Turbine blade coatings feel the heat
Molten deposit damage to gas turbine engine coatings poses a continuous challenge to the aerospace industry. In his PhD research at Cranfield University, UK, Ngunjoh Ndamka, explored how a new technique could help to assess degradation mechanisms and severity of damage, to improve engine efficiency.
Demand for bigger, better and faster air travel has never been greater. And thanks to new studies investigating the degradation mechanisms of the coatings applied to engine turbine blades, this has never been more achievable.
The aerospace industry has seen much success over recent years in developing low thermal conductive (low k) thermal barrier coatings (TBCs) to protect aero and industrial gas turbine engine components operating in the hot section (blades, vanes, seals and combustion panels) from thermally induced damage. However, attacks on these coatings by deposits entering the engine during flight remains an obstacle to the adoption of low k TBC technology, and the benefits it brings in terms of specific fuel consumption (SFC) and greenhouse gas emissions cannot be fully implemented.
As well as allowing higher engine operating temperatures, a secondary function of TBCs lies in restricting the access of deposits and contaminants to the substrate materials, which improves engine durability and energy efficiency. During flight, environmental contaminants such as airborne dust, sand, fly ash, volcanic dust, concrete dust and fuel residue are sucked into the aeroplane’s engine. Under certain turbine operating conditions, these materials accumulate on the engine turbine’s hotter surfaces – generally on the concave side of the blade – and melt when the operating temperature reaches 1,240°C. As the predominant oxides produced are that of calcia (CaO), magnesia (MgO), alumina (Al2O3) and silica (SiO2), these molten deposits are generally termed Calcia-Magnesia-Alumina-Silicates (CMAS).
CMAS composition varies from region to region across the world, be it from runway debris, sand or a volcanic eruption. However, different types of deposit melt at different temperatures, and this lack of continuity has proven problematic for the aerospace industry when attempting to find a solution to degradation of turbine blade coatings.
CMAS degradation of TBCs has been extensively studied, with researchers reporting a similar degradation mechanism with varying degrees of severity. Historically, mitigating solutions to TBC degradation caused by CMAS have mainly been reactive, rather than proactive. Incorporating molten degradation challenges during the design stage is now required. In an effort to reduce erosion damage, engine manufacturers are now customising engine hardware depending on the part of the world in which an aircraft will operate. To aid this transition from a reactive to a predictive approach, a recent study at Cranfield University, in Beford, UK, looked at a new approach to address this issue – the concept of Basicity Index (BI).
The Basicity Index (BI) model is commonly used in industries including mining, to detect the quality of metallurgical coke, and in welding, where it is used to describe the impact toughness of the weld metal. The Cranfield researchers used a modified version of BI to predict the severity of engine TBC damage expected in various regions of the world, based on the aggressiveness of the respective sand chemistry.
Back to basicity
The amphoteric oxides of a CMAS melt exhibit either acidic or basic roles. As illustrated in the equation above, BI represents the molar ratio between basic and acidic oxides, and is calculated by dividing the sum of the percentages of the basic constituents by the sum of the acid constituents.
Airborne contaminants entering gas turbine engines will either be a low, near-neutral or high basicity deposit. Where a CMAS melt is basic in nature, the amphoteric oxides behave as an acid until the BI reduces to 2. Where a melt is acidic, the amphoteric oxides behave as a base until the BI increases to 2. In neutral melts, amphoteric oxides are considered neutral. Three categories of CMAS compositions are identified based on soil compositions across the world – low BI refers to BI<1.8, mid-range BI (near-neutral deposits) to 1.8–2.2, and high BI to 2.2 or above. The neutral BI is based upon the neutral composition of CaO-SiO2 dominant melt. In a binary CaO-SiO2 slag system, the neutral composition corresponds to the formation of the ortho-silicate composition 2CaO-SiO2, hence the mid-BI value of 2. Understanding the behaviour of these individual categories enables prediction of the likely degradation morphology of zirconia-based TBCs from various parts of the world. In previous studies, BI of the CMAS has been classed as a single unit (mechanisms observed and severity of damage), which has led to variations in the results (see graph opposite). To avoid this, the Cranfield study used a modified version of the BI model that not only evaluated the degree of damage caused by molten deposits to TBCs, but also categorised different modes of attack. Applying this concept to previous authors’ work on CMAS, the study calculated BI for molten deposit composition and variations within those compositions, with the following results.
Low BI: acidic deposit
Almost all CMAS compositions used in previous studies fall into this BI category. The thermochemical mechanism for this mode of attack has been extensively studied and the observed damage mechanisms have always been very similar. The results observed in the Cranfield study for this class of deposit are similar to those published, as illustrated by micrograph images above. The micrograph on the left shows degradation of the TBCs at the columnar tips and along the column boundaries, with little degradation within the bulk of the coating and towards the thermally grown oxide. This commonly observed type of degradation is termed classic CMAS attack and matches that described by other authors, with the Y2O3 leaching out of the zirconia columns and forming dark globules of CMAS reacted phase.
Mid-range BI: Near-neutral deposit
Mid BI CMAS was deposited onto the TBC surface and heat treated for four hours at 1,480°C. The resulting micrograph shows the different morphology of attack observed with this near-netural deposit. Here, the attack is mainly along the TBC columns, with almost no dark, globular pores of CMAS reacted phases that were observed in the low BI attack. These deposits also showed a change in damage mechanism from primary column attack to columns, plus sub-boundary attack through the TBC thickness. This results in some recrystallisation within the innermost part of the columns, forming more equiaxial grained microstructures.
High BI: basic deposit
The micrograph shows complete dissolution of the entire columnar coating after four hours at 1,458°C, as well as recrystallisation through the coating thickness. The presence of significant amounts of zirconia in the melt pool confirms the partial or complete dissolution of bulk TBC into the CMAS melt, which in turn changes the local chemistry of the melt. The resulting microstructure consists of dense ceramic equiaxial, rounded grains. This can be attributed to the relative wetting characteristics of the TBC/CMAS system that effectively minimise its surface energies.
These results strongly suggest that wetting behaviour of zirconia-based TBC and CMAS system can be expected to vary with the BI of the slag/ deposit. Using the BI modelling tool to classify different types of deposits, it is therefore possible to predict the severity of damage each is likely cause to a coating. This will allow gas turbine engines to be custom-designed to mitigate a specific type of corrosion according to the region of the world in which the plane will operate.
In light of the desire to increase turbine entry temperature, BI is expected to play a vital role in the design and manufacture of the aerospace engines of the future.
Ngunjoh Ndamka is Advanced Materials Technologist at Rolls-Royce, UK. For more information, email Ngunjoh.Ndamka@Rolls-Royce.com