No abstract

No abstract

As a result of recent investigations on nanocrystalline (nc) materials, extensive experimental data on the deformation behavior of these materials have become available. In this article, an analysis of these data was performed to identify the requirements that a viable deformation mechanism should meet in terms of accounting for the mechanical characteristics and trends that are revealed by the data. The results of the analysis show that a viable deformation mechanism is required to account for the following: (1) an activation volume the value of which is in the range 10 to 40 b 3 ; (2) an activation energy that is close to the activation energy for boundary diffusion but that decreases with increasing applied stress; (3) the magnitudes of deformation rates that cover wide ranges of temperatures, stresses, and grain sizes; (4) inverse Hall-Petch behavior; and (5) limited ductility. The validity of available deformation mechanisms for nc materials is closely examined in the light of these requirements.

Electrodeposited nanocrystalline (nc) Ni having an average grain size of 20 nm was annealed at 443 K for different holding times. An examination of the microstructure following annealing showed three important features. First, all annealed samples exhibited abnormal grain growth, which was manifested by the presence of large grains that were surrounded by regions of small grains (bimodal grain distributions). Second, annealing twins existed in the large grains of the samples that showed a bimodal grain distribution. Third, by estimating the density of annealing twin, it was found that annealing nc-Ni at 443 K resulted in a maximum twin density after 5h. Following annealing treatment, specimens with different volume fractions of twins were tested under uniaxial tension at 393 K and a strain rate of 10-4 s-1. The results showed that both strength and ductility in nc-Ni attained maximum values after annealing for 5h. The role of both bimodal grain distributions and annealing twins in enhancing ductility and strength was discussed.

Good sample preparation is essential for acquiring successful electron backscattered diffraction (EBSD) patterns in the SEM. Mechanical polishing to obtain the required surface quality with minimal sub-surface defects and deformation that does not interfere with the quality of the diffraction data is, more often than not, an art form. Special polishing techniques, such as low force lapping fixtures, electrochemical-mechanical polishing, and vibratory polishing, have been used to minimize the sub-surface damage, but have not eliminated it. Ion polishing has been used to reduce the damage layer further. However, the commercially available ion systems suffer several drawbacks, including: 1) small area treatment (≤ 1 cm) decreasing beam current density with accelerating voltages, 3) inability to process non-conducting samples. Barna and Pecz have shown that at 5° and 3 keV, approximately 25 nm of ion damage occurs in Si and GaAs, but at 250 eV, there is less than 1 nm of damage [1]. They also showed that a glancing angle across the surface is essential for removing topographic features [2].A Kaufman and Robinson, Inc. 1 cm ion source (KDC-10) was adapted to fit the etch port of a South Bay Technology, Inc. sputter coating/etching system (SBT IBS/e). This ion source overcomes the above limitations. It has a 1 cm beam size that can be collimated, or be convergent or divergent. It can be operated between 100 eV and 1200 eV with a high beam current, and a neutralizer can be used to produce a neutral beam of gas ions that can polish non-conducting samples. This study reports the use of the KDC-10 in the IBS/e to ion polish samples for EBSD analysis. Soft metal samples, typically difficult to polish mechanically, were used such as copper and solder balls. The initial EBSD mapping was done using an Oxford Instruments (HKL) Nordlys camera interfaced to a Zeiss LS15 EVO SEM and latter samples were examined with the same EBSD system on a Zeiss Ultra Plus SEM. Optimal conditions were used for each microscope. Mapping was done using a pixel step size of 0.1µm. Figure 1A,B shows an EBSD pattern of copper from an integrated circuit package substrate after a mechanical polish with 0.3 µm Al 2 O 3 suspension followed by a short 0.05 µm silica polish. A pattern from the same sample is shown in Fig. 1C,D after ion polishing with 250 eV Ar for 40 min at an angle of incidence of 5° and a total beam current of 2.5 mA. Figure 2 shows a band contrast map and a Z-oriented inverse pole figure colorization of a region of the copper within the printed circuit board. The ion polishing step improved the EBSD "hit rate" (fraction of pixels with successful indexing) from about 41% to 81% that is attributed to a significant reduction in grain smearing due to poor mechanical polishing. With some noise reduction, the software was readily able to identify boundaries, including CSL boundaries in the copper. The ion polishing is also being investigated for improved EBSD pattern identification in aluminum and Ti 6AL 4V alloys.