In 1992, researchers at Mobil Research and Development created a new class of porous silicates, most notably MCM-41. This material features a hexagonal arrangement of pores and surface areas in excess of 1000 m2 g-1. Typically, MCM-41 exhibits pore diameters around 2.5 nm and a d-spacing around 4.0 nm, but slight variations in the synthesis of MCM-41 can produce a range of pore sizes from 2 to 10 nm. Zeolites, in comparison, have pore openings smaller than 1.3 nm and surface areas of a few hundred square meters per gram.

Recently, Aronson et al. have been studying the grafting of TiO2 to MCM-41. TiO2 is a versatile photocatalyst in many selective organic oxidations and other reactions, including the degradation of environmental toxins. by dispersing TiO2 on the internal surface of a mesoporous support one can increase the active surface area significantly and introduce some size-selectivity.

Alloying additions have been found to enhance the mechanical properties of NiAl through solid solution strengthening and the precipitation of 2°d phases. The precipitation of fine, homogeneously nucleated ß' (Ni2Al (Ti,Hf)) has been observed and characterized by dark-field imaging and electron diffraction. The exact composition of these and other precipitates, however, remains unknown. In this study, the composition of the fine ( <20 nm) ß’ precipitates in NiAl alloys is measured and compared to that of well-developed ß’ plates in the matrix.

Two directionally solidified NiAl single-crystal alloys, NiAl-3Ti-0.5Hf and NiAl-7Ti (at.%) were grown by a Bridgeman technique using high purity alumina crucibles. Both ingots were homogenized for 32 h at 1644 K. In addition, the NiAl-3Ti-0.5Hf sample was aged at 1255 K for 6 h and furnace cooled. Samples for transmission electron microscopy were prepared from 3 mm diameter cylinders electro-discharge machined from the heat-treated ingots.

Overview Scanning transmission electron microscopy (STEM) coupled with energy dispersive x-ray analysis (EDX) and electron energy-loss spectroscopy (EELS) has been used to characterize the elemental composition and oxidation conditions of various soot samples. The STEM employed in this investigation was the Vacuum Generators HB603-MIT, with a microanalyical resolution approaching 1 nm, that allowed the analysis of individual soot particles and aggregates.

The aim of this research is quantification of the EDX spectra which is possible after background and absorption corrections. This information can then be used for comparative studies of different fuels and combustion processes.

EELS has been employed to determine the amount of graphitic carbon in a soot particulate, and the detection of trace elements of low atomic number. It has been shown in soot that for Carbon the energy-loss of the p shell electrons increases with the amount of oxidation at high temperatures.

Analysis and characterization of gas turbine soot, collected from an engine exhaust duct of a 737-300 aircraft showed an abundance of different elements.

Quantitative CBED techniques are now capable of making low-order structure factor measure-ments with sufficient accuracy to study bonding effects in crystalline materials. The main limitation of these techniques has been identified as the accuracy with which one knows the Debye-Waller factor(s) (DWF(s)). Even where X-ray measurements exist, values have usually been determined at room temperature whereas we often want to perform our electron diffraction experiments at liquid nitrogen temperatures to reduce the effects of thermal diffuse scattering (TDS). Attempts to calculate theoretical DWF values have been shown to have limited accuracy when compared to experimental measurements.

This has led to a search for new electron diffraction methods for DWF determination such as the use of HOLZ line segments, high-index systematic rows and electron precession patterns. The aim should be to measure the DWF(s) under identical conditions to those used for the charge density studies, e.g. the same sample thickness, temperature and microscope settings.

The introduction of quantitative CBED techniques in recent years has led to experimental studies of charge densities in crystalline materials with unprecedented accuracy. Despite these successes, the development process is still on-going with the aim of gaining a better understanding of the techniques. With this in mind, we have undertaken a systematic study of the low-order structure factors of nickel using the ZAPMATCH zone-axis pattern matching technique of Bird and Saunders. In this approach, a set of low-order structure factors (both elastic and absorptive components) is adjusted until a best-fit is obtained between a many-beam simulation and an elastic filtered zone-axis pattern. Additional higher-order structure factors are included to converge the scattering potential but are kept fixed at neutral atom values during the fit. The questions we set out to address are (i) how to choose the correct number of structure factors to refine from a given data-set, (ii) are the absorptive structure factor components we refine of any use, and (iii) can we determine Debye-Waller factors at the same time as we measure the charge density? The first two points are discussed here.

Electron diffraction patterns taken in the transmission electron microscope (TEM) provide information about crystal structures and orientations in small sample areas. Extracting this information and manipulating local crystal orientations has become a great deal easier with the availability of desktop computer programs that allow simulation matching of experimental patterns and crystallographic control of sample tilting in the TEM. This presentation will illustrate an application of computer-aided crystallography for analyzing oriented crystallites in an experimentally complex material.

The surface corrosion films that form on reactive metals such as hafnium or zirconium in hot water provided our example. Cross-sectional examinations of the corrosion films revealed a columnar microstructure of monoclinic HfO2/ZrO2 grains extending normal to the metal/corrosion-film interface.The columnar grains were only about 50 nm in width, and thus were too small to analyze individually by selected-area diffraction. Local strains in the films smeared the diffraction fine structure so there was little hope for analysis by convergent-beam diffraction methods.

Electron diffraction patterns taken in transmission electron microscopes are widely used for phase identification and orientation determination of crystallites as small as hundreds or even tens of nanometers in size. The analyses typically require rather tedious measurement of the patterns, and matching of calculated d-spacings and angles with those of known phases. Recently, the analysis of these patterns has been facilitated by powerful desktop computer programs that use digitally captured images for on-line measurement and simulation matching of" the diffraction spot and Kikuchi line patterns. This presentation will illustrate an application of computer-aided pattern simulation and matching for precise determination of crystal orientations.

Two experimental TV camera arrangements were used to record diffraction patterns in a TEM. These included a cooled-CCD camera located on the electron optical axis below the microscope viewing chamber, and a simple TV-rate CCD that recorded directly from the inclined fluorescent viewing screen of the microscope.

Overview Radial distribution function (RDF) information obtained from amorphous glasses can provide important near and medium range structure information, determination of this information has important consequences for the encapsulation and the long term stability of nuclear waste material.

RDF data has been obtained from samples of waste glass and amorphous quartz. This data was obtained using a Vacuum Generators HB603 FEG-STEM, equipped with a GATAN Digi-PEELS. Post specimen scanning was employed to minimize the effect of spherical aberration that is present in the incident beam rocking mode.

Previous data obtained using the main-scan coils and rocking the incident beam on the specimen, diffraction information collected from out to 16 nm-1. The intensity of the information falls of rapidly as the inverse forth power of the scattering angle, out to the point where the signal to noise of the detector becomes unity. This point is limited by the sensitivity of the detector and the spherical aberration of the objective lens which limits the ultimate collection angle, regardless of the sensitivity of the detection system.

The phasing of electron diffraction data poses considerable problems for traditional direct or heavy atom crystallographic methods since the data are incomplete or at less than atomic resolution (i.e. 1.1Å), and subject to errors arising from dynamical scattering effects. The Bricogne formalism for phasing diffraction data using maximum entropy (ME) and Bayesian methods has proved especially useful in these situations1 since this formalism is stable irrespective of data resolution and completeness, and is robust with respect to errors on the measured diffraction intensities.

Successes with this formalism this include a wide range of structures:

1. (1) The ab initio phasing of diketopiperazine C4H6N2O2 (Fig 1).

2. (2) The structure solution of CuCl2.3Cu(OH)2 (Fig 2).

3. (3) The ab initio phasing of two 2-D membrane data sets at ca. 6Å resolution without the use of image phases (Halorhodopsin and Omp F Porin4 - Figs 3 and 4). In addition, we have had some success in phasing the 3-D data of bacteriorhodopsin at 6Å.

A basic problem in surface science is that, unlike bulk materials, very few of the atomic scale structures are known despite many years of effort. Even the advent of STM has not helped that much since it is difficult to impossible to interpret images without a-priori information. Transmission electron diffraction (or x-ray diffraction) data from surfaces represent a unique opportunity for the application of Direct Methods to solve surface structures, opening up new avenues for science. Diffraction from a surface when the bulk is only weakly dynamical is at least 90% kinematical, but it may be the case that strong (surface) scattering is buried under bulk reflections including the lxl or “surface forbidden bulk-allowed” reflections. In addition, further out in the diffraction pattern subsurface relaxation effects become very important, and these cannot be considered using conventional Direct Methods; instead of sharp atomic-features the true exit wave will have small positive/negative dipoles.

In principle, the availability of high-resolution micrographs in electron crystallography is a direct solution of the phase problem that has been used to great advantage for the study of proteins. However, as the resolution of the determination increases, the Fourier transform of the micrograph becomes a less accurate phase source. Hence, alternative direct methods for phase determination have been evaluated, if only to extend the resolution of most reliable lower resolution phases to the limit of the electron diffraction pattern. The first demonstration of its feasibility was published in a study of bacteriorhodopsin extending 15 Å image phases to beyond 3 Å by maximum entropy and likelihood procedures i. Later studies demonstrated that convolutional methods also can be effective.

In protein crystallography, there is always an interest in carrying out a true ab initio determinations, if only because of the challenge to traditional direct methods that become statistically less reliable as the number of atoms in the unit cell increases.

The phase problem in X-ray crystallography is one of the most interesting and most studied mathematical problems that exist. There is still today no general solution of the phase problem, but partial solutions have resulted in at least 8 Nobel Prizes, even though one of the most prominent solutions, the Patterson function, was not awarded the Prize.

It has frequently been claimed that there should be a phase problem also in electron microscopy. It may be more correct to say that there is no phase problem, only a phase confusion problem in electron microscopy. The phase confusion problem has arisen since the word phase is used for (at least) two very different physical entities in the field of electron microscopy. When crystallographers speak about phases, they mean the crystallographic structure factor phases, while physicists mean the wave front phases. In order to resolve the phase confusion problem, it is necessary to define clearly which phases are meant.

Exit waves can be reconstructed from through focus series of HREM images or by off-axis holography [1]. We have applied the through focus method to reconstruct exit waves, using algorithms developed by Van Dyck and Coene [2]. Electron microscopy was performed with a Philips CM30ST electron microscope with a field emission gun operated at 300 kV. The high resolution images were recorded using a Tietz software package and a 1024x1024 pixel Photometrix CCD camera having a dynamic range of 12 bits. The reconstructions were done using 15-20 images with focus increments of 5.2 nm. The resulting exit waves were corrected posteriorly for the three fold astigmatism.

The exit wave is complex; consequently it contains phase and amplitude. Since in the very thin regions the specimen acts as a thin phase object, such a thin area will show little contrast, an example of which is shown in Figure 1.

Aquaporin-1 (AQP1) is a class of water channels within the aquaporin superfamily. These channel proteins have been found in a wide variety of tissues such as kidney, lung, eye secretory gland and intestinal epithelium as well as in vacuolar membranes of plants. The major function of these channel proteins is to transport water exclusively into and out of cells. Based on amino acid sequence, it has been predicted that aquaporins contain six lipid bilayer-spanning a-helices. Models for the molecular folding of AQP1 containing six and four transmembrane a-helices have been proposed previously. Our earlier projection map at 3.5Å resolution revealed eight high density peaks which we interpreted as the projections of seven transmembrane α-helices and an eighth possibly transmembrane segment. The juxtaposition of structures seen in the projection map prevented us from unambiguously determining the exact number of transmembrane helices based on the projection map alone. We report here the three-dimensional (3-D) map at ˜6 Å resolution as determined by electron crystallography.

Recently it has been a matter of controversy whether inelastically scattered electrons can yield interference fringes so as to obtain holograms, and in particular whether compensation of energy loss in the object by energy gain in the source will maintain coherence [1]. In discussions about coherence (and wave mechanisms in general) it is always dangerous to rely on intuitive arguments (exchange of energy, time of interaction, etc.). In this work we will start from the most general approach, which is inspired by the treatment of inelastic electron diffraction crystals by Yoshioka in 1957 [2]. Energy exchanges are always described quantummechanically by an Hamiltonian. Therefore we can only investigate the balance between energy exchange properly if electron, object, and source are described by one global Hamiltonian. With source we mean the whole electron generating system (emitter, accelerator, condensor).

Consider a global system consisting of an electron, with position vector r, an object with particle vectors ri, and a source with particles at rα.

The acrosomal bundle is an intracellular quasi-crystalline organelle in the head of the sperm of Limulus polyphemus . It is a long (up to 60 μm) straight bundle ∼1000 Å in diameter1 consisting of two major proteins in a 1:1 stoichiometric ratio: actin (42 kDa) and scruin (102 kDa) with a minor calmodulin-like protein (14 kDa) presumably bound to scruin molecule. Previous helical reconstructions of single filaments in the bundle showed actin-scruin interactions. We recorded tilt series from single bundles in a 400 kV cryomicroscope and have developed a novel crystallographic technique to reconstruct a unit cell from the bundle as a whole to reveal interfilament interactions.

Acrosomal bundles were purified as described elsewhere and embedded in vitreous ice over holes on a holey carbon film. Tilt series images were collected in a JEOL 4000EX electron cryo-microscope at 400 kV using 10,000x microscope magnification. Total dose was ∼16-18 electrons/Å per series covering a tilt range of ± 60° with 5° angular increment.

The ability to produce features of nanometer scale offers the possibility of phase manipulation of electron waves. The first conclusive results of phase manipulation by nanometer-scale diffraction gratings directly cut by the finely focused electron beam in a scanning transmission electron microscope (STEM) have already been demonstrated. This was achieved by using a grating with a “wedge” profile. This produced an asymmetrical diffraction pattern (in violation of Friedel's law). This violation is expected only if the grating acts mostly as a strong phase object. In this paper, a demonstration of solid state Pixelated Fresnel Phase (PFP) lenses for electrons will be presented. An array of these electron lenses can be easily formed and may be utilized, for example, as compact electron-beam forming lenses for parallel electron beam lithography.

In general, an incident plane wave traveling along the optic axis of a lens experiences a phase shift. A conventional Fresnel phase lens, consisting of concentric zones with a modulo of 2π phase structure, focuses an incoming electron plane wave to its focal point.

It has long been thought that (111) surfaces of rock salt oxides microfacet to neutral surfaces upon annealing because of the very large energies involved in bulk terminating a layer of like ions. However in a recent reflection electron microscopy (REM) study Gajdardziska-Josifovska et al. found that MgO(lll) surfaces annealed in flowing oxygen furnaces at 1500°C not only did not microfacet, but displayed a √3×√3R30° surface periodicity that was stable in air. To determine the structure of this unusually stable surface MgO (111) transmission electron microscopy (TEM) samples were annealed in a vacuum furnace in the present study and their transmission electron diffraction (TED) patterns were analyzed with direct phasing methods.

The TEM samples were prepared by orienting a MgO single crystal and sawing lmm wafers along a (111) plane. Disk samples were then ultrasonically drilled, dimpled, mechanically polished and/or hot nitric acid etched, and milled with 5 KeV Ar+ ions.

Direct methods were applied to transmission electron diffraction data to solve the previously unknown In on Si(111)4x1 surface structure. The structure consists of zig-zag chains of In atoms separated by regions of silicon including dimer chains (Fig 1.). The 4x1 structure is one of several stable surface structures formed with increasing In coverages on the Si(111) surface. The √3x√3 structure consists of 1/3 of a monolayer of In, the √31x√31 occurs at a slightly higher coverage and the 4x1 structure appears before the formation of In islands on the surface . While the √3x√3 surface has been extensively studied, relatively little is known about the √31x√31 and 4x1 structures. Knowledge of the atomic positions in the 4x1 structure is an important step in understanding metal/semiconductor epitaxy and interface formation.

Two data sets were used in this study -- the first recorded on film and reduced in Tokyo, the second recorded on Imaging Plate in Tokyo and reduced at Northwestern. Twenty-seven independent intensities were measured.

The Si(111)-(3x1) reconstruction, observed for submonolayer coverage of all alkali metals and Ag, has been an attractive subject to study. Their almost identical low-energy electron diffraction (LEED) I-V curves suggest a similar surface structure, adsorbate size-independent. Moreover, filled-state scanning tunneling microscopy for Li, Na, Ag show similar features, double rows of maxima, resembling zigzag chains, and a pi surface unit cell symmetry. The atomic structure of this reconstruction is still controversial. Several models, most of them based on a π-bonded silicon chain, such as the Seiwatz or extended Pandey chain, have been proposed. This study attempts to determine this structure using direct phasing of transmission electron diffraction (TED) patterns.

Transmission electron microscopy silicon samples were cleaned and the Si(111)-(3x1)-Ag structure obtained in a surface science system, SPEAR, under ultrahigh vacuum conditions. Using the attached Hitachi UHV H-9000 microscope, through exposure series diffraction patterns were recorded and reduced with a cross-correlation technique.