Sharp leading edges and nose caps on hypersonic vehicles, re-entry vehicles and reusable launch vehicles are items of current research interest for enhanced aerodynamic performance and maneuverability. The unique combination of mechanical properties, physical properties, thermal / electrical conductivities and thermal shock resistance of ZrB2 make it a promising candidate material for such applications. In the recent past, a lot of work has been carried out on ZrB2-based materials towards processing as well as characterization of their mechanical, oxidation and thermal behaviour. ZrB2 based materials have been successfully processed by conventional hot pressing, pressureless sintering, reactive hot pressing and spark plasma sintering. Densification of ZrB2 gets activated when the oxide impurities (B2O3 and ZrO2) were removed from particle surfaces, which minimized coarsening. B4C is widely used as a sintering additive for ZrB2 because it reduces ZrO2 at low temperature. It is found that full densification in ZrB2 based materials by hot pressing is achieved either at 2000 C and higher temperatures with moderate pressure of 20-30 MPa or at reduced temperature (1790-1840 C) with much higher pressure (800-1500 MPa). But no study is available that identifies the dominant hot pressing mechanism at different temperatures and pressures. On the other hand, reinforcement of SiC in ZrB2 is known to increase flexural strength, fracture toughness and oxidation resistance. It has been shown that oxidation resistance of ZrB2-SiC composites is superior to that of monolithic ZrB2 and SiC. For high temperature applications in air, the residual strength (room temperature strength after exposure in air at high temperatures) of non oxide ceramics after oxidation is important. A few reports are available on residual strength of ZrB2 –SiC composite after thermal exposure at high temperatures. In contrast to the literature on composites, there are no reports available on the residual strength of monolithic ZrB2 after exposure to high temperatures. Also, previous studies on residual strength of ZrB2-SiC composites have been limited to a single temperature of exposure. But there is a need to measure the residual strength after exposure to a range of temperatures since the oxide layer structure changes with temperature. The room temperature thermal conductivity data for ZrB2 and ZrB2-SiC composite shows a wide scatter in value as well as a dependence on microstructural parameters, especially porosity and grain size. Also, there is insufficient data available for the high temperature thermal conductivity of ZrB2-SiC. Therefore, it is difficult to evaluate the effect of SiC content on thermal conductivity of ZrB2-SiC composites at high temperatures. The present thesis seeks to address some of these gaps to better understand the suitability of ZrB2 and ZrB2-SiC composites for ultra-high temperature applications. In the present work, hot pressing is used for densification of ZrB2 and ZrB2-SiC composites. Different amounts of B4C (0, 0.5, 1, 3 & 5 wt %) were used as sintering additives in ZrB2 and hot pressed at 2000 C with 25 MPa applied pressure. The hot pressed samples are characterized for their microstructural, mechanical properties and oxidation behaviour. By addition of B4C, density as well as micro-hardness increased. For lower B4C content (0.5 & 1 wt %), hot pressed ZrB2 has shown considerable improvement in flexural strength after exposure in air at 1000 C for 5 hours, while higher B4C content (3 & 5 wt %) leads to marginal or no improvement. Due to the better mechanical and oxidation behavior of composites containing SiC, the densification behavior during hot pressing was studied. The densification behaviors as well as the microstructures for hot pressing of ZrB2-20 % SiC composite were found to change in a very 0 narrow temperature range. During hot pressing at 1700 C, the densification was found to be mechanically driven particle fragmentation and rearrangement. On the other hand, thermally activated mass transport mechanisms started dominating after initial particle fragmentation and rearrangement after hot pressing at 1850 C and 2000 C. At 2000 C, the rate of grain boundary diffusion was enhanced which resulted into annihilation of dislocation. The effect of SiC contents (10, 20 & 30 vol %) on mechanical and oxidation behavior of ZrB2-SiC composite were also studied. The average micro-hardness and fracture toughness of ZrB2-SiC composites increased with SiC content. But the flexural strength of ZrB2-20 vol % SiC composites was found to be the highest. Oxidation and residual strength of hot pressed ZrB2 -SiC composites were evaluated as a function of SiC contents after exposure over a wide temperature range (1000-1700 C). Multilayer oxide scale structures were found after oxidation. The composition and thickness of these multilayered oxide scale structures were found to depend on exposure temperature and SiC content. After exposure to 1000 C for 5 hours, the residual strength of ZrB2 -SiC composites improved by nearly 60 % compared to the as-hot pressed composites with 20 & 30 vol % SiC. On the other hand, the residual strength of these composites remained unchanged after 1500 C for 5 hours. A drastic degradation in residual strength was observed in composites with 20 & 30 vol % SiC whereas strength was retained for ZrB2-10 % SiC composite after exposure to 1700 C for 5 hours in ZrB2 –SiC. Therefore, residual strength of ZrB2-10 % SiC composite was measured at different exposure times (up to 10 hours) at 1500 0C. An attempt was made to correlate the microstructural changes and oxide scales with residual strength with respect to variation in SiC content and temperature of exposure. Since the ZrB2-20 vol % SiC composite showed the maximum strength, the dependence of strength on various microstructural as well processing parameters was also studied. It was found that porosity, grain size as well as surface residual stress due to grinding influenced the strength of ZrB2-20 vol % SiC composites. Finally, thermal diffusivity and conductivity of hot pressed ZrB2 with different amounts of B4C and ZrB2-SiC composites were investigated experimentally over a wide temperature range (25 – 1500 C). Both thermal diffusivity as well as thermal conductivity was found to decrease with increase in temperature for all hot pressed ZrB2 and ZrB2-SiC composites. At around 200 C, thermal conductivity of ZrB2-SiC composites was found to be composition independent. Thermal conductivity of ZrB2-SiC composites was also correlated with theoretical predictions of the Maxwell-Eucken relation. The dominated mechanisms of heat transport for all hot pressed ZrB2 and ZrB2-SiC composites at room temperature were determined by Wiedemann-Franz analysis using measured room temperature electrical conductivity of these materials. It was found that the electronic thermal conductivity dominated for all monolithic ZrB2 whereas the phonon contribution to thermal conductivity increased with SiC contents for ZrB2-SiC composites. The heat conduction mechanism at high temperature was also studied by measuring the high temperature electrical conductivity of ZrB2 and ZrB2-SiC composites. The effect of porosity on thermal diffusivity and conductivity was also studied for ZrB2-20 vol % SiC composites.

Sharp leading edges required for hypersonic vehicles improve the maneuverability as well as reduce aerodynamic drag. However, due to the sharp design, increased surface temperatures require materials that can withstand these extreme conditions. Ultra-high temperature ceramics are a material group being considered for the leading-edge material, specifically ZrB₂/SiC (ZBS) which has a high thermal shock resistance, melting temperature, and thermal conductivity. Studies done by Tan et. al. has shown that adding samarium (Sm) as a dopant to ZBS has an emittance of 0.9 at 1600^oC and develop oxide scales that have excellent ablation performance. However, it remained unknown how the Sm doped oxide scale formed as well as how the emittance and ablation performance are affected by the microstructure. This study investigates the oxide scale development of 3 mol% doped Sm-ZBS billets as well as how differences in microstructure affect the emittance and ablation performance. Samples were prepared via chemical infiltration of samarium nitrate into spray-dried powders of 80 vol.% ZrB₂/20 vol.% SiC; powders were then pressed into billets and pressureless sintered. Samples cut and polished from these billets were then oxidized for 10, 60, or 300 s, respectively, using an oxyacetylene torch. X-ray diffraction was used to determine the sequence of oxidation of Sm-ZBS, beginning with the formation of ZrO₂ and Sm₂O₃. The final oxide scale was determined to be c₁-Sm_0.2Zr_0.8O_1.9, with a melting temperature exceeding 2500^oC. SEM and EDS were also used to investigate the microstructural formation that occurs from the bursting of convection cells. Samples with different microstructures revealed similar topographical microstructures post-ablation due to the sequence of the oxide formation. However, samples with rougher surfaces and higher porosities had a higher concentration of trapped glass in the cross-sectional oxide scale. It was also found that due to differences in heating the sample during emittance testing compared to ablation testing, the oxide developed was identical for all the samples. It was also found that variances in microstructure had no effect on the spectral emittance of Sm-ZBS at ultra-high temperatures. The fabrication of c₁-Sm_0.2Zr_0.8O_1.9 (SZO) as a bulk billet was also investigated to use as a thermal barrier coating (TBC) in replacement of Sm-ZBS.

Specimens of ZrB2 containing various concentrations of B4C, SiC, TaB2, and TaSi2 were pressureless-sintered and post-hot isostatic pressed to their theoretical densities. Oxidation resistances were studied by scanning thermogravimetry over the range 1150 - 1550 degree C. SiC additions improved oxidation resistance over a broadening range of temperatures with increasing SiC content. Tantalum additions to ZrB2-B4C-SiC in the form of TaB2 and/or TaSi2 increased oxidation resistance over the entire evaluated spectrum of temperatures. TaSi2 proved to be a more effective additive than TaB2. Silicon-containing compositions formed a glassy surface layer, covering an interior oxide layer. This interior layer was less porous in tantalum-containing compositions. The oxidation resistances of ZrB2 containing SiC, TaB2, and TaSi2 additions of various concentrations was studied using isothermal thermogravimetry at 1200, 1400, and 1500 degree C, and specimens were further characterized using x-ray diffraction and electron microscopy. Increasing SiC concentration resulted in thinner glassy surface layers as well as thinner ZrO2 underlayers deficient in silica. This silica deficiency was argued to occur by a wicking process of interior-formed borosilicate liquid to the initially-formed borosilicate liquid at the surface. Small (3.32 mol%) concentrations of TaB2 additions were more effective at increasing oxidation resistance than equal additions of TaSi2. The benefit of these additives was related to the formation of zirconium-tantalum boride solid solution during sintering, which during oxidation, fragmented into fine particles of ZrO2 and TaC. These particles resisted wicking of their liquid/glassy borosilicate encapsulation, which increased overall oxidation resistance. With increasing TaB2 or TaSi2 concentration, oxidation resistance degraded, most egregiously with TaB2 additions. In these cases, zirconia dendrites appeared to grow through the glassy layers, providing conduits for oxygen migration.

In this thesis, growth of ZrB2 thin films by direct current magnetron sputtering is investigatedusing a high vacuum industrial scale deposition system and an ultra-high vacuum laboratory scalesystem. The films were grown from ZrB2 compound targets at temperatures ranging from ambient (without external heating) to 900 °C and with substrate biases from -20 to -120 V. Short deposition times of typically 100 or 300 s and high growth rates of 80-180 nm/min were emphasized to yield films with thicknesses of 300-400 nm. The films were characterized by thinfilm X-ray diffraction with the techniques θ/2θ and ω scans, pole figure measurements andreciprocal space mapping, scanning and transmission electron microscopy, elastic recoil detection analysis and four point probe measurements. The substrates applied were Si(100), Si(111),4H-SiC(0001) and GaN(0001) epilayers grown on 4H-SiC. The Si(111), 4H-SiC(0001) substrates and GaN(0001) epilayers were chosen given their small lattice mismatches to ZrB2 making them suitable for epitaxial growth.The films deposited in the industrial system were found to be close to stoichiometric with a low degree of contaminants, with O being the most abundant at a level of < 1 at.%. Furthermore, the structure of the films is temperature dependent as films deposited in this system without external heating are fiber textured with a 0001-orientation while the films deposited at 550 °C exhibitrandom orientation. In contrast, epitaxial growth was demonstrated in the laboratory scale system on etched 4H-SiC(0001) and Si(111) deposited at 900 °C following outgassing of the substrates at 300 °C and in-situ heat treatment at the applied growth temperature to remove the native oxides. However, films grown on GaN(0001) were found to be 0001 textured at the applied deposition conditions, which make further studies necessary to enable epitaxial growth on this substrate material. Four point probe measurements on the films deposited in the industrial system show typical resistivity values ranging from ˜95 to 200 μΩcm with a trend to lower values for the films deposited at higher temperatures and at higher substrate bias voltages.

Spark plasma sintering (SPS) is a powder consolidation technique used to rapidly densify a variety of material systems. SPS is capable of precisely controlling material microstructures and achieving non-equilibrium phases due to rapid heating and cooling rates through the simultaneous application of pressure and direct current. Due to these characteristics, SPS is an ideal processing technique for high temperature ceramics which require processing at temperatures greater than 1500°C. Due to the desirable properties obtained on small diameter materials processed by SPS, larger and more complex geometries are desired while maintaining sample microstructures. In order to accurately scale ceramics produced by SPS, a finite element model must be developed that can be used as a predictive tool. My research focuses on developing a finite element model for the spark plasma sintering furnace at the University of Arizona and validating modeled results using experimentally obtained data. Electrical and thermal conductivity as functions of temperature vary widely among different grades of commercially available electrode grade graphite at constant density. Modeled material properties are optimized in order to calibrate modeled results to experimentally obtained data (i.e. measured current, voltage, and temperature distributions). Sensitivity analysis is performed on the model to better understand model physics and predictions. A calibrated model is presented for 20mm ZrB2 and Si3N4 discs. Sample temperature gradients are experimentally confirmed using grain size and β-Si3N4 phase composition. The model is used to investigate scale up from 20mm to 30mm discs and 30mm rings as well as effects of processing conditions on β-Si3N4 content.

The aim of this study was to produce zirconium diboride (ZrB2) and other metal borides such as lanthanum hexaboride (LaB6) and cerium hexaboride (CeB6) by magnesiothermic reduction (reaction of metal oxide, boron oxide and magnesium) using volume combustion synthesis (VCS) and mechanochemical process (MCP). Production of ZrB2 by VCS in air occurred with the formation of side products, Zr2ON2 and Mg3B2O6 in addition to MgO. Formation of Zr2ON2 was prevented by conducting VCS experiments under argon atmosphere. Wet ball milling was applied before leaching for easier removal of Mg3B2O6. Leaching in 5 M HCl for 2.5 hours was found to be sufficient for removal of MgO and Mg3B2O6. By MCP, 30 hours of ball milling was enough to produce ZrB2 where 10% of excess Mg and B2O3 were used. MgO was easily removed when MCP products were leached in 1 M HCl for 30 minutes. Complete reduction of ZrO2 could not be achieved in either production method because of the stability of ZrO2. Hence, after leaching VCS or MCP products, final product was composed of ZrB2 and ZrO2. Formation of LaB6 and CeB6 were very similar to each other via both methods. Mg3B2O6 appeared as a side product in the formation of both borides by VCS. After wet ball milling, products were leached in 1 M HCl for 15 hours and pure LaB6 or CeB6 was obtained. As in ZrB2 production, 30 hours of ball milling was sufficient to form these hexaborides by MCP. MgO was removed after leaching in 1 M HCl for 30 minutes and the desired hexaboride was obtained in pure form.

Zirconium- and titanium- based compounds (borides, carbides and nitrides) are of importance because of their attractive properties including: high melting temperature, high-temperature strength, high hardness, high elastic modulus and good wear-erosion-corrosion resistance. The ultra high temperature ceramics (UHTCs) - zirconium diboride (ZrB2) and zirconium carbide (ZrC) in combination with SiC are potential candidates for ultra-high temperature applications such as nose cones for re-entry vehicles and thermal protection systems, where temperature exceeds 2000°C. Titanium nitride (TiN) and titanium diboride (TiB2) composites have been considered for cutting tools, wear resistant parts etc. There are problems in the processing of these materials, as very high temperatures are required to produce dense composites. This problem can be overcome by the development of composites through reactive hot processing (RHP). In RHP, the composites are simultaneously synthesized and densified by application of pressure and temperatures that are relatively low compared to the melting points of individual components. There have been earlier studies on the fabrication of dense ZrB2-ZrC, ZrB2-SiC and TiN-TiB2 composites by the following methods: Pressureless sintering of preformed powders at high temperatures (1800-2300°C) with MoSi2, Ni, Cr, Fe additions Hot pressing of preformed powders at high temperatures (1700-2000°C) with additives like Ni, Si3N4, TiSi2, TaSi2, TaC Melt infiltration of Zr/Ti into B4C preform at 1800-1900°C to produce ZrB2-ZrC-Zr and TiB2-TiC composites RHP of Zr-B4C, Zr-Si-B4C and Ti-BN powder mixtures to produce ZrB2-ZrC, ZrB2-SiC and TiN-TiB2 powder mixtures at 1650-1900°C Spark plasma sintering of powder mixtures at 1800-2100°C There has been a lack of attention paid to the conditions under which ceramic composites can be produced by simple hot pressing (~50 MPa) with minimum amount of additives, which will not affect the mechanical properties of the composites. There has been no systematic study of microstructural evolution to be able to highlight the change in relative density (RD) with temperature during RHP by formation of sub-stoichiometric compounds, and liquid phase when a small amount of additive is used. The present study has been undertaken to establish the experimental conditions and densification mechanisms during RHP of Zr-B4C, Zr-B4C-Si and Ti-BN powder mixtures to yield (a) ZrB2-ZrC, (b) ZrB2-SiC, (c) ZrB2-ZrC-SiC and (d) TiN-TiB2 composites. The following reactions were used to produce the composites: (1) 3 Zr + B4C → 2 ZrB2 + ZrC (2) 3.5 Zr + B4C → 2 ZrB2 + 1.52rCx- 0.67 (3) (1+y) Zr + C → (1+y) ZrCx- 1/ (1+y) (y=0 to 1) (4) 2 Zr + B4C + Si → 2 ZrB2 + SiC (5) 2.5 Zr + B4C + 0.65 Si → 2 ZrB2 + 0.5 ZrCx + 0.65 SiC (6) 3.5 Zr + B4C + SiC → 2 ZrB2 + 1.5 ZrCx + SiC (5 to 15 vol%) (7) (3+y) Ti + 2 BN → (2+y) TiN1/(1+y) + TiB2 (y=0 to 0.5) (a) ZrB2-ZrC Composites: The effect of different particle sizes of B4C (60-240 μm, <74 μm and 10-20 μm) with Zr on the reaction and densification of composites has been studied. The role of Ni addition on reaction and densification of the composites has been attempted. The effect of excess Zr addition on the reaction and densification has also been studied. The RHP experiments were conducted under vacuum in the temperature range 1000-1600°C for 30 min without and with 1 wt% Ni at 40 MPa pressure. The RHP composites have been characterized by density measurements, x-ray diffraction for phase analysis and lattice parameter measurements, microstructural observation using optical and scanning electron microscopy. Selected samples have been analyzed by transmission electron microscopy. The hardness of the composites has also been measured. The results of the study on the effect of different particle sizes B4C and Ni addition on reaction and densification in the stoichiometric reaction mixture as follows. With the coarse B4C (60-240 μm and <74 μm) particles the temperature required are higher for completion of the reaction (1600°C and above). The microstructural observation showed that the material is densified even in the presence of unreacted B4C particles. The composite made with 10-20 μm B4C and 1 wt% Ni showed completion of the reaction at 1200°C, whereas composite made without Ni showed unreacted B4C (∼3 vol%) and the final densities of both the composites are similar (5.44 g/cm3). Increase in the temperature to 1400°C resulted in the completion of the reaction (without Ni) accompanied with a relative density (RD) of 95%. The composites produced with and without Ni at 1600°C had similar densities of 6.13 g/cm3 and 6.11 g/cm3 respectively (~97.3% RD). The Zr-Ni phase diagram suggests that the addition of Ni helps in formation of Zr-Ni liquid at ~960°C and leads to an increase in the reaction rate up to 1200°C. Once the reaction is completed, not enough Zr is available to maintain the liquid phase and further densification occurs through solid state sintering. The grain sizes of ZrB2 and ZrC phases after 1200°C are 0.4 μm and 0.3 μm, which are much lower than those reported in literature (2-10 μm), and may be the reason for reducing the densification temperature to 1600°C for stoichiometric ZrB2-ZrC composites. The effect of excess Zr (0.5 mol), over and above the stoichiometric Zr-B4C powder mixture, on reaction and densification of the composites is as follows. The formation of ZrB2 and ZrC phases with unreacted starting Zr and B4C is observed at 1000°C and with increase in temperature to 1200°C the reaction is completed. Since microstructural characterization reveals no indication of free Zr, it is concluded that the excess Zr is incorporated by the formation of non-stoichiometric ZrC (ZrCx-0.67). This observation is supported by lattice parameter measurements of ZrC in the stoichiometric and non-stoichiometric composites which are lower than those reported in the literature. X-ray microanalysis of ZrC grains in the stoichiometric and non-stoichiometric composites using transmission electron microscopy confirmed the presence of carbon deficiency. The composite produced at 1200°C showed the density of 6.1 g/cm3 (~97% RD), whereas addition of Ni produced 6.2 g/cm3 (~99% RD). The reduction in densification temperature for the non-stoichiometric composites is due to the presence of ZrCx even in the absence of Ni. The mechanism of densification of the composites at 1200°C is attributed to the lowering of critical resolved shear stress with increasing non-stoichimetry in the ZrC, which leads to plastic deformation during RHP. An additional mechanism may be enhanced diffusion through the structural point defects created in ZrC. The hardness of the composites are 20-22 GPa, which is higher than those of reported in literature due to the presence of a dense and fine grain microstructure in the present work. In order to verify the role of non-stoichiometric ZrC the study was extended to produce monolithic ZrC using various C/Zr ratios (0.5-1). Here again, stoichiometric ZrC does not densify even at 1600°C, whereas non-stoichiometric ZrC can be densified at 1200°C. (b) ZrB2-SiC Composites: Since ZrB2 and ZrC do not have good oxidation resistance unless they are reinforced with SiC, the present study has been extended to produce ZrB2-SiC (25 vol%) composites using Zr-Si-B4C powder mixtures. The samples produced at 1000°C showed the formation of ZrB2, ZrC and Zr-Si compounds with unreacted Zr and B4C and as the temperature is increased to 1200°C only ZrB2 and SiC remained. A fine grain (~0.5 μm) microstructure has been observed at 1200°C. During RHP, it was observed that the formations of ZrC, Si-rich phases and fine grain size at low temperatures was responsible for attaining the high relative density at a temperature of ~1600°C. The relative densities of the composites produced with 1 wt% Ni at 40 MPa, 1600°C for 30 min is 97% RD, where as composites without Ni showed a small amount of partially reacted B4C; extending the holding time to 60 min eliminated the B4C and produced 98% RD. The hardness of the composites is 18-20 GPa. (c) ZrB2-ZrC-SiC Composites: Since ZrC plays a crucial role in densification of ZrB2-ZrC and ZrB2-SiC composites, the study has been extended to reduce the processing temperature for ZrB2-ZrCx-SiC composites by two methods. In one of the methods, Si is added to the non-stoichiometric 2.5Zr-B4C powder mixture which is resulted in ZrB2-ZrCx-SiC (15 vol%) composites with ~98% RD at 1600°C. In another method, SiC particulates are added to the non-stoichiometric 3.5Zr-B4C powder mixture to yield ZrB2-ZrCx-SiCp (5-15 vol%) composites at 1400°C. The density of the 5 vol% SiC composite is 99.9%, whereas addition of 15 vol% SiC reduced the density to 95.5% RD. The mechanisms of densification of the composites are similar to those observed in ZrB2-ZrC composites. The hardness of the composites is 18-20GPa (d) TiN-TiB2 Composites: ZrB2, ZrC, TiB2, and TiN are members of the same class of transition metal borides, carbides and nitrides; however, their densification mechanisms appear to be different. In earlier work, the RHP of stoichiometric 3Ti-2BN powder mixtures yielded dense composite at 1400-1600°C with 1 wt% Ni addition, whereas composites without Ni required at least 1850°C. The major contributor to better densification at 1600°C (with Ni) appeared to be the formation of local Ni-Ti liquid phase at ~942°C (Ti-Ni phase diagram). The present work explores the additional role of non-stoichiometry in this system. It is shown that Ti excess can lead to a further lowering of the RHP temperature, but with a different mechanism compared to the Zr-B4C system. Excess Ti allows the transient alloy phase to remain above the liquidus for a longer time, thereby permitting the attainment of a higher relative density. However, eventually, the excess Ti is converted into a non-stoichiometric nitride. Thus, the volume fraction of a potentially low melting phase is not increased in the final composite by this addition. The contrast between these two systems suggests the existence of two classes of refractory materials for which densification may be greatly accelerated in the presence of non-stoichiometry, either through the ability to absorb a liquid-phase producing metal into a refractory and hard ceramic structure or greater deformability. Conclusions: The study on RHP of ZrB2-ZrC, ZrB2-SiC, ZrB2-ZrC-SiC and TiN-TiB2 composites led to the following conclusions: • It has been possible to densify the ZrB2-ZrC composites to ~97 % RD by RHP of stoichiometric Zr-B4C powder mixture with or without Ni addition. The role of B4C particle size is important to complete both reaction as well as densification. • Excess Zr (0.5 mol) to stoichiometric 3Zr-B4C powder mixtures produces dense ZrB2-ZrCx composite with 99% RD at 1200°C. The densification mechanisms in these non-stoichiometric composites are enhanced diffusion due to fine microstructural scale / stoichiometric vacancies and plastic deformation. • In the case of ZrB2-SiC composites, the formation of a fine microstructure, and intermediate ZrC and Zr-Si compounds at the early stages plays a major role in densification. • Starting with non-stoichiometric Zr-B4C powder mixture, the dense ZrB2-ZrCx-SiC composites can be produced with SiC particulates addition at 1400°C. • Non-stoichiometry in TiN and ZrC is route to the increased densification of composites through enhanced liquid phase sintering in TiN based composites that contain Ni and through plasticity of a carbon-deficient carbide in ZrC based composites.

Les composites de type Cf/C sont utilisés en tant que pièces structurales dans les propulseurs spatiaux du fait de leurs excellentes propriétés mécaniques dans le domaine des très hautes températures. Néanmoins, l’atmosphère oxydante et corrosive créée lors du décollage des lanceurs et les hauts flux gazeux dégradent ces matériaux. Afin d’améliorer les performances de ces matériaux vis-à-vis de l’oxydation/corrosion, une protection composée de céramiques ultra-réfractaires (dites UHTC) peut être appliquée. Pour une efficacité de protection optimale, des phases UHTC ont été introduites en tant que constituants de la matrice. Dans ces travaux de thèse, la matrice a été réalisée par l’intermédiaire d’un procédé d’élaboration en phase liquide combinant : (i) l’introduction de poudres et (ii) la densification par infiltration réactive d’un métal fondu. La composition de la matrice appartient au système (B;C;Si;Zr). La caractérisation des matériaux après élaboration a permis de comprendre les mécanismes d’infiltration et les réactions permettant de mieux contrôler la composition chimique et la répartition des phases. Des essais sous torche oxyacétylénique ont été utilisés pour se placer dans des conditions proches de l’application visée. La caractérisation post-test des matériaux a permis d’évaluer l’efficacité de la protection dans le cas d’une utilisation unique et également d’une possible réutilisation. Les résultats en oxydation/corrosion ont permis de classer les matériaux en fonction de leur efficacité de protection. / Since many decades, Carbon/Carbon composites are used as structural parts in rocket engines due to their excellent thermomechanical properties. However, under highly oxidizing/corrosive atmosphere and high gas flow rates, carbon suffers from severe oxidation. To improve oxidation resistance of these composites, Ultra High Temperature Ceramics (UHTC) can be used as a protection. To protect the whole composite, the introduction of UHTC as a matrix has been done using a liquid phase process combining: (i) slurry infiltration process and (ii) reactive melt infiltration. Matrix constituents belong to the (B;C;Si;Zr) system. Material characterisation allowed a better understanding of the infiltration mechanisms and of the phase distribution and composition in respect to the processing conditions. To select the best composition, oxyacetylene torch testing has been done to recreate spacecraft launch environmental conditions. Post-test characterisation has been done to evaluate protection efficiency of each matrix composition for single use and possible reuse. Finally, advantages and drawbacks assessment of each composition allowed to highlight the most protective composition and phase distribution.