JAm. Ceram.Soc,84]7677402001) journal Toughened Oxide Composites Based on Porous Alumina-Platelet Interphases Sang. Jin Lee*T and Waltraud M. Kriven** Department of Materials Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, Illinois 61801 A novel mechanism for debonding at a weak interphase in an More recently, the formation of transformation-weakened inter- all-oxide composite is introduced. This methodology involves phases that result from a negative volume change in the crystal the use of alumina platelets that have a diameter of 10-15 or structures has been demonstrated to be a viable interphase debond 5-10 um and a thickness of 1 um. The platelets induce Ing mechanism. 25-2 constrained sintering of the ceramie powder, which results in The use of porous coatings has been proposed as a universal and permanent porosity. For room-temperature properties, onl simple way to obtain weak interphases. , 4 However, several minor additions(0-3 vol %)of matrix powder yield sufficiently weak debonding interphases. The platelets lie in random, eventual closure of pores that accompanies continued sintering three-dimensional orientations and provide a de bonding mech- during prolonged operation at high temperatures. Porous mullite anism that is independent of temperature, in chemically com (3AlO3'2SiO,) and alumina(Al,O3) matrices that have been patible matrixes. Laminated composites with two types of matrixes-mullite and alumina-have been fabricated with reinforced with uncoated Al,O3 fibers show promising results for modified fibrous monoliths of alumina in a triple-layer"core/ porous composites that have relatively simple fabrication require ments. However, the extensive matrix porosity limits the overall terphase/matrix"arrangement. In the laminated systems, mechanical strength of the material and, hence, its use in load the intimate mixing of strong versus tough microstructures h been tailored by alternating va Is matrix: interphase thick- bearing applications Thus, this work addresses the need for a tough, flaw-tolerant, ness ratios. Preliminary load-displacement curves clearly all-oxide composite that is relatively dense and is capable of demonstrate characteristics of"graceful failure"and notabl improvements in the work of fracture. Scanning electron sustained performance in an oxidizing environment. The intended microscopic observation of the crack paths confirms the approach uses the phenomenon of"constrained sintering"-to viability of platelets for producing permanently porous, produce permanent porosity in an interphase that should still be debondable interphases at elevated temperatures in air. sufficiently weak to deflect a propagating crack. In this method- ology, there are competing effects of grain growth versus densifi- cation that limit the complete densification of a two-phase mixture, lting in a rigid bu orous body. In recent studies laboratory, 37-39 AL,, platelets of various ranges of size an B RITTLENESS and unreliability each still are difficult problems in volume fraction were distributed in 3AL, 0 2SiO, and zirconia of ceramics. Attempts to impart" graceful failure (ZrO2) matrixes. The resulting microstructures confirmed the which is analogous to ductility in metalshave been partially constrained-sintering effect, with the formation of a stable, uni successful with the use of composites. I Toughening is now well formly porous material in the case of the 3A1203 2SiO, matrix established to result from debonding at an interface (or within an In this paper, we propose that a suitable mixture of Al2O interphase) between a matrix and a reinforcement, or between platelets and ceramic powder, which constitutes an interphase, will laminates in a composite. -Alternatively, in the absence of be weak and porous enough to deflect a crack along the interphase reinforcements, different geometric configurations that incorporate in a composite, as illustrated schematically in Fig. 1. This weak interphase and crack-deflecting or crack-energy-dissipatin henomenon will result in crack blunting, macroscopic-crac systems have been produced. 0-2 These systems include lami deflection and frictional work to be done. which leads to overall nated composites such as silicon carbide-graphite(SiC-C)o, I toughening and flaw tolerance of the composite. This mechanism and silicon nitride-silicon nitride-silicon nitride whiskers(Si, N- has an advantage in that it is independent of temperature, thus, it (Si N(30-vol%Si N, whiskers). 2 Fibrous monoliths such as can function, for example, at temperatures up to 1700C in an Sic-C and silicon nitride/boron nitride(si3Na-BN)have also beer AL2O -rich mullite(2Al O, SiO2 ),4 or up to -1900C in a fabricated; the latter system has been studied extensively. itable matrix such as yttrium aluminate garnet(Y3AlsO12,or Other oxide systems that are based on monazite (LaPO)8-20 and YAG), yttria-stabilized zirconia(3-mol%-Y2O3-ZrO2, or 3Y related xenotime(YPO4 )2-24 interphases also have emerged. TZP), or nickel aluminate(NiAl, O4). The other salient feature of this design is that the porosity is relatively permanent, as opposed to the transient porosity that is achieved in current, fiber- reinforced, porous interphases or matrices, which are made using P. Padture--contributing editor a fugitive graphite phase. Because of the interphase porosity, is anticipated that residual-stress issues that are due to thermal mismatch between the matrix and the ase will be Manuscript No 189060 Received October ed October 31. 2000 he concept of a porous interphase already has been demon- platelets or inclusions is known to retard densification. However, the combination of these two effects to produce a weak, perma nently porous, debondable interphase is a novel approach for National University, Chyunggye-myun, Muan-gun, Chonnam, Republic of Korea. producing tough, high-temperature oxide compositesToughened Oxide Composites Based on Porous Alumina-Platelet Interphases Sang-Jin Lee* ,† and Waltraud M. Kriven** Department of Materials Science and Engineering, University of Illinois at Urbana–Champaign, Urbana, Illinois 61801 A novel mechanism for debonding at a weak interphase in an all-oxide composite is introduced. This methodology involves the use of alumina platelets that have a diameter of 10–15 or 5–10 mm and a thickness of 1 mm. The platelets induce constrained sintering of the ceramic powder, which results in permanent porosity. For room-temperature properties, only minor additions (0–3 vol%) of matrix powder yield sufficiently weak debonding interphases. The platelets lie in random, three-dimensional orientations and provide a debonding mech￾anism that is independent of temperature, in chemically com￾patible matrixes. Laminated composites with two types of matrixes—mullite and alumina—have been fabricated with modified fibrous monoliths of alumina in a triple-layer “core/ interphase/matrix” arrangement. In the laminated systems, the intimate mixing of strong versus tough microstructures has been tailored by alternating various matrix:interphase thick￾ness ratios. Preliminary load–displacement curves clearly demonstrate characteristics of “graceful failure” and notable improvements in the work of fracture. Scanning electron microscopic observation of the crack paths confirms the viability of platelets for producing permanently porous, debondable interphases at elevated temperatures in air. I. Introduction BRITTLENESS and unreliability each still are difficult problems in the use of ceramics. Attempts to impart “graceful failure”— which is analogous to ductility in metals—have been partially successful with the use of composites.1 Toughening is now well established to result from debonding at an interface (or within an interphase) between a matrix and a reinforcement, or between laminates in a composite.2–9 Alternatively, in the absence of reinforcements, different geometric configurations that incorporate a weak interphase and crack-deflecting or crack-energy-dissipating systems have been produced.10–23 These systems include lami￾nated composites such as silicon carbide–graphite (SiC–C)10,11 and silicon nitride–silicon nitride–silicon nitride whiskers (Si3N4– (Si3N4–(30-vol% Si3N4 whiskers)).12 Fibrous monoliths such as SiC–C and silicon nitride/boron nitride (Si3N4–BN) have also been fabricated; the latter system has been studied extensively.13–17 Other oxide systems that are based on monazite (LaPO4) 18–20 and related xenotime (YPO4) 21–24 interphases also have emerged. More recently, the formation of transformation-weakened inter￾phases that result from a negative volume change in the crystal structures has been demonstrated to be a viable interphase debond￾ing mechanism.25–27 The use of porous coatings has been proposed as a universal and simple way to obtain weak interphases.28,29 However, several problems still remain in fiber-reinforced composites, such as the eventual closure of pores that accompanies continued sintering during prolonged operation at high temperatures. Porous mullite (3Al2O3z2SiO2) and alumina (Al2O3) matrices that have been reinforced with uncoated Al2O3 fibers show promising results for porous composites that have relatively simple fabrication require￾ments.30 However, the extensive matrix porosity limits the overall mechanical strength of the material and, hence, its use in load￾bearing applications. Thus, this work addresses the need for a tough, flaw-tolerant, all-oxide composite that is relatively dense and is capable of sustained performance in an oxidizing environment. The intended approach uses the phenomenon of “constrained sintering”31–38 to produce permanent porosity in an interphase that should still be sufficiently weak to deflect a propagating crack. In this method￾ology, there are competing effects of grain growth versus densifi￾cation that limit the complete densification of a two-phase mixture, resulting in a rigid but porous body. In recent studies in our laboratory,37–39 Al2O3 platelets of various ranges of size and volume fraction were distributed in 3Al2O3z2SiO2 and zirconia (ZrO2) matrixes. The resulting microstructures confirmed the constrained-sintering effect, with the formation of a stable, uni￾formly porous material in the case of the 3Al2O3z2SiO2 matrix. In this paper, we propose that a suitable mixture of Al2O3 platelets and ceramic powder, which constitutes an interphase, will be weak and porous enough to deflect a crack along the interphase in a composite, as illustrated schematically in Fig. 1. This phenomenon will result in crack blunting, macroscopic-crack deflection, and frictional work to be done, which leads to overall toughening and flaw tolerance of the composite. This mechanism has an advantage in that it is independent of temperature; thus, it can function, for example, at temperatures up to 1700°C in an Al2O3-rich mullite (2Al2O3zSiO2) 40,41 or up to ;1900°C in a suitable matrix such as yttrium aluminate garnet (Y3Al5O12, or YAG), yttria-stabilized zirconia (3-mol%-Y2O3–ZrO2, or 3Y￾TZP), or nickel aluminate (NiAl2O4). The other salient feature of this design is that the porosity is relatively permanent, as opposed to the transient porosity that is achieved in current, fiber￾reinforced, porous interphases or matrices, which are made using a fugitive graphite phase.30 Because of the interphase porosity, it is anticipated that residual-stress issues that are due to thermal￾expansion mismatch between the matrix and the interphase will be minimized. The concept of a porous interphase already has been demon￾strated to be a viable debonding mechanism, and the use of platelets or inclusions is known to retard densification. However, the combination of these two effects to produce a weak, perma￾nently porous, debondable interphase is a novel approach for producing tough, high-temperature oxide composites. N. P. Padture—contributing editor Manuscript No. 189060. Received October 25, 1999; approved October 31, 2000. This work was supported by the Argonne National Laboratory, with funding from the Defense Advanced Research Projects Agency (DARPA), through a Department of Energy Interagency Agreement, under Contract W-31-109-Eng-38. *Member, American Ceramic Society. **Fellow, American Ceramic Society. † Present address: Department of Materials Science and Engineering, Mokpo National University, Chyunggye-myun, Muan-gun, Chonnam, Republic of Korea. J. Am. Ceram. Soc., 84 [4] 767–74 (2001) 767 journal