High Resolution Electron Microscopy of Twist Grain Boundaries
- PDF / 2,381,892 Bytes
- 8 Pages / 420.48 x 639 pts Page_size
- 73 Downloads / 236 Views
HIGH RESOLUTION ELECTRON MICROSCOPY OF TWIST GRAIN BOUNDARIES William Krakow, IBM Corporation, T.J. Watson Research Center, P.O.Box 218, Yorktown Heights, New York 10598.
Abstract Preliminary experimental high resolution micrographs have been obtained for a Au thin film containing a [001], 2=5 (0=36.5 0) twist boundary. Other twist boundary cases such as the [001], 7.=13 (0= 22.60) and [111] low angle (30) .have also been fabricated. In order to assess whether the atomistic boundary structure is being imaged, extensive computer simulations of the E=5 bicrystal images have also been undertaken. Structurally, several contributions to the scattering must be considered which include: both upper and lower surface layers, bulk contributions above and below the interface and the relaxed interfacial structure at the composite film midplane. The effect of specimen orientation on image features is also evaluated for dynamical diffraction conditions. Likewise, the types of image features which arise by including bulk lattice reflections or alternately including only the reflections arising from the interface and surfaces will be discussed. Several schemes for analyzing twist boundary structures flat-on or near flat-on will be presented for a wide range of microscope conditions for a 400 kV microscope.
Introduction High resolution electron microscopy (HREM) has been successfully applied to elucidate the atomic structure of tilt grain boundaries in metals in several recent publications [1-3]. In addition to observing several here-to-fore new structures, matching was achieved between computer relaxed structural models and real image structures. Here columns of atoms at the interface are highly visible in projection because they run through the entire thin film thickness (-250X) along the direction of the electron beam. Viewing tilt boundaries in this manner is particularly suitable for the observation of dislocation core structures with Burgers vectors in the film plane. It should be noted that before the development of HREM, there was no suitable method for verifying the results of atomistic computer simulations since x-ray diffraction techniques were not suitable for analyzing a single boundary region edge-on. The case of imaging twist boundaries, where the electron beam is, most naturally, perpendicular to the interfacial plane, poses several problems due to the relatively small proportion of scattering from the interfacial region compared to the bulk crystalline lattice. The early experimental work on twist boundaries using electron diffraction from relatively large areas concentrated on measuring misfit dislocation array periodicities from the diffraction pattern fine structure [4]. It soon became clear however that double diffraction effects would dominate all but the thinnest bicrystalline samples [5]. Hence x-rays were used as an alternative radiation source where only single scattering (kinematical) occurs in thin films [6]. Full three dimensional x-ray structure determinations the -=5/ [001], 36.90 and 7.=13/[001], 2
Data Loading...
