Phase equilibria in the Ni-Al-Ti system at 1173 k
- PDF / 1,067,349 Bytes
- 4 Pages / 612 x 792 pts (letter) Page_size
- 53 Downloads / 223 Views
I.
INTRODUCTION
A L L O Y systems based on Ti and on Ni are of great importance for high temperature applications, particularly in the aerospace industry. Alloy development in this field relies on having a data base of phase equilibria which is readily available to the alloy designer. However, there is a deficiency of phase equilibria data in these types of alloys, and some of the available data is from early work when direct phase analysis was not available. In view of this, a determination of the phase equilibria in the Ni-A1-Ti system was undertaken using modern techniques to establish the phase boundaries accurately. A recent review of the Ni-A1-Ti system 1 pointed out the lack of data, especially in the regions away from the binary sides and corners. None of the experimental studies determined phase boundaries directly but used X-ray diffraction or optical microscopy which, in view of the complexity of the phase equilibria, cannot give an accurate representation of the constitution of this system, in addition, there is some discrepancy in the nickel-rich region between the calculated phase equilibria 2 where /32 + "0 ~- yt and /32 -~- H + y ' regions were found and experiments 3 where rt + y ' + H and/32 + H + rt regions were found.
II.
equilibration anneal at 900 ~ All samples were equilibrated at 900 ~ for one week and quenched into iced water. The specimens were mounted and polished to a 0.25 p,m alumina finish and etched in Kalling's reagent. Since the phases present have relatively high hardness, it was not considered likely that smearing of the phases would occur during mechanical polishing. The specimens were then examined by optical microscopy, SEM, and in some cases X-ray diffraction. Wavelength dispersive analysis of each phase present in the unetched specimens was performed using a JEOL 733 Superprobe and the results were corrected using a computer program Magic IV. Each phase was analyzed in three different locations and the results averaged with only those results being within 2 pet of the 100 pet sum of the concentrations being accepted. Analysis was performed well away from any inclusions which might have been picked up during specimen preparation. The 'as cast' structures were examined optically and by SEM to determine the fields of primary crystallization.
III.
RESULTS
Table I lists the alloy compositions used in this study together with their designations. The 1173 K isothermal section is shown in Figure 1 and the schematic liquidus in
EXPERIMENTAL PROCEDURE
The materials used to produce the alloys for this study were VP grade titanium and aluminum rod from Materials Research Corporation and 99.99 pct Ni shot from INCO Ltd. The pure materials were argon arc melted on a water cooled copper hearth using a nonconsumable tungsten electrode. Each alloy was produced as an approximately 20gram button which was remelted three times, turning it over between melts in order to ensure homogeneity. Part of each button was cut off for examination of the 'as cast' structure while the remainder was
Data Loading...
