Tailoring Grain-Boundary Segregation to Control Mechanical Properties
- PDF / 2,674,351 Bytes
- 8 Pages / 417.6 x 639 pts Page_size
- 82 Downloads / 173 Views
complement the analysis. However, the limitations of AEM are also important, including 1) the complete lack of depth resolution, 2) the need to thin the specimen and 3) the poor sampling (few boundaries can be analyzed). Typically AEM studies have made a few point analyses, or line profiles across specific points on a few grain boundaries While limitationsl) and 2) are fundamental to the TEM, we have recently developed quantitative high-resolution X-ray mapping (XRM) [3, 4] and applied it to the study of segregation of Bi in Cu and Cu in Al [5, 6] as shown in Figure 1.
6-
4-
32 0
A
5
10
distance (nm)
15
20
B
Figure 1. A) a quantitative X-ray map of the distribution of Cu (from 0.10 wt. %) on many grain boundaries in fine grained A1-4 wt. % Cu thin film. OverlO boundaries are visible in this one image. B) a line profile of the Cu variation across the boundary, which can be extracted from any point or region on any of the grain boundaries. Courtesy D.T. Carpenter. XRM is a fundamentally new tool for the study of GB segregation. Such maps provide direct correlation between electron images, electron diffraction and segregant distribution, and immediately reveal the well-known inhomogeneity of segregation, which cannot be studied fully with AES or other surface techniques, since low segregant levels may not result in intergranular failure. As shown in FigurelA, any inclination of the GB to the electron beam is also reflected in a wider projected image of the segregant. Segregation levels for many GBs can be determined from a single map, which is a significant improvement over the more traditional methods of
point analyses or line profiles taken across a specific point. As shown in FigurelB, it is also possible to extract a line profile across any given point in the many boundaries imaged in the map. Thus, the XRM technique combines the best aspects of AEM and AES. As noted, AES, while offering analysis of many GBs, requires brittle boundaries, which must be fractured in situ prior to study. Therefore, GBs with low segregation levels are not amenable to study by AES if the boundary does not fracture. Now, the first time, XRM can map quantitatively the composition of segregant(s) on large numbers (>100) of undisturbedboundaries [6] (i.e. we do not have to fracture the specimen). Both brittle and ductile material can be studied with good (-2 nm) spatial localization. Furthermore, in the same instrument, we can relate the amount of segregant to changes in the GB misorientation and plane, with standard diffraction techniques. Knowing the distribution of the segregants accurately is not, however, sufficient to answer the question why certain segregants cause brittle failure and why others do not. Losch [7] and Messmer and Briant [8] first proposed that GB embrittlement was due to bond hybridization or charge transfer respectively, although at that time there was no experimental technique available to test the theories. In their recent review, Hofmann and Lejcek [9] conclude that one of the fundamental questions th
Data Loading...
