ICMCTF2013 Session A1-1: Coatings to Resist High Temperature Oxidation, Corrosion and Fouling

Multiphase Mo-Si-B alloys with compositions, that yield the ternary intermetallic Mo₅SiB₂ (T₂) phase as a key microstructure constituent together with the Mo and Mo₃Si phases, offer an attractive balance of high melting temperature, oxidation resistance and mechanical properties. Mo-Si-B alloys respond to high temperature oxidation in two distinct stages. First, there is a transient stage with an initial high recession rate that corresponds to the evaporation of volatile MoO₃ due to the oxidation of the molybdenum rich phases. The steady state stage of the

oxidation begins when a borosilica layer that initiated in the transient period becomes continuous and protects the alloy from further rapid oxidation. Then, the oxidation rate is limited by oxygen diffusion through the borosilicate layer. In order to improve the oxidation performance of the Mo-Si-B alloys, it is necessary to minimize the transient stage. The three phases, Mo (solid solution), Mo₃Si (A15) and Mo₅SiB₂ (T₂), composing the Mo-Si-B alloys play different roles in

the transient stage. The interaction of the three phases with a reduced microstructure scale can reduce considerably the transient oxidation stage. As a further approach to inhibit the transient stage, a kinetic biasing strategy has been developed to capitalize on the reactions between different phases to develop useful reaction products and alloy compositions that evolve toward a steady state of a compatible system. In order to achieve a compatible interface coating together with enhanced oxidation resistance, a pack cementation process has been adopted to apply

diffusion coatings. From this basis kinetic biasing is used together with pack cementation to develop Mo-Si-B based multilayered coatings with an aluminoborosilica surface and in-situ diffusion barriers with self-healing characteristics for enhanced oxidation resistance. While a combustion environment contains water vapor that can accelerate attack of silica based coatings, the Mo-Si-B based coatings provide oxidation resistance in water vapor up to at least 1500°C. An exposure to hot ionized gas species generated in an arc jet confirms the robust coating performance in extreme environments. To extend the applications beyond Mo-based systems a two-stage process has been implemented to provide effective oxidation resistance for refractory metal cermets, SiC and ZrB₂ ultra-high temperature composites.

In the present work, both coatings and intermetallic single phases Ti₃X₃CrSi₆ were elaborated by uniaxial hot pressing. Their characterization (SEM+XRD) showed that the nature of the metal X influenced on the chromium content. Isothermal oxidation behavior at 1200°C was investigated for 100h in thermobalance using industrial dry air (1,5L/H). Steam oxidation test (=7,5% using a flow of 2,5L/h) at 1200°C during 100h and 500h were also carried out. XRD analysis and FEG-SEM observations highlighted that all phases developed a duplex chromia and silica protective layer, with a ratio chromia/silica depending on the nature of X. Formation of low chromium content phases under oxide scale (TiXSi₂ or Ti₄X₄Si₇) due to the selective oxidation of chromium were detected, ensuring the development of a thin protective silica layer in dry oxidation conditions. Under water vapour atmosphere, the protective behaviour of the oxide scale was function of the nature of X. Indeed Co- and Ni-containing specimens developed pure chromia during oxidation whereas Fe ones develop (Cr_xFe_1-x)₂O_3, reducing drastically the CrO₃ volatilization from the upper chromia scale, and ensuring the durability of the silica layer.

Photo-stimulated luminescence spectroscopy (PSLS), 3D microscopy and focused ion beam (FIB) SEM evaluations were made at increasing cyclic oxidation exposures on the same region of simple and Pt-modified aluminide bond coatings on several superalloy substrates. Each sample coupon was cut in half and one half was tested in 10%H2O and the other in dry air in order to understand the effect of water vapor on samples with a thermal barrier coating. With one exception, water vapor did not increase the roughness of the bond coating surface over the increase due to thermal cycling in air. The roughness increase for all samples was due to large and small bond coating grains rising and sinking, respectively, with thermal cycling. This caused the compressive stress in the Al2O3 scale to decrease on the bond coating grain boundary regions eventually leading to scale cracking in the simple aluminide bond coatings. Water addition retards the θ to α-Al2O3 phase transformation in the Pt-modified bond coatings but it is unclear if this affects scale adherence at later stages. Stress histograms produced by PSLS mapping help to elucidate Al2O3 scale damage accumulation in each sample while not identifying a consistent difference for all samples between wet and dry conditions.

NiAl-based intermetallic compounds have attracted increasing attentions because of their promising potential as candidates for metallic coatings or the bond coats in thermal barrier coating (TBCs) to protect the underlying superalloy against high-temperature oxidation and corrosion. Some important aspects should be considered before NiAl compounds are explored as the protective coatings for advanced single crystal (SC) superalloys, such as poor oxide scale adherence of NiAl coating when subjected to cyclic oxidation and severe interdiffusion between the coating and SC superalloy during high-temperature exposure. It is well admited that Secondary reaction zone (SRZ) and need-like topologically-closed packed phases (TCP) formed in SC alloys, partially due to interdiffusion, would result in a significant degradation of mechanical properties of the alloys. In this work, minor reactive elements doped RuNiAl coatings were produced on SC alloys by a combination of electro-plating and electron beam physical vapor deposition (EB-PVD). Cyclic oxidation and interdiffusion of the RuNiAl coated alloy were researched. 0.05at% Dy doped RuNiAl coating revealed not only much improved cyclic oxidation life but also lower oxidation rate as compared to the NiAl coating. And, Dy and Hf co-doping could even behave better in improving cyclic oxidation life than single RE doping. In the RuNiAl coated alloy, SRZ and needle-like TCP phases didn’t occur after 200 h annealing at 1373 K, whereas SRZ and TCP was observed in a NiAl coated alloy after 100 h annealing, with a thickness of ~100 μm, indicating that the RuNiAl coating effectively suppressed the formation of SRZ and TCP as a diffusion barrier. The associated mechanisms for the diffusion barrier effect of the RuNiAl coating was discussed via diffusion couples.

Keywords: NiAl; Oxidation; Diffusion; Secondary reaction zone (SRZ); Reactive element effect (REE).