STUDIES ON DUCTILE PHASE TOUGHENING OF MOLYBDENUM DISILICIDE BASED COMPOSITES

Abstract

Silicides are a new class of engineering materials, which offers the advantages of a ceramic as well as of a metal. Out of a large number of silicide compounds known, MoSi2 has been proposed as a model material for high temperature structural applications due to its unique physical and mechanical properties. It is brittle like ceramics at room temperature and undergoes plastic deformation and creep like metals at high temperatures. It has recently become a potential candidate material for several advanced high temperature aerospace applications. The current and potential applications of MoSi2 and its composites include heating elements, industrial gas burners, molten metal lances, components requiring contact with molten glasses, diesel engine glow plugs and components for aerospace gas turbine engine. The most promising materials for high temperature structural applications include Ni base superalloys, the aluminides of nickel, titanium and iron, Si base ceramics like SiC, Si3N4, SiC-SiC composites and aluminides of refractory metals. However, Ni base superalloys have a relatively higher density. Nickel, titanium and iron aluminides have melting points in the range of 1400-1600 °C. This limits the use of these materials to temperatures less than 1200 °C, although density wise these aluminides look potentially attractive. In addition to their relatively lower melting points, aluminides have poor oxidation resistance above 650 °C. Si-base ceramics are brittle over the entire temperature range, while refraCtory metal aluminides are brittle at room temperature and have low strength and creep resistance at required high temperatures. MoSi2 has the potential of meeting structural and oxidation requirements up to a temperature of 1600 °C and therefore has been proposed as an alternative to structural ceramics. Efforts to develop MoSi2 have been hampered by its extreme brittleness at temperatures below 1000 °C, coupled with relatively low creep resistance. For its effective use as a high temperature structural material, it becomes necessary to toughen the material at lower temperatures (within the MoSi2 brittle regime) while simultaneously improving the strength at higher temperatures. It has ii been recognized by most workers in the field that the best approach to solve these twin problems is to develop MoSi2 matrix composites with a variety of reinforcements and its alloying with other elements. MoSi2 is a line (no off stoichiometry) compound and does not easily alloy with other elements. WSi2 has been added due to its similarity in crystal structure and lattice parameters with MoSi2. WSi2 alloying leads to considerable improvements in high temperature strength and creep resistance but its effects on room temperature ductility are found to be very limited (Flinn et al. 1989, Petrovic and Honnell 1990, Frankwicz et al. 1993). MoSi2 can be engineered by addition of suitable reinforcements (brittle ceramics as well as ductile metals) to improve its mechanical properties. Ceramic reinforcements are aimed to improve its high temperature strength and creep resistance while the metallic reinforcements may be effective to improve its low temperature fracture toughness. Extensive studies have been carried out on MoSi2 matrix composites reinforced with high strength ceramics such as SiC (Gac and Petrovic 1985, Petrovic and Honnell 1990, Gibbs et al. 1987, Yang and Jeng 1990, Bhattacharya and Petrovic 1991), TiC (Yang et al., 1989), Si3N4 (Petrovic and Honnell 1990, Hebsur 1994), A1203 (Tuffe et al., 1993), and ZrO2 (Petrovic and Honnell 1990, Bhattacharya and Petrovic 1991, Petrovic et al. 1991). At high temperatures they improve creep resistance by inhibiting the excessive dislocation motion. Although the ceramic reinforcements result in considerable improvements in mechanical properties of MoSi2, their effect on improving the room temperature fracture toughness of MoSi2 is marginal. Only moderate room temperature toughening effects are derived with the addition of ceramic reinforcements. Almost all components used for high temperature structural applications are subjected to some degree of stresses at low temperatures, which could be catastrophic for materials having very poor toughness at lower temperatures. The room temperature fracture toughness (KO of MoSi2 has been reported to be in the range of 3-4 MPal/rn. Since a fracture toughness level of '12 to 15 MPaJrn is desirable for possible structural applications, other types of reinforcements, e.g., ductile phases need to be explored. The ductile phase may provide toughness at room and intermediate temperatures while the matrix provides for oxidation resistance. iii Ductile phase toughening of MoSi2 was proposed originally by V. D. Kristic (Kristic et al., 1981). It was first investigated by Fitzer and Remmele (1985). Other silicides, such as Nb5Si3, have exhibited dramatic increase in toughness when a pure Nb phase was incorporated in it. However, there are very limited fracture toughness data on the use of ductile reinforcements in MoSi2 matrix. The candidate ductile reinforcements are various refractory metals like Nb, Ta, Mo and W. Most of the work reported till date is on the use of Nb as a ductile reinforcement (Xiao 1991, Xiao and Abbaschian 1992, Alman and Stoloff 1995, Alman et al. 1992, Venkateswara Rao et al. 1992). The success achieved in improving the room temperature fracture toughness of MoSi2 by incorporating Nb motivates further studies on understanding the effect of other ductile reinforcements on microstructure and mechanical properties of MoSi2. The present study aims to explore and investigate the toughening of MoSi2 by different ductile phases. In the present investigation, an attempt has been made to understand the role of various ductile refractory metals like tungsten, molybdenum, tantalum and niobium, in toughening of MoSi2 matrix. A comparative study of their compatibility with MoSi2 and their effect on its room temperature mechanical behaviour has been carried out. The composites in the present investigation were made by Powder Metallurgy (PM) techniques using vacuum hot pressing (VHP). The ductile refractory metals were incorporated in MoSi2 matrix in particulate form (discontinuously reinforced) as well as in foil form (laminated approach). The observations and results obtained in this work have further added to the existing information base and thus results in enhanced understanding of the ductile phase toughening of MoSi2. Chapter-1 contains an introduction to the present study in the context of technological importance of the material. Chapter-2 begins with the structure and properties of MoSi2 including its historical background, thermo-physical and mechanical properties, slip systems, oxidation resistance and its potential applications. A comprehensive survey of literature has been carried out to understand critically the existing knowledge about various approaches to improve the mechanical properties of MoSi2 for its use as a high temperature structural material. These include alloying with other elements and iv reinforcing with ceramics to create composites. Literature survey has revealed that the idea of incorporating ductile phases in MoSi2 for improvement in its room temperature fracture toughness has found very limited attention. Therefore, the problem under present investigation i.e., ductile phase toughening of MoSi2 has been formulated. An exhaustive literature survey has been carried out on toughening mechanisms for brittle materials, which revealed that numerous concepts have been proposed for toughening of ceramic matrix composites often leading to confusion. Many of these concepts differ with each other only marginally and use different terminology to describe the similar toughening mechanisms. Therefore, an attempt has been made to regroup various toug hening mechanisms for brittle materials bringing out the exclusive and salient features of each one of them and is presented in a brief and systematic manner. Chapter-3 presents the experimental procedure and techniques employed for characterization of raw materials, processing of various MoSi2 based particulate and laminated composites, microstructural characterization and evaluation of room temperature mechanical behaviour of all the composite systems prepared in the present study. Various combinations of matrix (MoSi2 based) and the reinforcements (different ductile refractory metals) were used to prepare a variety of composite systems. Pure MoSi2, MoSi2 + 2 wt% Al and MoSi2 + 20 vol% SiC, were used as the matrix materials with W, Mo, Ta and Nb as reinforcements in particulate as well as in foil form to synthesize various particulate and laminated composites. Nb foil was used with and without application of an inert A1203 coating. Nb foil was also used in embrittled condition after allowing it to undergo partial oxidation during the processing. The various composite materials were processed by powder metallurgy route. The monolithic matrix materials were also produced by the same processing methods to provide a reference material against which the properties of the composites could be compared. Vacuum hot pressing was used as the main technique to consolidate the matrix and the reinforcement components together to make the different particulate and laminated composites. In the present investigation, the work was restricted to only model "tri-layer" laminates. The microstructure] characterization was carried out using optical microscopy (with and without polarized light), x-ray diffraction analysis (XRD), electron probe micro analysis (EPMA) and scanning electron microscopy (SEM). The various mechanical properties evaluated included density, elastic modulus, hardness, indentation fracture toughness, flexural strength and fracture toughness. In case of laminated composites, the flexural strength and fracture toughness were measured in crack arrester as well as in crack divider modes by a three-point bend test (SENB test specimen). Apart from the stress intensity factor based fracture toughness, energy driven fracture toughness in terms of 'work of fracture' was also evaluated. Micro-hardness measurements were made at the interface regions between matrix and the reinforcement to understand the mechanical characteristics of the various interfacial reaction products. The metallic foils used as the reinforcements were characterized for their hardness and tensile properties. Cracks generated by the hardness indentations and the crack paths were observed using optical and scanning electron microscopes to have an understanding of the crack propagation and interactions with different reinforcing phases such as SiC particulates and refractory metal foils. This led to an enhanced understanding of the toughening micro mechanisms. The fracture surfaces of the tested specimens were studied using scanning electron microscope. The unbroken specimens were also observed in scanning electron microscope as well as under the stereomicroscope to study the macroscopic crack path behaviour. Residual thermal stresses are very important in all composite materials. The high processing temperatures involved and a very low strain to fracture make the residual stress problems more severe in ceramic matrix composites. The subject of residual thermal stresses is an area, which has often been neglected in the development of new structural materials. In the present work, an attempt has been made to understand the nature and magnitude of residual thermal stresses in these composites by analytical methods as well as by FEM analysis. Chapter-4 describes in detail the analysis of residual thermal stresses carried out for MoSi2 + 20 vol% SiC particulate composite and various laminated composites prepared using MoSi2 + 20 vol% SiCp as the matrix layer. As the residual thermal stresses may significantly affect the mechanical behaviour of a composite material, a quantitative idea of residual stresses from the results obtained in the present study was found to be very useful to understand the ductile phase toughening behaviour of MoSi2 by different refractory metal foils. It was found that Nb foil reinforced composite has the lowest residual thermal stresses owning to its minimum thermal expansion mismatch with the matrix layer. The results obtained by FEM analysis were found quite similar to the values vi obtained by analytical method thus validating the analytical models used in the present work. Chapter-5 describes the results of the studies on ductile phase toughening of MoSi2 based composites by various refractory metal reinforcements. Addition of 20 vol% W, Mo and Nb particulates in MoSi2 resulted in a considerable improvement in hardness and flexural strength while retaining the fracture toughness over the monolithic MoSi2. Only marginal improvement in fracture toughness might be attributed to the conversion of W, Mo and Nb particulates into their silicides due to extensive diffusion of Si from MoSi2 during high temperature processing. The silicides formed were identified to be of R5Si3 (R = refractory metal) type. Hardness and strength improvement could be attributed to the formation of hard, brittle silicide phases while the same silicides did not result in improved toughness by way of crack bridging. Therefore, the strategy was changed and the ductile refractory metals were used in continuous (foil) form rather than in discontinuous (particulate) form. Model tri-layer laminated composites were made by sandwiching a single ductile refractory metal foil in between two layers of MoSi2 powder by vacuum hot pressing. 2 wt% Al was added in MoSi2 matrix. Al addition into MoSi2 matrix was found to substantially decrease the amount of SiO2 in MoSi2. Al reacted with SiO2 and formed A1203 in-situ. This configuration resulted in a considerable improvement in fracture toughness of laminated composite with Ta foil but there was extensive interfacial debonding in the laminated composite with Mo foil due to the large thermal expansion mismatch between MoSi2 and pure Mo foil. The coefficient of thermal expansion of MoSi2 is higher than the coefficient of thermal expansion of all the refractory metal foils used in the present study resulting in large residual thermal stresses. To address the problem of thermal expansion mismatch between the layers of the laminated composites, the strategy adopted was to add a moderate amount of 20 vol% SiC particulates in MoSi2 matrix. The coefficient of thermal expansion of SIC is much lower than the coefficient of thermal expansion of MoSi2. Addition of SiC into MoSi2 decreases the effective coefficient of thermal expansion of the matrix layer and thus results in lower residual thermal stresses in the composite. vii These hybrid laminated composites (with brittle SiC particulates and ductile refractory metal foils) exhibited very significant improvement in room temperature fracture toughness over the monolithic MoSi2 + 20 vol% SiCp. The highest fracture toughness measured from the peak load obtained in a three-point bend test on notched (SENB) specimen was of the order of 20 MPa'lm in case of laminated composite with Ta foil. Such improvements in fracture toughness could be attributed to the synergistic effect of brittle (SiC particulates) and ductile (refractory metal foils Mo, Ta and Nb) reinforcements together, as well as improved thermal compatibility between the layers of the laminated composites. However, interfacial reaction layers are formed due to diffusion of Si from MoSi2 towards the refractory metal foils at high processing temperatures. The thickness of the interfacial reaction layers was measured as 40 1..trn, 20 gni and 10 pm for Mo, Nb and Ta foil laminated composites, respectively. The various interfacial reaction products in different composites were identified as Mo5Si3, Mo3Si, Mo2C, Nb5Si3, Ta5Si3, and Ta2Si by extensive use of EPMA studies. To suppress the chemical interactions between MoSi2 and refractory metal foils at high processing temperatures, application of an inert diffusion barrier coating on the metal foil was tried. A1203 coating on Nb foil was applied by plasma spray method prior to consolidation into a tri-layer laminated composite by vacuum hot pressing. Microstructural and EPMA studies have revealed that A1203 coating on Nb foil effectively inhibited the Si diffusion across the interface and suppressed the matrix-reinforcement chemical interactions during the high temperature processing. Only a very thin Si rich layer was found at the A1203 coating / Nb foil interface due to limited diffusion of Si through the A1203. However, the A1203 coated Nb foil laminated composite exhibited a relatively lower increase in stress intensity factor based fracture toughness than uncoated Nb foil laminated composite over the monolithic MoSi2 + 20 vol% SiCp. However, the energy derived fracture toughness measured in terms of work of fracture for A1203 coated Nb foil composite was found to be higher than for the uncoated Nb foil composite. This is attributed to a weak interfacial bonding between MoSi2-A1203 coated Nb system. The formation of various interfacial reaction products in different laminated composites is analysed based on binary and ternary (if available) phase diagrams and viii thermodynamic calculations (standard free energy change vs. temperature) for several possible chemical reactions between the constituents of the various composite systems. An attempt has also been made to analyse the interfacial debonding ahead of a propagating crack based on the theories proposed by Cook and Gordon (1964), He and Hutchinson (1989) and Evans and Marshall (1989). The criteria for interfacial debonding seems to be satisfied in all the laminated composites prepared in the present work. However, observations of indentation crack paths under optical and scanning electron microscopes revealed no debonding at the interface upon impingement of the crack on the metal foils. The indentation cracks were arrested by ductile refractory metal foils. It is believed that the impingement of the crack on the interface caused local dislocation slip in the reinforcement instead of interfacial failure, leading to the release of the stress concentration. The present results suggested that interfacial failure was not the only mechanism of blunting cracks in case of ductile reinforcements. The slip capability of ductile reinforcement can play an important role. The fracture toughness of laminated composites measured in crack arrester mode was found to be much higher than the fracture toughness measured in crack divider mode. The improvement in fracture toughness in crack divider mode was found to be moderate for all the laminated composites prepared in the present study. The fracture toughness measured as "Work of fracture" exhibited a trend opposite to the "damage tolerance" calculated from the peak load. The mechanisms of crack propagation in crack arrester and crack divider modes were analysed based on the typical load-displacement curves obtained, photographs of the specimens taken after the completion of the test under stereo and scanning electron microscopes and the typical fracture surface features revealed by the SEM. Chapter-6 outlines the conclusions of the present investigation. In summary, it has been successfully demonstrated that ductile phase toughening of MoSi2 is a viable approach to improve fracture toughness of MoSi2 based materials. The experiments and models establish the necessary basis for understanding and designing the MoSi2 matrix composites with toughness levels larger than that of monolithic MoSi2. The principles and the methods of residual thermal stress analysis employed in the present work may hold equally good and useful for other ceramic matrix composite systems un