A selected area channeling pattern study of the correlation between boron content of the grain boundary and grain boundary structure in substoichiometric boron-doped Ni3Al

Author

Yee C. Lin

Date of Award

8-1991

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Metallurgical Engineering

Major Professor

Charlie R. Brooks

Abstract

The intermetallic compound Ni₃Al is very attractive as a new structural alloy since it shows an increase in strength and stability at elevated temperature. However, the low symmetry crystal structure leads to a very low ductility, and this alloy is prone to intergranular brittleness. Although single crystals are ductile, the polycrystalline alloy fractures with almost no ductility. However, it has been found that the addition of small amounts of boron to a slightly nickel-rich Ni₃Al alloy dramatically improves the ductility. The boron in these ductile alloys is found to segregate to the grain boundaries. Choudhury carried out a detailed study of the effect of bulk boron content and heat treatment on the grain boundary boron content. He found that increasing the bulk boron content increased the amount of boron on the grain boundaries and reduced intergranular fracture. Also in his study, he observed a variation in boron content from one grain boundary to the other for a given sample, and even with the location of the analysis on a given curved grain boundary. Thus the question arose as to whether these variations were due to experimental errors or were caused by the effect of the grain boundary structure on the equilibrium grain boundary boron content. One of the purposes of this research was to address this question by determining if any correlation exists between the boron content on the grain boundaries and grain boundary structure in Ni₃Al. At the beginning, this research relied on using Auger electron spectroscopy (AES) samples (which had been fractured in a Auger electron spectroscopy instrument and the intergranular fracture facets analyzed for boron content) from the research of Choudhury. Using Choudhury's samples, the orientation of mating grains of known grain boundary boron content would be determined from electron channeling patterns of the intergranular fracture facets. From this, the boundary structure would be determined, and related to the already experimentally determined grain boundary boron content. To proceed in this research, the plastic deformation on Choudhury's samples surfaces induced by fracture had to be eliminated. To solve this problem, a lot of effort was expended in annealing the Auger electron spectroscopy samples from 500°C to 1115°C. The annealing time varied from a few minutes to several weeks. However, to have the strain removed without the occurrence of recrystallization was very difficult, and recrystallization would completely change the structure of the grain boundary. Another problem was the presence of oxides picked up at high temperature, and which would retard obtaining a sharp channeling pattern. To remove the oxide layers, samples were electropolished. However, the fracture surface topology of the fractured sample was rounded after electropolishing. After numerous annealing and electropolishing processes had been carried out, it was concluded that the annealing technique did not successfully accomplish the required recovery process without recrystallization on Choudhury's samples. Then, another method was carried out to correlate the boron content to the boundary structure. A Ni₃Al alloy (IC-50) doped with boron was obtained from Oak Ridge National Laboratory. Samples were first heated at 11150°C for 48 hrs (to increase the grain size (up to 50-100μ)). water quenched, then annealed at 700°C for 6 hrs to enhance the boron content on the grain boundaries. Samples were then electropolished to remove the surface strain induced by mechanical polishing. The determination of the orientation was accomplished by calculation from the selected area channeling patterns obtained from a Cambridge Stereoscan 360. The Ni₃Al alloy samples were hydrogen-charged and then fractured in an Auger apparatus at room temperature. The boron contents of individual grain boundaries on fractured surfaces were then measured along the edge of the sample. The deviation of the boron/nickel ratio was about +/- 20%. The reproducibility of the Auger data is acceptable for qualitative analysis in this study. The last step was to match the grain boundary character of one half of the AES sample to that of the Scanning Electron Microscopy micrograph taken from the metallographic AES sample before fracture. This allowed location and identification of the grains on either side of the crack. After the trace of the crack was determined, the boron content on a specific grain boundary was correlated to the grain boundary structure which was characterized from the channeling patterns of two corresponding adjacent grains. This is the first experimental study of the correlation of the segregation of boron with the grain boundary structure in Ni3Al systems. All the grain boundaries analyzed in this research were random, which was determined by the calculated Δθ/Δθc ratios. The most close-to-ordered grain boundary has a Δθ/Δθc ratio of 1.3. The result of this study quite matches the observation of Hanada et al. whose study indicated that the cracks of two Ni₃Al samples were opened up mainly along random (or disordered) grain boundaries. The result indicates that the lower the Δθ/Δθc value (more ordered), the higher the boron/nickel ratio, which suggests that the more ordered grain boundaries absorb more boron.

Recommended Citation

Lin, Yee C., "A selected area channeling pattern study of the correlation between boron content of the grain boundary and grain boundary structure in substoichiometric boron-doped Ni3Al. " PhD diss., University of Tennessee, 1991.
https://trace.tennessee.edu/utk_graddiss/11165