To produce crack free powder compacts with desirable mechanical properties and uniform densities, a predictive finite element simulation of the powder metallurgy process is necessary (compaction and sintering). The finite element method, through the use of appropriate constitutive material models captures the microstructure-property history after compaction and sintering. A FC-0205 cylinder and FC-0208 automotive main bearing cap were compacted to investigate the microstructure changes at different locations within the parts. Measurements of the pore volume fraction, pore size, pore nearest neighbor, pore aspect ratio, and grain size were performed after compaction for the cylinder after compaction and after compaction and sintering for the MBC. An image analysis methodology was created to measure density in the main bearing cap, and to validate future model results. A comparison between the image analysis and the Archimedes immersion methods demonstrated the reliability of the methods.

In comparison to traditionally fabricated steels that can undergo extensive processing to produce a complex-shaped component, the powder metallurgy (PM) technique can provide a more efficient approach as it is capable of producing intricately-shaped components that require little to no additional processing and machining [1], [2]. A key factor in being able to do so pertains to quenching and utilizing an appropriate quenching agent that can provide dimensional stability to the part being quenched [3], [4]. To ensure that a PM component can perform equally well when being quenched by a quenchant of reduced cooling capability, the PM component should be if not more, then just as hardenable. Steel hardenability can inevitably be improved with the increase of overall alloying content [5], however, if overall alloying content is to be kept at a minimum, the concept of lean PM steel design is one worth investigating; where a lean steel entails that each and every alloying addition is utilized to its maximum potential. This study evaluates the homogenization behaviour of alloying elements in PM steels during sintering as well as the efficiency of wide-spread industrial practices involving the use of various master alloys and ferroalloys, and investigates the realm of liquid phase sintering to understand and optimize the homogenization behaviour of alloying elements and mechanical properties of PM steels. In the context of this work, multi-component master alloys contain at least three of non-ferrous metals as alloying elements and ferroalloys are master alloys containing iron in addition to typically a maximum of two other non-ferrous alloying additions. Part one of this study discusses a combination of thermodynamic software (DICTRA and Thermo-Calc), incremental sintering experiments and scanning electron microscopy (SEM) - wavelength dispersive spectroscopy (WDS) that were used in order to form a deeper understanding of the homogenization behaviour of alloying elements within PM steel during sintering. Electron microscopy analyses on partially and industrially sintered components provide elemental maps to track the evolution of alloying elements as they relax to homogeneity. Electron microscopy analyses for this portion of the study were conducted on an industryproduced automotive component that was sectioned and sintered industrially as well as experimentally at 1280°C for 30 minutes and 13.4 hours. DICTRA simulations carried out for this research provide a 1-D insight into the evolution of concentration profiles and phases throughout various sintering times for systems involving Cr, Mn, C and Fe. DICTRA simulation results of alloying sources were studied alongside alloying element profiles obtained by compiling point quantification from wavelength dispersive spectroscopy maps for the sintered automotive component. Computational results provided conservative, semi-quantitative recommendations on optimal alloy addition forms that lead to an improvement in homogenization. Part two of this study involves the approach of fabricating and testing multi-component master alloy additions. As these materials are widely employed in PM and are typically fabricated by solidification, their states are non-equilibrium and therefore have regions containing phases precipitating in the beginning of freezing which have higher melting temperatures than regions with phases forming later on. During heating, it is hypothesized that Scheil’s solidification path backtracks and as a result, a fraction of liquid in the ferroalloy can be estimated at sintering temperature. If the fraction is significant, the utilization of this ferroalloy implies liquid phase sintering. Through a combination of Thermo-Calc and Fortran softwares, multi-component ferroalloys with promising compositions were discovered in Fe-C-Cr-Mn, Fe-C-Cr-Mn-Ni, FeC-Mn-Mo, Fe-C-Mn-Mo-Ni and Fe-C-Cr-Mn-Mo-Ni systems for low temperature liquid phase sintering. Those of the Fe-C-Cr-Mn-Mo, Fe-C-Cr-Mn-Mo-Ni and Fe-Mn-Mo-Ni system were fabricated and tried in practice. Compositional maps and mechanical properties of PM steels made with variations of this specially tailored multi-component master alloys were compared with those for which traditional alloy additions were used. / Thesis / Master of Applied Science (MASc)

Elemental W and Ni powders were mechanically alloyed in a SPEX mill with WC grinding media for durations ranging from 5 to 50 hours, then compacted samples were sintered in hydrogen to generate bulk tungsten heavy alloys with 2, 4 and 6 wt.% Ni. Evidence of a bimodal grain size distribution was seen in X-ray diffractograms of sintered samples and confirmed by scanning electron microscopy. Grain sizes in the small-grained regions ranged from 200–600 nm; those in the large-grained regions ranged from 1–2 µm. Furthermore, the volume fraction of the small-grained region increased linearly as milling time increased. A slice from a sintered sample was prepared for examination by TEM, in which particles 30–100 nm in diameter were regularly observed on the boundaries of the 200–600 nm grains. EDS point analysis showed that the particles are WC. Therefore it is concluded that heterogeneously distributed contamination from the grinding media is continually incorporated during mechanical alloying and, during sintering, inhibits grain growth through Zener pinning. Densities of sintered samples increased as milling time increased to a maximum of almost 96% of the theoretical value. Density increases with respect to milling time were initially great but diminished upon further milling. While the samples with 4 and 6 wt.% Ni both approached 96% of the theoretical density value by 50 hours of milling, densities in the samples with 2 wt.% Ni were considerably lower. Thus it appears that the Ni that becomes incorporated into the bcc W structure during mechanical alloying activates W diffusion during sintering, though there is a limit to the amount of Ni that the W structure can accommodate. This is evinced in W lattice parameter values from the as-milled powders; while the lattice parameter drops considerable from 2 to 4 wt.% Ni, the difference between 4 and 6 wt.% Ni is much smaller and the Ni content limit surely falls between the two values. Otherwise-equivalent samples with added WC powder were also produced, but did not increase the volume fraction of the small-grained region – probably because the particles remained large and were homogeneously distributed. / Ph. D.

M. Tech. (Department of Metallurgical Engineering, Faculty of Engineering and Technology), Vaal University of Technology. / Pure Ti, Nb, Zr, Al and V powders were used as starting materials. Ti, Ti-6Al-4V and Ti-34Nb-25Zr materials produced by SPS were compared on the basis of density, microstructure, biocompatibility, tensile strength and corrosion resistance. In this study, powder metallurgy (PM) processing route was used to fabricate the alloys. The processing route was mechanical alloying (MA) and spark plasma sintering (SPS). Commercially pure metallic powders (Ti, Nb, Zr, V and Al) of different morphological features and different formulations were prepared. Powder mixing for ternary alloys with Ti as the matrix were conducted in a turbula mixer at a speed of 49 rpm. Followed by mechanical alloying of Ti, Ti-6Al-4V and Ti-34Nb-25Zr in a high energy ball mill for 5h at 500rpm, with a ball to powder ratio of 10:1. Spark plasma sintering of Ti, Ti-6Al-4V and Ti-34Nb-25Zr biomedical alloys was conducted using a hybrid spark plasma sintering furnace at a sintering temperature, heating rate, holding time and pressure of 1200°C, 100°C/min, 10min and 50MPa, respectively. Ti-34Nb-25Zr was fabricated in two ways, fully densified and porous samples. The fully densified sample was fabricated at a sintering temperature, heating rate and holding time and pressure of 1200°C, 100°C/min, 10min and 50MPa, respectively. Whereas, porous Ti-34Nb-25Zr was fabricated using NaCl space holder at a sintering temperature, heating rate, holding time and pressure of 750°C, 50°C/min, 5min and 50MPa, respectively. This was done to compare the solid and porous alloy biocompatibility behaviour. Microstructures, elemental compositions. Phase constitution of the sintered specimens were examined using a field emission scanning electron microscope (FE-SEM) equipped with energy dispersive x-ray spectrometer (EDS) and an x-ray diffractometer (XRD). The microstructure of Ti-34Nb-25Zr had pores and precipitates of niobium. Relative density, micro-hardness, biocompatibility and corrosion test was also conducted on the metallographically polished cross sections of sintered specimens. Ti, Ti-6Al-4V and Ti-34Nb-25Zr alloys made from the irregularly shaped Ti powders and sintered on the hybrid sintering machine yielded higher densifications reaching up to 100 % relative densities. Hardness values ranging from 300-600Hv at a load of 0.5kg. The corrosion resistance of the alloys was higher in the range of 2-4 nA/cm2 exhibiting a passive behaviour in simulated body fluids, such as Hank’s, 0.9wt.% NaCl and eagles minimum essential + 10% fetal bovine serum (E-MEM+ 10% FBS). Biocompatibility tests were conducted (cytotoxicity by WST-1 with SaOS-2 human osteosarcoma cells, protein adsorption and surface wettability). Fibronectin adsorption was less for solid Ti and Ti-34Nb-25Zr (<2ng/mm) compared to Ti-34Nb-25Zr porous and Ti-6Al-4V (4 ng/mm). Albumin adsorption was the highest on Ti substrate (3 ng/mm) than on the fully densified and porous Ti-34Nb-25Zr surfaces followed by less adsorption on Ti-6Al-4V. Surface wettability of Ti and Ti-6Al-4V showed a high contact angle of between 93-98° compared to 86° for the Ti-34Nb-25Zr solid alloy, indicating that Ti-34Nb-Zr alloys exhibited hydrophilic behaviour. The surface wettability results correlated well to less fibronectin adsorption on Ti-34Nb-25Zr solid alloy and excellent adsorption for Ti-6Al-4V. Solid and porous Ti-34Nb-25Zr showed less cell proliferation (0.06 and 0.02% cell viability) which was possibly linked to fibronectin adsorption results. Biocompatibility behaviour of Ti-34Nb-25Zr solid and porous alloys was poorer than Ti (0.20% cell viability) and Ti-6Al-4V (0.23% cell viability). There was poor protein adsorption and cell proliferation on all the alloy substrates.

Thesis (MScEng) -- Stellenbosch University, 2014. / ENGLISH ABSTRACT: In this study, commercially pure titanium foam was produced using space holder powder metallurgy techniques. Titanium foam is attractive as a scaffolding material for bone replacement and implants in the body. The porous morphology of the foam promotes osteogenesis, while the mechanical behaviour of the foam is closer to that of bone, which has an elastic moduli range of 5 - 40 GPa. Titanium foam was manufactured from powder mixtures of commercially pure titanium (CPTi) powder mixed with 41.4 wt% ammonium bicarbonate (ABC) powder and 1.45 wt% polyethyl glycol (PEG) powder. In this study, two CPTi powders with different particle size distributions, < 75 μm (-200 mesh, designated TiAA) and < 200 μm (-100 mesh, designated TiG), were mixed with the space holder ABC powder, that had been sieved into specified particle size ranges. The size ranges of space holder material studied were: 0 - 710, 250 - 425, 425 - 560, and 560 - 710 μm. This allowed foams with different large or macropore distributions to be produced from the different mixtures. The mixtures were uniaxially compacted at 100 MPa into transverse rupture bars. The ABC and PEG was then removed by thermal debinding in air for 5 hours at 100 °C and 1 hour at 330 °C each, consecutively. The debound samples are then sintered under high (10-6 mbar) vacuum on yttria-stabilised zirconia substrates, heating at 5 °C/min to 1200 °C, with a 2 hour hold at temperature. The microstructures of the different foams were evaluated by examining the polished samples using light optical microscopy. Three point bend tests were conducted on the sintered bars in order to determine the flexural strength and flexural modulus of the different foams. The produced foams had a relative density range between 37.5 - 62.5 % and average macro pore size range between 300 - 500 μm. The foams were found to have an elastic modulus similar to that of bone, 2 - 7 GPa. Finally, the mechanical properties of the foams were compared to known open foam mechanical models and other research projects. It was found that: (i) changes in either metal or space holder powder influences the sintering behaviour of metal foams, (ii) sintered titanium foams with similar densities but different macro/micropore size distributions have different mechanical responses to stress and (iii) the Ashby-Gibson model, based on foam density alone, gives a rough estimate of mechanical properties for the titanium foams studied, but does not capture variations due to pore size distribution. / AFRIKAANSE OPSOMMING: In hierdie studie is kommersiële suiwer titaanskuim geproduseer met behulp van ruimtehouer poeier metallurgie tegnieke. Titaanskuim is aantreklik as 'n raamwerkmateriaal vir beenvervanging en -inplantings in die liggaam. Die poreuse morfologie van die skuim bevorder osteogenese, terwyl die meganiese gedrag van die skuim naby aan dié van been is, met ‘n elasticiteitsmodulus tussen 5 - 40 GPa. Titaanskuim is vervaardig van ‘n poeier mengsel van kommersiële suiwer titaan (CPTi) poeier gemeng met 41,4 gew% ammonium bikarbonaat (ABC) poeier en 1.45 gew% poli-etileenglikol (PEG) poeier. In hierdie studie is twee tipes CPTi poeiers met verskillende deeltjiegrootteverspreiding, < 75 μm (-200 stofdigtheid, TiAA genoem) en <200 μm (-100 stofdigtheid, TiG genoem), met die ruimtehouer ABC-poeier, wat in bepaalde deeltjiegroottereekse gesif is, gemeng. Die wisselende groottes van ruimtehouer wat bestudeer is, was: 0 - 710, 250 - 425, 425 - 560, 560 - 710 μm. Dit het die vervaardiging van skuim met verskillende groot of macroporeuse vanaf die verskillende mengsels toegelaat. Die mengsel is teen 100 MPa in een rigting gekompakteer. Die ABC en PEG is dan verwyder word deur termiese ontbinding in lug vir 5 uur by 100 °C en 1 uur by 330 °C elk, onderskeidelik. Die ontbinde monsters is dan onder hoë (10-6 mbar) leemte op yttrium-gestabiliseer zirconia-substraat, met verwarming teen 5 °C/min tot 1200 °C met 'n verdere 2 uur by 1200 °C, gesinterd. Die mikrostrukture van die verskillende skuim is geëvalueer deur gepoleerde monsters met behulp van ‘n ligmikroskopie te ondersoek . Driepunt draaitoetse is op die gesinterd stawe uitgevoer om die buigsterkte en buigmodulus van die verskillende skuime te bepaal. Die vervaardigde skuime se relatiewe digtheid het tussen 37,5 - 62,5 % gewissel en die gemiddelde makroporiegrootte tussen 300 - 500 μm gewissel. Die skuim het 'n elastisiteitsmodulus soortgelyk aan dié van been getoon, 2 – 7 GPa. Ten slotte is die meganiese eienskappe van die skuim met bekende oop skuim meganiese modelle en ander navorsingsprojekte vergelyk. Daar is bevind dat: (i) veranderinge in óf metaal of ruimtehouer poeier beïnvloed die sinteringgedrag van metaalskuime, (ii) gesinterd titaniumskuim met soortgelyke digthede, maar verskillende makro / mikroporeuse verdelings, toon verskillende meganiese reaksies op stres en die Ashby-Gibson model, gebaseer op die skuimdigtheid alleen, (iii) wat 'n rowwe skatting van die meganiese eienskappe vir die bestudeerde titaniumskuime gee, maar nie die variasies ingrootteverspreiding van porieë ondervang nie.

"The goal of heat treating manufactured steel components is to enhance the characteristics of the metal so that the components meet pre-specified quality assurance criteria. However, the heat treatment process often creates considerable distortion, dimensional change, and residual stresses in the components. These are caused mainly by thermal stresses generated by a non-uniform temperature distribution in the part, and/or by transformation stresses due to the volume mismatch between the parent phase and product phases that may form by phase transformation. With the increasing demand for tighter dimensional tolerances and better mechanical properties from heat treated components, it is important for the manufacturer to be able to predict the ability of a component to be heat treated to a desired hardness and strength without undergoing cracking, distortion, and excessive dimensional change. Several commercial softwares are available to accurately predict the heat treatment response of wrought steel components. However, these softwares cannot be used to predict the heat treatment response of steel components that are made by powder metallurgy (PM) processes since these components generally contain pores which affect the mechanical, thermal, and transformation behavior of the material. Accordingly, the primary objective of this research is to adapt commercially available simulation software, namely DANTE, so that it can accurately predict the response of PM steel components to heat treatment. Additional objectives of the research are to characterize the effect of porosity on (1) the mechanical properties, (2) the heat transfer characteristics, and (3) the kinetics of phase transformation during heat treatment of PM steels."

The electric Joule heating of solid materials through direct current excitation can be used to generate a temperature profile throughout a powdermetallic (P/M) compact. When recording the surface temperature distribution with an infrared (IR) camera important information regarding the integrity of the sample can be gained. This research will concentrate on the formulation of a mathematical model capable of predicting the temperature distribution and heat flow behavior in P/M parts and its relations to the supplied current, injection method, geometric shape as well as the thermo-physical properties. This theoretical model will subsequently be employed as a tool to aid in the actual measurements of infrared signatures over the sample surface and their correlation with the detection of surface and subsurface flaws. In this work we will develop the theoretical background of IR testing of green-state and sintered P/M compacts in terms of stating the governing equations and boundary conditions, followed by devising analytical and numerical solutions. Our main emphasis is placed on modeling various flaw sizes and orientations in an effort to determine flaw resolution limits as a function of minimally detectable temperature distributions. Preliminary measurements with controlled and industrial samples have shown that this IR testing methodology can successfully be employed to test both green-state and sintered P/M compacts.

Mo-Si-B alloys are a leading candidate for the next generation of jet turbine engine blades and have the potential to raise operating temperatures by 300-400°C. The alloys of interest are a three-phase mixture of the molybdenum solid solution (Moss) and two intermetallic phases, Mo3Si (A15) and Mo5SiB2 (T2). A novel powder metallurgical method was developed which uses the reaction of molybdenum, silicon nitride (Si3N4) and boron nitride (BN) powders to synthesize a fine dispersion of intermetallics in a Moss matrix. The covalent nitrides are stable in oxidizing environments up to 1000ºC, allowing for fine particle processing. The process developed uses standard powder processing techniques to create Mo-Si-B alloys in a less complex and expensive manner than previously demonstrated. This powder metallurgy approach yields a fine dispersion of intermetallics in the Moss matrix with average grain sizes of 2-4μm. Densities up to 95% of theoretical were attained from pressureless sintering at 1600°C and full theoretical density was achieved by hot-isostatic pressing (HIP). Sintering and HIPing at 1300°C reduced the grain sizes of all three phases by over a factor of two. Microstructure examination by electron back-scatter diffraction imaging was used to precisely define the location of the phases and to measure the volume fractions and grain size distributions. Microstructural quantification techniques including two-point correlation functions were used to quantify microstructural features and correlate the BN reactant powder size and morphology to the distribution of the intermetallic phases. High-temperature tensile tests were conducted and yield strengths of 580MPa at 1100°C and 480MPa at 1200°C were measured for the Mo-2Si-1Bwt.% alloy. The yield strength of the Mo-3Si-1Bwt.% alloy was 680MPa at 1100°C and 420MPa at 1300°C. A review of the pertinent literature reveals that these are among the highest yield strengths measured for these compositions. The oxidation resistance in air at 1000 and 1100°C was examined. The protective borosilicate surface layer formed quickly due to the close spacing of intermetallic particles and pre-oxidation treatment was developed to further limit the transient oxidation behavior. An oxidation model was developed which factors in the different stages of oxidation to predict compositions that minimize oxidation.

M. Tech. (Department of Metallurgical Engineering, Faculty of Engineering Technology), Vaal University of Technology. / This work is an investigation in the development and characterisation of porous Ti-Sn-SiC and Ti-Nb-SiC composites. Pure Titanium (Ti), Tin (Sn), Niobium (Nb) and Silicon carbide (SiC) powders were used as starting materials. The Ti-Sn-SiC and Ti-Nb-SiC composites were produced by powder metallurgy (PM) press-and-sinter route. The Sn is an α-phase stabilizer while Nb is a β-phase stabilizer in Ti alloys. A systematic study of binary Ti-Sn and Ti-Nb alloys was conducted with the addition of SiC particles. The addition of Sn influences the microstructure of the titanium alloy. With increasing the percentage of Sn content, the density of the samples decreases on the Ti-Sn alloys. An increase in the Sn content from 10 to 25 wt. % content resulted in decreased hardness. The Ti-Sn binary revealed stability of the HCP phase with increasing composition of the Sn content. The porous structures of the Ti-Sn-SiC composites were evenly distributed throughout the materials. The sintered densities increase from 94.69% to 96.38%. XRD analysis detected the HCP crystal lattice structure for the Ti5.4Sn3.8SiC and Ti5.6-Sn3.8-SiC composites. XRD pattern of the Ti5.8-Sn3.8-SiC reveals both the HCP and FCC crystal structures. The HCP phase has lattice parameters a= 2.920 Å; c=4.620 Å with smaller c/a ratio of 1.589. Additionally, FCC lattice parameter a=5.620 Å Fm-3m # 225 was obtained both for Ti5.8Sn3.8SiC and Ti6.0Sn3.8SiC XRD patterns. On the other hand, Optical microscopy analysis of the Ti-Nb alloys revealed the equiaxed grains composed of the light β-phase segregating on the grain boundaries. The Ti9Nb1 has low Vickers hardness of all alloys while Ti8Nb2 and Ti7.5Nb2.5 alloys are harder due to high amount of Nb content. Generally, the densities of the Ti–Nb alloys increased with increasing Nb content. HCP and BCC phases have the lattice parameters a = 2.951 Å, c = 4.683 Å and 3.268 Å, respectively. An HCP (α′) phase was detected in the Ti8.5Nb1.5 alloy with lattice parameters a = 5.130 Å, c = 9.840 Å while a BCC phase had a = 3.287 Å. The sintered Ti8Nb2 alloy also had the α′-phase with a = 5.141 Å, c = 9.533 Å and BCC phase with a = 3.280 Å lattice parameters. On the contrary, the Ti7.5Nb2.5 alloy formed the α′-phase of a = 5.141 Å, c = 9.533 Å and BCC with a = 3.280 Å lattice parameters. For the 10 and 15 wt.% Nb alloys, very porous structures were observed. The pores appear spherical and widely distributed. As the Nb content is increased to 20 wt.% (Ti7Nb2SiC) and 25 wt.% (Ti7Nb2.5SiC), porosity was minimized. The sintered densities of the Ti-Sn alloys are decreasing from 95.90% to 92.80% with increased amount of Sn in the Ti, while the sintered densities of Ti-Sn-SiC are increasing from 94.69% to 96.38%. The high porosity, which developed in Ti7Nb1SiC and Ti7Nb2.5SiC, affected the densities of these composites. The sintered densities of Ti-Nb alloys are increasing from 92.08% to 97.65% with increased amount of Nb in the Ti. In terms of hardness Ti7Nb1SiC and Ti7Nb2.5SiC resulted in the lowest while Ti7Nb1.5SiC and Ti7Nb2SiC composites were 511.74 HV and 527.678 HV. The porosity levels were increased by the addition of SiC in the Ti-Sn-SiC and Ti-Nb-SiC composites. The XRD analysis revealed phase transformation on the Ti-Nb alloys and Ti-Nb-SiC composites.

The development of molybdenum - silicon - boron (Mo-Si-B) composites having a combination of high temperature strength, creep, and oxidation residence has the potential to substantially increase the efficiency of gas turbines. The refractory nature of the αMo, Mo3Si (A15), and Mo5SiB2 (T2) phases results in good strength and creep resistance up to 1300°C. At this temperature, the formation of a borosilicate surface scale from the two intermetallic phases is able to provide oxidation resistance. However, realization of these advantages has been prevented by both a high brittle to ductile transition temperature and difficulty in forming the initial surface borosilicate to provide bulk oxidation resistance. This dissertation addresses two factors pertaining to this material system: 1) improvements to powder processing techniques, and 2) development of compositions for oxidation resistance at 1300°C. The processing of Mo-Si-B composites is strongly tied to their mechanical properties by establishing the αMo matrix, limiting impurity content, and reducing silicon supersaturation. These microstructural aspects control the brittle to ductile transition temperature which has traditionally been too high for implementation of Mo-Si-B composites. The processing here built upon the previously developed powder processing with silicon and boron nitrides which allowed for a low oxygen content and sintering of fine starting powders. Adjustments were made to the firing cycle based upon dew point measurements made during the hydrogen de-oxidation stage. Under a relatively high gas flow rate, 90% of the total water generated occurred during a ramp of 2°C /min between 450 and 800°C followed by a hold of 30 minutes. The oxidation resistance of Mo-Si-B composites was studied for a wide range of compositions. Silicon to boron atomic ratios were varied from 1 to 5 and iron, nickel, cobalt, yttria, and manganese were included as minor additions. In all these compositions, the αMo volume fraction was kept over 50% to ensure the potential toughness of the composite. For the oxidized surface glass, a silica fraction of 80 to 85% was found to be necessary for the borosilicate to have a sufficiently high viscosity and low oxygen permeability for oxidation resistance at 1300°C. For the Mo-Si-B bulk composition this corresponds to a Si/B atomic ration of 2 to 2.5. Higher viscosity compositions failed due to spallation of poorly attached, high silica scales. Lower viscosity compositions failed from continuous oxidation, either through open channels or repetitive MoO3 bubble growth and popping. Additionally, around 1% manganese was necessary for initial spreading of the borosilicate at 1300°C. In conjunction with flowing air to prevent MoO3 accumulation, oxidation weight loss rates below 0.05 mg/cm2-hr were measured. Finally, a theory is proposed here to describe the mechanisms responsible for the development of oxidation resistance. This theory involves three stages associated with: 1) generation of an initial surface borosilicate, 2) thickening of the borosilicate layer, and 3) slow parabolic oxidation controlled by the high silica surface scale.