens ondon,282,508-20(1964) are loaded in flexure. This stress determines the mechanism of T. W. Button, and J D. birchall failure that is observed at both room and elevated temperatures. 455-57(1990) Nature(London), 357 1Oct. 4behave similarly to conventional layered materials. It is there￾fore interesting to compare and contrast their properties. (A) Strength: The strength of fibrous monoliths and con￾ventional layered materials made from the same base materials and with the same compositions are comparable.11,35,51 It is expected that the strength of layered materials would be slightly higher in flexure because more of the stiffer, load￾bearing phase is farther from the neutral axis. However, for specimens containing many layers, this effect is negligible. A more important influence on strength comes from the fact that the strength of layered materials is determined by the largest flaw anywhere on the surface of the specimen. In contrast, for fibrous monoliths, a single, large flaw on the tensile surface causes failure of only a single cell. The peak load is achieved when the failure of individual cells accumulates to the point where a layer of cells no longer can bear the applied load, and the failure process is similar to the failure of a bundle of fila￾ments. Thus, compared to layered materials, the strength of fibrous monoliths should be less sensitive to defects that are present on the specimen surface. Experimental evidence con￾firms that defects, such as indentations, have little effect on the strength of fibrous monolithic ceramics,6 whereas they have a large effect on the strength of conventional, two-dimensional laminates.52 (B) Energy Absorption: Both layered materials and fi￾brous monoliths rely on the creation of interfacial crack area and frictional sliding to dissipate energy. However, because of the cellular nature of the architecture, there is significantly more interfacial crack area per unit volume in fibrous mono￾liths compared to layered materials. Thus, more crack area is created in fibrous monoliths during fracture, and, once cracking occurs, there are more sliding interfaces to dissipate energy. Thus, it is expected that the energy absorption capability of fibrous monoliths should exceed that of layered materials. Ex￾perimental results confirm that the work-of-fracture is 30%– 50% higher for fibrous monoliths compared to layered mate￾rials made from the same materials. VII. Conclusions In many respects, the fracture process in fibrous monoliths is similar to that of laminates. A modified laminate theory, used to predict the elastic response of fibrous monoliths, can be used to calculate the stress within any layer of cells when specimens are loaded in flexure. This stress determines the mechanism of failure that is observed at both room and elevated temperatures. Shear failure is favored when the interfacial fracture resistance is low or when the span-to-depth ratio of the specimen is large. When the interfacial fracture resistance is higher or when the span-to-depth ratio is lower, failure occurs when the tensile stress exceeds a critical value. The strength is determined by the orientation and strength of the individual cells with respect to the loading axis. When cells are oriented parallel or nearly parallel to the loading axis, the strength is determined primarily by the strength of the Si3N4 cells. If, however, cells are oriented perpendicular or nearly perpendicular to the loading axis, fail￾ure is determined by the strength of the BN cell boundaries. The energy absorption capacity of fibrous monoliths is a result of two energy dissipation mechanisms: cracking and fric￾tional sliding. Both of these mechanisms are more effective when extensive delamination occurs prior to fracture of the individual cells. Long delamination distances are achieved only when delamination cracks remain in the BN interphase and do not kink into the surrounding Si3N4 cells. It has been shown that such crack kinking occurs if the interfacial fracture resis￾tance is high and/or large flaws are present in the Si3N4 cells. A model has been verified that successfully predicts the values of the flaw size and interfacial fracture resistance necessary to avoid crack kinking. It also has been shown that frictional sliding plays an important role in dissipating energy during the fracture process. Our calculations and measurements indicate that approximately one half of the energy dissipation capacity of fibrous monoliths results from frictional sliding. The insight we have gained from uniaxially aligned fibrous monoliths has allowed us to design fibrous monoliths with a variety of multiaxial architectures. This has led to the ability to design and manufacture fibrous monoliths with arbitrary archi￾tectures for specific applications. Models to predict the elastic properties and the load–deflection response of these materials have been presented and verified. Typical properties for an architecture that exhibits in-plane elastic isotropy are a strength of 285 MPa and a work-of-fracture of 4600 J/m2 . Acknowledgments: The authors thank Advanced Ceramics Research, Tucson, AZ, for providing green material. We also thank G. Allen Brady for manufacturing some of the specimens used in this study. References 1 J. Cook and J. E. Gordon, ‘‘A Mechanism for the Control of Crack Propa￾gation in All-Brittle Systems,’’ Proc. R. Soc. London, 282, 508–20 (1964). 2 W. J. Clegg, K. Kendall, N. McN. Alford, T. W. Button, and J. D. Birchall, ‘‘A Simple Way to Make Tough Ceramics,’’ Nature (London), 357 [Oct. 4] 455–57 (1990). Fig. 22. Strength is plotted versus test temperature for monolithic Si3N4 and for fibrous monoliths. In both cases, 6 wt% Y2O3 and 2 wt% Al2O3 were added to the Si3N4 as a sintering aid. Overlapping data at 1000°C has been offset slightly for clarity. October 1997 Fibrous Monolithic Ceramics 2485