Characterization of TiAl Alloys by Secondary Ion Mass Spectrometry

Abstract

Titanium aluminides are arising as a valuable alternative to superalloys in applications where the ratio resistance/density is important. Together with excellent mechanical and corrosion properties at high temperatures, such characteristics are very attractive for applications in the aeronautical, aerospace and automotive industries. However, the current high selling price, due to high costs of production and raw materials and the need of very specific equipment, are limitative factors for further applications. With the end of the cold war, and the decrease of traditional markets of TiAl, the strategy to develop other applications, strongly depends on the decrease of production costs. An alternative to the present production routes might be the use of traditional casting techniques, by induction melting of the alloy in a ceramic crucible and pouring into ceramic moulds, made by the investment casting process. However, due to the high reactivity of Ti alloys, the use of traditional ceramic materials cannot be used, as they lead to oxide formation and oxygen pick up both from the crucible and the moulding materials. In this work, the relative oxygen concentration of Ti-48Al castings was measured by SIMS — Secondary Ion Mass Spectrometry. This technique provides a direct measurement of the isotopic composition with high sensitivity. The cylindrical samples were specially prepared to allow the analysis of the area close to the border. Oxygen profiles were acquired for samples obtained by with different mould materials. The comparison of such profiles with hardness profiles, give insight in the significance of the oxygen concentration in the properties of the alloy and in the choice of the most suitable materials for TiAl production Introduction During the last decade several intermetallic alloy systems have been identified as potential candidate materials for applications at high temperatures. Among those systems, γ-TiAl was found to be a very promising material, but its currently high cost limits significantly its successful industrial-scale application. Besides, titanium aluminides suffer from quite low ductility at room temperature, and their application is always dependent from the microstructure modification. The best compromise between high temperature properties and room temperature ductility, is achieved with a two-phase microstructure (α2 + γ), for a volume ratio α2 /γ between 5 and 15%. This is achieved by decreasing the aluminium content of the alloy, and adding small amounts of alloying elements, such as Cr, Mn, V and B, which were found to refine the grain size and stabilize the lamellar structure [1, 2]. On the other hand, γ-TiAl is extremely reactive with almost every known elements, with particular emphasis to oxygen and nitrogen, which dissolves interstitially, leading to a strong hardness increase and alloy embrittlement. The maximum solubility of oxygen in the γ phase is about 250 at.ppm and more than 2,1 at.% in α2, and it does not depend on the γ-TiAl alloy composition [3]. According to some researchers [1,3], such difference in the oxygen solubility limits is the most important factor to explain the higher ductility of two-phase γ−TiAl alloys, when compared with single phase γ-TiAl: since the α2 phase tends to contain more interstitials, namely oxygen, than the γ phase, a scavenging effect in this phase, lowering its interstitials concentration, results in an increase on the dislocation mobility on this phase, and an increase in the alloy ductility. On the other hand, in single phase γ-TiAl, as soon as the oxygen concentration reaches 250 at.ppm, precipitation of titanium oxides starts, until total consumption of oxygen. On two-phase γ-TiAl alloys, the excess of oxygen is absorbed in α2, avoiding oxides formation, and enhancing ductility. As TiAl is extremely reactive, its production by foundry processes using traditional crucible and mould materials (usually metal oxides) is a difficult task, because ceramic-alloy interactions are always present, in a certain extent. As a consequence, it is almost impossible to obtain titanium alloys castings free from contaminants. During the last five years, authors have done extensive research work on this field, and a melting and moulding technique, using ceramic materials, was developed, and results published elsewhere [4, 5]. The main purpose of such work was to develop a casting procedure, with which sound γTiAl components could be produced at competitive prices without, or with controlled residual elements contamination. Metal-mould interaction After pouring, and during the cooling and solidification stages, metal-mould interactions might occur. While the metal is in the liquid state, it can dissolve or react with the ceramic mould, into a certain extent. As a consequence, the oxygen and other elements picked up from the mould material, or the reaction products, go into solution with the liquid metal. When solidification starts, the interaction continues, and composition gradients start to develop, from the interface to the inside of the castings. Such gradients lead to diffusion mechanisms, in order to reach the equilibrium situation. As the temperature decreases, diffusion rates become smaller, until they become negligible at low temperatures. As a consequence, chemical composition gradients can be found, at room temperature, from the surface to the inside of castings, leading to significant changes in their mechanical and technological properties, like a surface “hard case”, usually called “alpha-case”. As the diffusion rate of oxygen in TiAl is very high, and its presence is known to affect the ductility and hardness of the alloy, its measurement and the knowledge of its variation, are crucial to select proper ceramic materials, for crucible/mould production. Samples preparation Ti-48Al was induction melted in multi layered ZrO2+Y2O3 crucibles, and cylindrical shaped samples were poured in ZrO2, SiO2 and multi-layered ZrO2+Y2O3 moulds (ZrO2 based, with Y2O3 inside contact layer), using the techniques described elsewhere [4, 5, 6]. Microhardness was evaluated on a Shimadzu hardness tester, using a 50g load, for 15 seconds. Overall oxygen content was measured by the IGF technique (Inert Gas Fusion), and SIMS (Secondary Ion Mass Spectrometry) was used to determine the diffusion profile of oxygen. Cast specimens were cylinders 20 mm in diameter and 85 mm long, and were poured at about 1550oC. Samples for characterization were collected from the middle section of them, by cutting the cylinders at half their height, and prepared using traditional metallographic techniques. As oxygen measurement was needed at distances from the samples surface less than the beam diameter (50 μm), samples were specially prepared for SIMS characterization by cutting through a plane making a 15o angle with the cylinder axis as described in Figure 1. In this way, analysis could be performed in several points in depth, without the need of long erosion times to reach the bulk. Experimental procedure SIMS has been performed in a Multitechnique Surface Analysis System detailed described elsewhere [7]. A 4 keV Ar + , 200 nA beam focused in a 50μm diameter was used. Secondary ions produced by sputtering where mass analyzed by a quadrupole type mass spectrometer. Both positive and negative spectra were taken. Base pressure during analysis was in the order of 10 -9