Design of Al diffusion coatings for Fe-based and Ni-based alloys

Components, which consist of metallic materials and operate at high temperatures, can degrade due to high temperature corrosion attack. The most frequently occurring phenomenon is oxidation. When metallic materials are exposed to an oxidizing atmosphere, it is desired that elements, which are called protective oxide forming elements, diffuse from the material interior to the material surface and form a dense and slow-growing oxide scale, which acts as a diffusion barrier and decelerates further oxidation. Since for several reasons most high temperature alloys contain only a limited amount of protective oxide forming elements, coatings are applied with higher amount of these elements. The concept of the diffusion coatings is to enrich the substrate surface with one or more protective oxide forming elements (e.g. Al). Aluminization of Fe- and Ni-based alloys leads to Al diffusion coatings, which can consist of one phase or several stacked phases, depending on the Al activity within the intermetallic phase, according to the Fe-Al and Ni-Al phase diagrams. In this work, the aim was to develop a predictive design procedure for the manufacturing of pack cementation Al coatings on austenitic steels and Ni-based alloys. The pack cementation process is a CVD (chemical vapour deposition) process, where the substrate to be coated is embedded in a powder mixture, consisting of the deposition element (e.g. Al), an activator (e.g. NH4Cl) and a filler (e.g. Al2O3) and is heated in a tube furnace for several hours in an Ar/5%H2 inert atmosphere. The coating design is based on thermodynamic and kinetic considerations of the pack cementation process. Thermodynamic considerations were conducted by calculations with the thermodynamic software FactSage®, to determine the Al activity (total partial pressure of Al carrying halides) within the pack powder as a function of process temperature and powder composition. Furthermore, the determination of the full range of the binary phase diagram of the Fe-Al and Ni-Al systems and the Al activities of these systems were calculated as a function of temperature and mole fraction. Kinetic values, as the diffusion coefficient, which affect the resulting coating thickness, have been determined via a limited amount of experiments for each alloy system, followed by Matano analysis. It was shown that based on the model considerations and the collection of the thermodynamic and kinetics data for a material/deposition element couple, the coating design approach developed enables a quantitative prediction and adjustment of the resulting coating properties (intermetallic phases and coating thickness) for a wide range of process parameters. Coating experiments on austenitic steels (AISI 321, AISI 314 and Alloy 800) and a Ni-based alloy (Alloy 601) have shown that low pack process temperatures (up to 900°C) promote the formation of HA coatings. This observation is in agreement with the thermodynamic calculations, but also kinetic considerations show that a high process temperature promotes the interdiffusion of Al from the coating to the interior of the material during the coating process, which promotes the formation of LA coatings. The coating design postulates that it is possible to determine the entire kinetic values (the pre-factor D0, the activation energy EA and the constant q) for a deposition element/substrate couple by means of three “calibration” pack experiments at three different temperatures. The determined diffusion coefficients for the coating procedure on AISI 321, AISI 314, Alloy 800 and Alloy 601 at 800, 900 and 1000°C showed good agreement with literature values. The collection of the entire thermodynamic and kinetic information made it possible to predict the coating microstructure for these four materials and to compare the predicted and experimentally formed coating properties, which showed good agreement. On the other side, the coating design contains limitations. For example, an extensive activator amount in the pack, which would theoretically cause a higher Al activity in the pack, leads to an attack of the substrate by the hydrogen halides. The co-deposition of another element to the main deposition element Al reduces the Al activity within the pack, since the activator is consumed by both deposition elements. Experiments have shown that Si and Hf can be co-deposited to an Al coating. The coating thickness is reduced in comparison to a mono-element Al coating, which is not only caused by the Al activity reduction due to co-deposition. Diffusion coefficient determinations of Al and the co-deposition elements (Si and Hf) have shown that also the diffusion coefficient of Al was reduced, because the co-deposition element occupies Al lattice sites. Cyclic oxidation experiments in an oxidizing and reducing atmosphere at 1000°C have shown that Si co-deposited Al coatings enhance the high temperature corrosion resistance, since the Al activity within this two-element coating is lower in comparison with the mono-element Al coating. Thus, the Al interdiffusion to the interior substrate and the coating brittleness is reduced. Also, Hf co-deposited Al coatings have shown an enhancement compared to the mono-element Al coating in a way that oxide scale thickness is lower, which indicates a slower oxide scale growth rate. The design concept has successfully been applied to a combustion chamber in a reformer system and is available for further use in coating technology