Low mass platinum aluminide bondcoat for thermal barrier coating

Abstract

During the last 30 years, Thermal Barrier Coating systems (TBCs) have been extensively used to protect the hottest part of aero-engines. They can extend significantly the lifetime of high pressure turbine blades and combustor walls by decreasing the superalloy substrate temperature by up to several hundreds o f degree C. TBCs are duplex systems consisting of a thermal insulative ceramic toplayer and an intermediate metallic bondcoat layer, whose function is to protect the substrate against corrosion and oxidation and to promote the ceramic adherence by forming an alumina scale at the interface with the ceramic. The lifetime of the TBCs is however limited by chemical, mechanical and thermal stresses in the coatings due to bondcoat oxidation and the mismatch of thermal expansion coefficient (CTE) between the ceramic, the bondcoat and the substrate. The bondcoat consideration is therefore of a substantial importance for the TBCs lifetime extension, and the present work has been focused on the development of a novel and innovative intermetallic overlay bondcoat, having a much thinner thickness than conventional bondcoats, acting as a diffusion barrier for substrate harmful elements, and promoting the formation of a pure, slow-growing and adherent alumina scale. The low-mass bondcoat system has been based on a 3-15 microns thick PtAh intermetallic layer, with the ternary addition of a reactive element (Hafnium, Zirconium, or Yttrium). Aluminium and Platinum are deposited sequentially by the sputtering process (Physical Vapor Deposition). The bondcoat is thus a multi-layer coating, and the layers react one with another exothermically by diffusion after a subsequent heat treatment at a relatively low temperature. The temperature of reaction between the layers and the stability of the obtained intermetallics has been studied by using Differential Thermal Analysis. Different platinum aluminides have been developed as bondcoats and the number of layer has been varied (up to 350 layers) in order to study the influence on the coating structure. Finally, the most successful systems have been cyclically tested to be compared to industrial bondcoats systems. These experimentations have led to the development of a highly controllable bondcoat deposition and formation process. Different morphologies and compositions can be accurately obtained by varying the individual layer thickness and Al/Pt thickness ratio within the coatings. A reactive element, which consists of either zirconium, yttrium or hafnium has been introduced into the aluminium layer by sputtering co-deposition and it has been therefore demonstrated the possibility of improving the efficiency of the low-mass bondcoat by adding such an element evenly through the coating. Whatever the composition or its structure, the low-mass bondcoat is adherent to the substrate and does not interact with the substrate during the deposition and the formation process. The bondcoat is thermally stable for a significant time of aging at 700°C, 900°C and 1100°C, but do not withstand cyclic oxidation testing better than industrial bondcoats. Nevertheless, to really assessed the potential of the low mass bondcoat, a cyclic oxidation test has to be performed after ceramic topcoat deposition, which would modify the local stress gradients on the thermally grown oxide, during cooling.