As a part of the component development process for the particle bed reactor (PBR), it is necessary to develop coatings which will be time and temperature stable at extremely high temperatures in flowing hydrogen. These coatings must protect the underlying carbon structure from attack by the hydrogen coolant. Degradation which causes small changes in the reactor component, e.g. hole diameter in the hot frit, can have a profound effect on operation. The ability of a component to withstand repeated temperature cycles is also a coating development issue. Coatings which crack or spall under these conditions would be unacceptable. While refractory carbides appear to be the coating material of choice for carbon substrates being used in PBR components, the method of applying these coatings can have a large effect on their performance. Two deposition processes for these refractory carbides, chemical vapor deposition (CVD) and chemical vapor reaction (CVR) have been evaluated.

Screening tests for these coatings consisted of testing of coated 2-D and 3-D weave carbon-carbon in flowing hot hydrogen at one atmosphere. Carbon loss from these samples was measured as a function of time. Exposure temperatures up to 3000 K were used and samples were exposed in a cyclical fashion, cooling to room temperature between exposures. The results of these measurements are presented along with an evaluation of the relative merits of CVR and CVD coatings for this application.

As a part of the component development process for the particle bed reactor (PBR), it is necessary to develop coatings for fuel particles which will be time and temperature stable. These coatings must not only protect the particle from attack by the hydrogen coolant, but must also help to maintain the bed in a coolable geometry and mitigate against fission product release. In order to develop these advanced coatings, a process to produce chemical vapor reaction (CVR) coatings on fuel for PBRs has been developed.

The initial screening tests for these coatings consisted of testing in flowing hot hydrogen at one atmosphere. Surrogate fuel particles consisting of pyrolytic graphite coated graphite particles have been heated in flowing hydrogen at constant temperature. The carbon loss from these particles was measured as a function of time. Exposure temperatures ranging from 2500 to 3000 K were used and samples were exposed for up to 14 minutes in a cyclical fashion, cooling to room temperature between exposures. The rate of weight loss measured as a function of time is compared to that from other tests of coated materials under similar conditions. Microscopic examination of the coatings before and after exposure was also conducted and these results are presented.

Several novel transition metal nitrides were synthesized via ammonolysis of solid state oxide precursors at temperatures ranging from 700°C-900°C and reaction times ranging from 12 hours to 4 days. Both intermetallic nitrides, Fe3Mo3N and Co3Mo3N, and ionic/covalent nitrides, FeWN2, MnWN2, Ta5N6 and Nb5N6, were prepared by this method. The products were characterized by powder X-ray diffraction and their structures were determined by powder X-ray Rietveld refinement. The intermetallic nitrides were found to be isostructural with the eta-carbide structure, Fe3W3C, while the ionic/covalent nitrides have layered structures, with metals in octahedral and trigonal prismatic coordination environments. Two polymorphs of the MnWN2 composition, α-MnWN2 and β-MnWN2, were isolated after ammonolysis at 700°C and 800°C, respectively. While the alpha phase can be converted into the beta phase by heating to 800°C under ammonia, annealing the beta phase at 700°C did not result in a structural transformation. Magnetic measurements show that FeWN2 orders antiferromagnetically at 45K. The magnetic ordering temperature was confirmed by M6ssbauer spectroscopy. All the other nitrides were paramagnetic down to 5K. Conductivity measurements show that FeWN2 and MnWN2 are metallic.

The ternary nitride, FeWN2, has been synthesized via ammonolysis of the transition metal tungstate, FeWO4. FeWN2 is hexagonal with a unit cell of a = 2.8724(1) Å and c =10.973(4) A. Rietveld refinements were performed in the space group P3–1C with Rwp=9.91%, and Rp=7.23%. The structure consist of alternating octahedrally coordinated iron-nitrogen layers and trigonal prismatically coordinated tungsten-nitrogen layers. Mössbauer data at 250K indicates an isomer shift δ=0.85 and shows no quadrupole splitting, which is consistent with the octahedral iron coordination in the proposed structure.

Manufacturing of high temperature materials is an energy expensive process, and conventional techniques require long periods of homogenization to ensure uniform chemical composition in large batches of material. Several covalent ceramics, and high temperature materials can be manufactured by utilizing the exothermicity of the reactants. Some of the salient features of combustion synthesis, approaches to densify the combustion synthesized product, and other processes that utilize exothermic reactions for the synthesis of materials have been discussed.

Particle bed reactors (PBRs) have the potential for providing compact power sources for both space power and propulsion applications. The reactors operate in the temperature range of 2000 to 3000 K and, for propulsion systems, utilize hydrogen as a coolant. Given these overall operating conditions, the need exists to develop and verify the performance of various reactor components, particularly those which must operate at the more extreme conditions of temperature and hydrogen flow rate. These components must undergo multiple thermal cycles between cryogenic and exit temperatures. In addition, some of the components are subject to a thermal gradient of thousands of degrees Kelvin over a few centimeters.

A number of approaches directed towards meeting these needs have been developed. These include both monolithic ceramics (e.g. boron nitride) as well as coated composites consisting of refractory carbides (ZrC, NbC and TaC) on carbon-based substrates. Full and partial scale components have been fabricated and tested under prototypic conditions. The results of this developmental effort will be described.

In this paper the differences of properties of stoichiometric and nonstoichiometric IVBand VB-Group carbides will be discussed. On the base of newer investigations of their physical and chemical properties it seems to be possible to improve material properties by using them instead of stoichiometric ones in some applications.

The preparation, characterization and pyrolytic conversion to silicon-containing ceramics of two types of polymers – a series of poly(ureidosilazanes) and poly(silylenecarbodiimides) – are described. The former appears to have some promise as a preceramic polymer.

A copolymer prepared using methylcyclosiloxanes and methylcyclosilazanes has been pyrolysed in N2 and a NH3/N2 mixture. Thermogravimetric analysis (TGA) in both atmospheres and thermogravimetry-Fourier transform infrared (TG-FTIR) studies in N2 are discussed. The latter results have been used to suggest a mechanism for the thermal decomposition of the copolymer.

Polymethylsilane (-[MeSiH]x-) synthesized from a single step catalytic polymerization of methylsilane (MeSiH3), provides a new route to near stoichiometric, microcrystalline silicon carbide fibers. The final composition is controlled by selected additives. The entire process simply contains 3 simple steps: (1) a single step synthesis of polymer precursor, (2) precursor fiber spinning, and (3) a single step pyrolytic conversion to target ceramic fibers.

Selected organosilanes were examined as precursors for the deposition of amorphous hydrogenated silicon carbide in an argon/hydrogen plasma. The effect of process variables on the quality of the films was established by means of FTIR, Auger spectroscopy, XPS, XRD, chemical analysis, and weight losses upon pyrolysis. For a give power level there is a limiting feeding rate of the precursor under which operation of the system is dominated by thermodynamics and leads to high quality silicon carbide films that are nearly stoichiometric and low in hydrogen. Beyond that limit, carbosilane polymer formation and excessive hydrogen incorporation takes place. The hydrogen content of the plasma affects the deposition rate and the hydrogen content of the film. In the thermodynamically dominated regime the nature of the precursor has no effect on the quality of the film, it affects only the relative utilization efficiency.

Cubic β-SiC was processed under conditions of self-propagating high-temperature synthesis, based on the exothermic reaction between elemental Si and C powders. The set up for the synthesis of SiC using a mixture of Si and C is described. X-Ray and Raman spectroscopy studies were performed to characterize the quality of the β-SiC produced using the new set up. Scanning Electron Microscopy and Transmission Electron Microscopy studies were also carried out to study the mechanism of the formation of β-SiC from the elemental powders. The density of the compact and grain size of the graphite was found to be critical in the formation of the SiC. Diamond seeds were also implanted on the SiC pellet while compaction and this implanted diamond provided the necessary seed for the growth of a thick diamond film. The adhesion of the film is good because of its growth from the implanted diamond. Diamond films were grown on SiC using HFCVD. The quality and the adhesion of the diamond films on SiC were studied using SEM.

Rapid self-sustaining solid-state precursor methods have proven to be an effective route to non-oxide ceramics. With the appropriate choice of precursors, fine powders of crystalline BN, AIN as well as composites such as TiB2-TiN-BN can be formed easily within seconds. These materials are characterized by X-ray diffraction, scanning electron microscopy and elemental analysis (XPS, EDS, and conventional chemical methods). The synthetic techniques will be discussed along with characterization and numerical simulation results. Further applications of this technique will be proposed.

Boron nitride presents many interesting properties for high temperature applications: high specific modulus, high electrical resistivity, electromagnetic transparency and strong chemical stability. However synthesis and densification by liquid or gas phase reactions (CVD, CVI or plasma processes) required for high technology applications (thermal shield, protective coating, microwave windows) are difficult to bring into play because material characteristics are strongly dependent on elaboration conditions. Boron nitride ultrafine powder synthesized by gas phase reaction in RF plasma (using NH3 and BCh3 at 800-1400°C) and composites obtained using a rapid densification process (vapor phase infiltration using trichloroborazole compound as precursor) have been studied. In both cases post synthesis heat treatment had to be performed in order to get stabilization of the material, and to avoid its hydrolysis. The stability has been related to the physical characteristics of the material. Thermodynamical studies using Gibbs energy minimization calculations were helpful to determine synthesis and heat treatment conditions.

The ternary nitride MgSiN2 crystallizes in a diamond-like structure with a band gap of 4.8 eV. These characteristics make the compound promising for a variety of applications. Recently we succeeded in preparing fully dense ceramics by sintering at 1550°C. Phase composition, chemical properties and the mechanical properties of the as-prepared ceramics are described. The ceramics are resistant to oxidation in air at least up to 920°C. The thermal conductivity at room temperature has been found to be 20 W/m·K. A considerable improvement of the thermal conductivity is expected when improved processing conditions, well within reach, are realized. A reasonable strength of 270 MPa and a fairly good fracture toughness of about 4.3 MPa·m½ are obtained. A hardness of about 15 GPa and a Young's modulus of 235 GPa have been measured. These new ceramics show that it is still possible to find improved materials as compared to the usual oxide and non-oxide ceramics.

Thin films of diamond and diamond-like carbon (DLC) are technologically important materials that serve as hard, scratch resistant and chemically inert coatings for tools and optics. Recent calculations suggest that β-C3N4 should be harder than diamond. We have deposited carbon nitride (CNx) thin films by pulsed laser deposition. The films were grown from a graphite target in a nitrogen background. The nitrogen source was either (a) a N2 gas atmosphere, or (b) a N2+/N+ ion beam generated by a Kaufman ion gun. A wide range of deposition parameters were investigated, such as deposition pressure (0.3-900 mTorr N2), substrate temperature (50 and 600°C), and laser fluence (1-4 J/cm2) and laser repetition rate (1-10 Hz). The films have been characterized by Rutherford Backscattering Spectroscopy, thin-film X-ray diffraction, scanning electron microscopy, and micro-Raman spectroscopy. In general, the films were nitrogen deficient with a maximum nitrogen to carbon ratio (N/C) of 0.45 and a shift in the G band Raman peak consistent with amorphous CNx (a-CNx).

Continuous fiber ceramic composites are enabling new, high temperature structural applications. Chemical vapor infiltration methods for producing these composites are being investigated, with the complexity of filament weaves and deposition chemistry merged with standard heat and mass transport relationships. Silicon carbide-based materials are, by far, the most mature, and are already being used in aerospace applications. This paper addresses the state-of-the art of the technology and outlines current issues.

Historically, texture analysis has been useful in explaining processing-induced property anisotropy in metals and deformation history in geological materials. Texture analysis, however, has not been fully utilized in explaining similar phenomena in ceramic systems. The scarcity of texture analysis in ceramic systems could conceivably stem from the classical conception that textures rarely develop in ceramic materials because of their lack of plasticity. While it is true that slip-based textures occur in only a limited number of ceramic systems, textures occur in a multitude of ceramic systems in which dislocation plasticity is not observed. The most prominent non-slip-based texture development mechanism in ceramic systems is grain rotation during processing. Other possible mechanisms for texture development in ceramics are seeded solid solution formation and oriented grain growth. These non-slip-based texture development mechanisms and other phenomena related to the measurement of textures in covalent ceramics are discussed in the context of the A1N-SiC system.

The role of boron and carbon during densification of sintered β-SiC was investigated through the combined approach of in-situ dilatometry and CTEM/AEM inspection of TEM-foils referring to well-defined densification events. Preliminary data obtained indicate that in the early stages of densification, boron is not enriched in the continuous carbon-rich surface layer covering the β-SiC powder particles nor does it segregate to internal interfaces in high quantities. Small boron quantities are dissolved in the SiC grains- Simultaneously with the β- to α- phase transformation, decomposition of foam-like B4C aggregates releases small B4C particles, which are bound intragranularly to α-SiC.

During an HREM/AEM investigation, two different oxygen sinks were identified for a variety of SiC-ceramics involving different sintering mechanisms: segregation of oxygen to i) internal interfaces, and ii) “excess-carbon” secondary phases consisting of basically turbostratic graphite and an intermediate amorphous Si-C-O phase. Si-C-O glasses and polycarbosilanebased nonstoichiometric SiC materials (Nicalon fibers) are well suited as model systems to evaluate the structure and the stability range of the amorphous Si-C-O phase in SiC-based ceramics.