Formation process of silicide at Pt-Si (111) interface

Author(s):  
Y. yokota ◽  
H. Hashimoto

The initial stage of the reaction process forming the platinum silicide at Pt/Si (111) interfaces have been investigated by a high resolution electron microscope in both “flat-on” [1] and “cross-sectional” mode.First, platinum of 11 nm thickness was deposited on a Si (111) wafer. For observing the specimen in “flat-on” mode, the top surface and edge of the specimen were covered by paraffin and etched chemically using CP-4 reagent from backside. The specimen for observing in “cross-sectional” mode was prepared as follows. The Si wafer was cut into slips of 3mm width. After stacking 4 to 6 slips together with epoxy resin, the stacks were sliced to a thickness of 0.3-0.4mm by a diamond saw. The slices were mechanically polished to a thickness lower than 0.05mm and then thinned by Ar ion beam.The formation process of platinum silicide (PtSi) was observed in a cross-sectional specimen.

Author(s):  
D. X. Li ◽  
P. Pirouz ◽  
A. H. Heuer ◽  
S. Yadavalli ◽  
C. P. Flynn

MgO films were deposited on the (sample A), (0001)Al2O3 (sample B), and the (sample C) planes of sapphire by Molecular Beam Epitaxy (MBE). Cross-sectional UREM specimens were prepared using standard techniques and examined in a top-entry JEOL 4000FX high resolution electron microscope. Image simulations were performed using the SHRLI programs developed by O'Keefe.


Author(s):  
D. X. Li ◽  
P. Pirouz ◽  
A. H. Heuer ◽  
S. Yadavalli ◽  
C. P. Flynn

Nb/MgO/Nb/Al2O3 superlattices were prepared by MBE growth of Nb and MgO films on a (012)Al2O3 sapphire substrate. Cross sectional HREM specimens were prepared using standard techniques involving mechanical grinding to a thickness of 0.13 mm, dimpling to a thickness of 20 μm and ion beam milling at 6 kV. The samples were subsequently examined in a top-entry JEOL 4000EX high resolution electron microscope with a point-to-point resolution of ∼0.19 nm. Image simulations were performed using the SHRLI programs developed by O'Keefe.


1992 ◽  
Vol 283 ◽  
Author(s):  
Toyohiro Chikyow ◽  
Nobuyuki Koguchi

ABSTRACTA selective growth of GaAs microcrystals was demonstrated on a Se-terminated GaAlAs surface. Ga molecules were supplied to the Se-terminated GaAlAs surface at first. The surface consisted of Ga droplets and bared Se-terminated GaAlAs surface. After the following As molecule supply to the surface, a selective GaAs microcrystal growth from Ga droplets was observed. The cross sectional investigations by the high resolution electron microscope revealed epitaxial growth of GaAs microcrystals with (111) facets and a possibility of (GaAl)2Se3, layer formation at the GaAs/Se-terminated GaAlAs interface.


1993 ◽  
Vol 300 ◽  
Author(s):  
Toyohiro Chikyow ◽  
Michihisa Lijima ◽  
Nobuyuki Koguchi

ABSTRACTA selective growth of GaAs micro crystals was demonstrated on a Se-terminated GaAIAs surface by sequential supplies of Ga and As molecules for the quantum well box structure. After the growth, the surface consisted GaAs micro crystals with (111) facets and some Se clusters. The cross sectional investigations by the high resolution electron microscope revealed an epitaxial growth of GaAs micro crystals on the surface and a mixture of Ga2Se3 and A12Se3 layer formation at the interface of GaAs/Se-terminated GaAIAs. The selenidation process seems to be a reaction limited one. The Se cluster segregation could be avoided by selenidation in As molecule atmosphere.


Author(s):  
O.C. de Hodgins ◽  
K. R. Lawless ◽  
R. Anderson

Commercial polyimide films have shown to be homogeneous on a scale of 5 to 200 nm. The observation of Skybond (SKB) 705 and PI5878 was carried out by using a Philips 400, 120 KeV STEM. The objective was to elucidate the structural features of the polymeric samples. The specimens were spun and cured at stepped temperatures in an inert atmosphere and cooled slowly for eight hours. TEM micrographs showed heterogeneities (or nodular structures) generally on a scale of 100 nm for PI5878 and approximately 40 nm for SKB 705, present in large volume fractions of both specimens. See Figures 1 and 2. It is possible that the nodulus observed may be associated with surface effects and the structure of the polymers be regarded as random amorphous arrays. Diffraction patterns of the matrix and the nodular areas showed different amorphous ring patterns in both materials. The specimens were viewed in both bright and dark fields using a high resolution electron microscope which provided magnifications of 100,000X or more on the photographic plates if desired.


Author(s):  
Mihir Parikh

It is well known that the resolution of bio-molecules in a high resolution electron microscope depends not just on the physical resolving power of the instrument, but also on the stability of these molecules under the electron beam. Experimentally, the damage to the bio-molecules is commo ly monitored by the decrease in the intensity of the diffraction pattern, or more quantitatively by the decrease in the peaks of an energy loss spectrum. In the latter case the exposure, EC, to decrease the peak intensity from IO to I’O can be related to the molecular dissociation cross-section, σD, by EC = ℓn(IO /I’O) /ℓD. Qu ntitative data on damage cross-sections are just being reported, However, the microscopist needs to know the explicit dependence of damage on: (1) the molecular properties, (2) the density and characteristics of the molecular film and that of the support film, if any, (3) the temperature of the molecular film and (4) certain characteristics of the electron microscope used


Author(s):  
T. A. Welton

An ultimate design goal for an improved electron microscope, aimed at biological applications, is the determination of the structure of complex bio-molecules. As a prototype of this class of problems, we propose to examine the possibility of reading DNA sequence by an imaginable instrument design. This problem ideally combines absolute importance and relative simplicity, in as much as the problem of enzyme structure seems to be a much more difficult one.The proposed technique involves the deposition on a thin graphite lamina of intact double helical DNA rods. If the structure can be maintained under vacuum conditions, we can then make use of the high degree of order to greatly reduce the work involved in discriminating between the four possible purine-pyrimidine arrangements in each base plane. The phosphorus atoms of the back bone form in projection (the helical axis being necessarily parallel to the substrate surface) two intertwined sinusoids. If these phosphorus atoms have been located up to a certain point on the molecule, we have available excellent information on the orientation of the base plane at that point, and can then locate in projection the key atoms for discrimination of the four alternatives.


Author(s):  
David J. Smith

The electron microscope has evolved to the level where it is now straightforward to record highresolution images from thin samples (t∼10 to 20nm) that are directly interpretable in terms of atomic arrangements. Whilst recorded images necessarily represent two-dimensional projections of the structure, many defects such as dislocations and interfaces may be linear or planar in nature and thus might be expected to be amenable to detailed characterization. In this review, we briefly consider the recent significant progress that has been made in quantitative defect analysis using the high-resolution electron microscope and then discuss some drawbacks to the technique as well as potential scope for further improvements. Surveys of defect modelling for some small-unit-cell materials and interfaces have recently been published, and reference should be made to other papers in this symposium for further examples.The technique of structure imaging originated in the early '70s with observations of large-unit-cell block oxides.


Author(s):  
G.Y. Fan ◽  
O.L. Krivanek

Full alignment of a high resolution electron microscope (HREM) requires five parameters to be optimized: the illumination angle (beam tilt) x and y, defocus, and astigmatism magnitude and orientation. Because neither voltage nor current centering lead to the correct illumination angle, all the adjustments must be done on the basis of observing contrast changes in a recorded image. The full alignment can be carried out by a computer which is connected to a suitable image pick-up device and is able to control the microscope, sometimes with greater precision and speed than even a skilled operator can achieve. Two approaches to computer-controlled (automatic) alignment have been investigated. The first is based on measuring the dependence of the overall contrast in the image of a thin amorphous specimen on the relevant parameters, the other on measuring the image shift. Here we report on our progress in developing a new method, which makes use of the full information contained in a computed diffractogram.


Sign in / Sign up

Export Citation Format

Share Document