The HVEM “critical voltage” effect can provide quantitative information about intrinsic crystal parameters that cannot be obtained easily on a microscopic scale by other techniques. For pure elements, more accurate atomic scattering amplitudes and Debye temperatures have been determined. In binary alloys, the occurrence of segregation, short or long range order or clustering can be evaluated and charge distribution or transfer and atomic bonding changes detected in favorable cases.

The analytical electron microscope (AEM) has been used to measure the concentration gradients which form in single and multiphase diffusion couples. The procedures used to collect x-ray microanalytical data and reduce that data to elemental compositions are typical of those used to quantify x-ray data generated in thin films. The primary difficulty in analyzing diffusion couples with the AEM is sample preparation. The principle advantage of the AEM as a tool to measure diffusion induced concentration gradients is its high spatial resolution, approximately 20 to 100 times better than that of the electron microprobe which has been traditionally used to measure these concentration profiles. As a consequence, the AEM data can yield diffusivities as much as 5 orders of magnitude smaller than those obtained from electron microprobe data. This paper will review the fundamental principles of the determination of diffusion coefficients from the concentration gradients measured in single and multiphase diffusion couples and the basic considerations of thin foil x-ray microanalysis. With this understanding of the basic concepts, recent studies of diffusion in Ta-W and U-Nb will be discussed.

The sensitization resistance of austenitic stainless steels can be improved by replacing some of the C with N. Electrochemical potentionkinetic reactivation (EPR) tests indicate that this is effective up to ∼0.16 wt.%N, but that above this level sensitization is enhanced. Thermodynamic calculations indicate that N should continue to reduce sensitization up to at least 0.25 wt.%N, as it retards the growth kinetics of Cr carbides. Analytical electron microscopy was used to investigate this apparent conflict and showed that, while N did decrease the volume diffusion coefficient of Cr beyond 0.16 wt.%, an increase in the amount of discontinuous precipitation of carbides with increasing N was responsible for the sensitization at higher N levels.

Determination of phase diagrams to understand the equilibria among phases in metallic and ceramic systems has been of paramount interest in many areas of materials research. Traditionally, phase diagrams have been determined by (i) selecting several experimental compositions to map the region in the system of interest, (ii) exposing the alloys to temperature to achieve equilibrium, and (iii) characterizing one or more of the physical properties of the bulk material such as the crystal structure(s), magnetic property, metallography, etc., to determine the number and type of phases present in each alloy. Phase boundaries are then determined by delineating the regions of alloy compositions with respect to the number and types of phases present. The traditional methods suffer from several drawbacks, the major ones being (i) the need for numerous samples, (ii) difficulties in accurately determining multi-phase regions, (iii) assumption that the bulk samples are macroscopically and microscopically homogeneous, and (iv) difficulties in establishing the attainment of equilibrium. The limitations of these traditional methods mainly stem from their inherent inability to characterize the crystal structure and composition of the individual phasesCI). More recently, methods have been developed to determine phase boundaries by x-ray microprobe analysis of bulk diffusion couples (2). Though this approach offers some advantages over the conventional methods, it is still limited by the poor spatial resolution of x-ray microprobe analysis.

In order to fully exploit the potential of Ni3Al alloys containing iron, it is necessary to characterize the phase relationships at the Ni-rich region of the Ni-Al-Fe ternary phase diagram. Due to the fine scale structures observed, a technique with high spatial resolution is required. X-ray microanalysis in the analytical electron microscope (AEM) provides a sufficiently high spatial resolution to determine the composition of the various phases occurring in this system. In this study, the isothermal section at 1000°C was evaluated in a series of alloys having nickel contents greater than 65 wt%. In Ni3Al, Fe can substitute for either Ni or Al in the ordered face-centered cubic lattice having an Ll2 structure. However, in substituting Fe for Ni one finds that the solubility limit is quickly exceeded at about 5 wt% Fe with the resultant formation of a ordered body-centered cubic (B2) phase at high temperature, which transforms martensitically upon quenching. Also, in substituting Fe for Al, the solubility limit is exceeded at about 5 wt% Fe with the formation of a disordered, face-centered cubic Ni-rich terminal solid solution. Thus, a very narrow range of alloy compositions exist where Fe can substitute for Ni and Al and preserve a single-phase L12 structure. It was found that the maximum solubility of Fe in Ni3Al at 1000°C was 17.3 wtpercnt; at 70.3 wt% Ni and 12.4 wt% Al.

X-ray microanalysis in an analytical electron microscope is a proven technique for the measurement of solute segregation in alloys. Solute segregation under equilibrium or nonequilibrium conditions can strongly influence material performance. X-ray microanalysis in an analytical electron microscope provides an alternative technique to measure grain boundary segregation, as well as segregation to other defects not accessible to Auger analysis. The utility of the technique is demonstrated by measurements of equilibrium segregation to boundaries in an antimony containing stainless steel, including the variation of segregation with boundary character and by measurements of nonequilibrium segregation to boundaries and dislocations in an ionirradiated stainless steel.

This study points out how symmetry properties inherent in precipitation reactions can be exploitet to facilitate and improve the accuracy of a morphological analysis by TEM. A method is given for the simple determination of the direction and distribution of needle precipitates and the habit plane of plate precipitates based on the use of symmetry properties of the matrix crystal. The bicrystal symmetry of different orientation relationships between cubic crystals and its connection with observed and equilibrium shapes is illustrated with examples of high resolution micrographs of Ge precipitates in an Al matrix.

Small precipitates of high lattice mismatch with respect to the matrix may be investigated by high resolution and high contrast moiré fringe images in TEM. One group of such precipitates are the MX-type precipitates with M = Ti, V, Zr, Nb, Hf, Ta, and X = C,N that form in austenitic steels. The usefulness of moiré images for the investigation of these MX-type precipitates is demonstrated and compared to other electron imaging and diffraction methods. The moiré technique has been used to study (a) nucleation and growth of precipitates, (b) the evolution of dual precipitate structures at grain boundaries, (c) morphology and faceting of small precipitates (∼10 nm diameter), (d) lattice parameter differences due to variation in precipitate chemical composition, (e) interaction of precipitates with dislocations, (f) gas atom trapping at precipitates, and (g) irradiation induced precipitate nucleation at point defect clusters.

Recent advances in transmission electron microscopy instrumentation and technique now make it possible to study the shape-evolution of precipitates in metallic alloys at the atomic level. This investigation demostrates how a combination of transmission electron microscopy techniques; namely, high-resolution electron microscopy, image simulation, energy-dispersive x-ray spectroscopy and convergent-beam electron diffraction are used to characterize the atomic structures, chemistry and growth mechanisms of γ' precipitate plates in an Al-4.2 a/o Ag alloy aged for 30 min. at 350°C. The complimentary information obtained from each of these techniques allows modelling of the growth process at the atomic level, thus providing insight into the basic precipitation behavior of alloys.

A number of dispersed phases in Al-Fe-Ce alloys, varying structurally and chemically, have been observed by TEM. Of particular interest were rod-like particles which showed five-fold symmetry and irrational spacings. These dispersoids contain linear features, a periodic, columnar structure parallel to the five-fold axis, aperiodic spacings perpendicular to the five-fold axis and were coherent in one direction with the Al matrix. The evidence indicates a two-dimensional Penrose tiling with a columnar structure, possibly a stacking of eclipsed pentagonal units.

Matrix and grain boundary precipitation in austenite has been studied in two Nb/V microalloyed steels with different V levels, which were given a number of different heat treatments after hot deformation. The steel with a higher V level was found to have a greater number of fine carbonitride precipitates in the matrix. In both steels, the precipitates became more Nb-rich as the aging temperature was increased from 854°C to 1066°C. For a given steel and heat treatment, the matrix precipitates were generally more Nb-rich than the grain boundary precipitates.

Vasco X-2M is a low carbon tool steel which is being considered for critical gear applications. Since toughness is an important property for these applications, the Charpy V-notch impact energy was tested as a function of tempering temperature and found to drop considerably at 480C temper and above. A TEM investigation was carried out in order to better understand the results of this test. The precipitation sequence of carbides was examined along with their morphology, structure and orientation relation to the matrix. The precipitation of the intralath M3C carbides occurred on quenching from austenite. These carbides were then present in all tempered samples. M2C carbides were observed at 480C and 590C temper, and evidence for precipitation of these carbides was based on hardness data at 480C temper and TEM observations at 590C temper. After aging 200h at 590C, the ceitientite was observed to be in the process of transforming in an in-situ manner to an M7C3 carbide. The minimum in the Charpy impact energy was found to occur when both coarse M3C and a very fine dispersion of M2C were present in the microstructure.

The sizes and compositions of precipitates in two weldable HSLA steels have been determined using analytical electron microscopy. It has been shown that the balance between the microalloy additions must be carefully controlled if the formation of very large precipitates in the cast slab is to be avoided. These large precipitates are very stable, some having been shown to be stable on reheating to about 1200°C and others are stable up to the melting point of the steel. The significance of these observations is briefly discussed in terms of the influence of these precipitates on the properties of the microalloyed steels.

Ti-Ni alloys have been made amorphous over a broad composition range by sputter deposition, ion implantation and electron irradiation. Structural analysis of these alloys was made by electron diffraction techniques. Microdensitometer traces of diffraction patterns produced scattering profiles from which radial distribution functions (RDF's) were derived. The results from this analysis were comparable to those from x-ray diffraction studies on similar alloys. It was found that the positions of the intensity maxima vary systematically with alloy composition. However, values of coordination number were less precise due to experimental uncertainties. Furthermore, no dramatic differences were observed in the RDF's of the samples amorphized by the three techniques.

The migration kinetics of antiphase boundaries (APBs) in long-range ordered Fe-Al alloys are analyzed using in-situ hot-stage transmission electron microscopy, making it possible to observe the interfacial migration directly as it occurs. The observed domain structures are essentially two dimensional due to the thin foil configuration of the specimen, facilitating analytical interpretation of the data. By measuring the local curvatures and migration distances from a series of micrographs taken at a constant temperature, the proportionality constant relating velocity to curvature can be determined. This proportionality constant, called the mobility for APB migration, is found to be independent of the curvature for <100> APBs in Fe-27 atomic percent Al at 745K, having the value of 4.8×10−19 m2/s.

A technique has been developed which employs high resolution transmission electron microscopy (HRTEM) for the observation of the atomic mechanisms associated with solid state phase transformation as they occur at elevated temperatures. It consists of the annealing in-situ of cross-section transmission electron microscopy (TEM) specimens that have been favorably oriented for lattice fringe imaging and the video-recording of dynamic events as they occur in real-time. By means of this technique, we report the first video-recorded lattice images of crystallographic defect motion in silicon, viz. the motion of dislocations and stacking faults, as well as the first such images of the atomic mechanisms responsible for the amorphous to crystalline (a-c) phase transformation, viz. heterogeneous nucleation of crystal nuclei, coalescence of crystal nuclei by co-operative atomic processes, ledge motion at the growth interface, and normal growth in silicon. This technique holds great potential for the elucidation of the atomic mechanisms involved in reaction kinetics in the solid state.

Electron irradiation induces many intermetallic compounds to become amorphous. As this irradiation is capable of producing only isolatedpoint defects for each collision event, an understanding of this amorphisation proqess has a special significance with respect to the understanding of amorphous materials. This paper will report the results of HREM observations of an intermetallic compound in the Cu-Ti alloy system. High resolution micrographs will be presented to demonstrate that the energy associated with the irradiation induced chemical disordering is the major driving force in the amorphisation process. Also, the utility of the HREM in the study of amorphisation and the amorphous structure will be discussed.

Solute partitioning during rapid solidification is being studied in cellular growth structures produced via pulsed-laser melting of binary alloys. This rapid solidification process can generate significant departures from local equilibrium at the liquid-solid interface which directly affects the resultant microsegregation patterns. Calculated compositional profiles are obtained from a solidification model incorporating the effects of both solute trapping and solute buildup with a suitable tip stability criterion. These are compared to experimentally determined campositional profiles generated across the steady-state cellular regrowth region of the melt pool using STEM x-ray microanalysis.

Transmission electron microscopy has been used to study phase transformations in a Ni4Mo alloy containing 2.08 wt pct chromium following isothermal annealing at 850, 800 and 700°C. At these temperatgres, the DO22 phase forms initially, which then transforms to Ni3Mo. At 700°C, the DO22 phase forms from bands composed of Ni4Mo and Ni2Mo aligned parallel to {111} fcc planes, and is relatively stable. A second mechanism for the formation of Ni3Mo occurs at 700°C. Large Ni4Mo domains form at grain boundaries. These su~sequently transform to Ni3Mo, with the simultaneous presence of stacking faults on (010) Ni3Mo planes.

These observations are correlated to the optical microstructure. Grain size influences the type, volume, and distribution of second phases. Sequences and mechanisms of transformations are related to the experimental observations.

Commensuration and discommensuration characteristics of the long period superlattice in the Ag-Mg alloys near Ag3Mg with short modulation periods (M ≃ 2) are investigated. When quenched from 650°C, the modulation periods M of these alloys is found to take theoretical values determined by the size of the Fermi surface, while upon further annealing at low temperatures, alloys assume the commensurate structure M=2, except for those alloys with theoretical M values substantially different from 2. These alloys approach the periodic discommensuration structures of the type <2j1> or its variants whose average periods M are closest to the theoretical incommensurate values. This behavior leads to the conclusion that the formation of the long period superlattice in these alloys is basically due to the CDW formation and that the commensuration and the discommensuration characteristics are determined by the free energy of the modulation boundaries.