Automated Crystal Orientation Measurement/Mapping (ACOM) as used in the SEM by unattended interpretation of Backscatter Kikuchi Patterns from bulk surfaces (BKP)* has become an invaluable new technique for the quantitative characterization of microstructure on a grain-specific level. The principal objectives are to relate grain orientations and misorientations directly to grain location in the specimen surface, to determine the character of misorientations and grain/phase boundary planes, and to calculate derived entities such as microtexture and anisotropic local materials properties. These quantities are commonly depicted in pseudo-colors on the scanning grid to form Crystal Orientation Maps (COM) or similar “images” of the microstructure. Stereological as well as orientation related data, as most sensitive indicators of the production process and use of a material, are readily available for basic research and quality control. The three-dimensional representation of microstructure by consecutive sectioning is still a challenge to enable the study of inhomogeneity beneath the surface and the three-dimensional connectivity of multiphase material. Dynamic experiments are a further promising application of this new technique in the SEM.

Electron Back-Scatter Diffraction (EBSD) is a technique used for obtaining microtextural information from bulk samples in the scanning electron microscope (SEM). number of good commercial systems provide hardware and software to acquire and analyze the electron back scatter patterns (EBSP). The systems come with tools to visualize crystal orientation, and map grain and grain boundaries. However money is always in short supply, and for studies which do not require mapping, we have found it more economical to build our own EBSP system using readily available off-the-shelf components.

An EBSP's are formed by the intersection of the diffracted back scattered electrons with a phosphor screen placed in their path. We acquire this pattern by viewing the phosphor.screen through a view port with a CCD video camera which is configured for on-chip integration. On-chip integration allows the signal to be amplified without extensive electronic noise. We made a port cover for the right-hand port of the DSM-962 SEM on which a 2” diameter view window is fitted.

This paper describes a new form of orientation imaging microscopy: Adaptive Orientation Imaging Microscopy or AOIM. In standard OIM experiments the local orientation of the crystal lattice is determined with respect to a suitable reference frame by acquiring and indexing electron back-scattering (EBS) patterns on a square or hexagonal grid of sampling points. Since its invention, OIM has been extensively used to characterize polycrystalline materials. With the fast development of computers and CCD camera systems in the past few years, the standard OIM can now acquire and index of the order of 4000 EBS patterns per hour.

In a typical OIM experiment one is usually interested in either the overall texture of. the sample, or in the types and relative frequencies of grain boundaries. In the former case the proper way of acquiring orientation data involves the use of an equidistant grid of sampling points. In the latter case, however, a large fraction of the acquisition time is wasted sampling the grain interior, which does not provide any information about the grain boundaries.

This work discusses using the automated Electron BackScatter Pattern (EBSP) technique of electron diffraction in the Scanning Electron Microscope (SEM) for measuring grain size in polycrystalline materials. A quantitative analysis of 2-D grain size from a standard metallographic section can be made in less than one hour using an automated-EBSP system on a traditional SEM and dedicated analysis software. Thus, the automated EBSP technique provides a useful and cost-effective way to determine grain size. A common measure used to describe grain size is mean intercept length (MIL) and the present discussion will be limited to determining MIL using EBSP.

Grain size analysis using optical microscopy is strongly influenced by sample preparation, which involves etching of the polished section to reveal boundaries. Boundaries that are to be avoided in determining MIL, such as sub-grain boundaries and twin boundaries, may be revealed and mistaken for random high angle boundaries.

Recent advances in cameras and computers have made it possible to build electron backscattering pattern (EBSP) cameras which can give crystallographic information from diffraction patterns in the scanning electron microscope (SEM) on a routine basis.[1] There are a few hundred such systems world wide and the number is growing fast. In the case of crystalline samples (nearly all applications of SEM outside the biomedical field), it will surely soon be considered essential to fit an SEM with an EBSP system, just as it is now considered essential to have the SEM equipped with an energy-dispersive x-ray spectroscopy (EDS) system. There are at least four commercial manufacturers of EBSP systems. In the next few years, then, I anticipate that many owners of existing SEMs as well as buyers of new instruments will be faced with the problem of selecting an EBSP system. This paper presents some of the issues involved in making such a choice.

As in many technical decisions, different people will have different needs and put different priorities on the specifications to be met. An instrument which is to be used to do repeated analyses of aluminum for beer cans will have different needs from a system used mostly to teach crystallography, and those will be different in turn from the needs of a system used to determine which phases are present in geological samples.

This work describes an investigation of the microstructure of a γ/γ' Ni-base superalloy using the automated Electron BackScattering Pattern (EBSP) technique with simultaneous energy dispersive x-ray spectrometry (EDS). The goal of the investigation was to determine if EDS measurements necessary to discriminate γ' from γ' could be obtained when acquired simultaneously with a normal, high speed automated EBSP scan. The acquisition conditions that impact the EDS collection are: surface tilted ∼70 degrees from the beam axis, 70mm spectrometer to specimen distance limited by interference from the EBSP detector and approximately 300 ms available acquisition time.

Thermo-mechanical processing of superalloys can produce coarse γ' that appears both within grains and at grain boundaries. In some conditions, the coarse γ' may be as large as many γ grains. As the crystal structure of γ and γ' are very similar, it is not possible to discriminate between the two phases using typical EBSP analysis. However, one needs to discriminate these phases in order to measure properties of the γ matrix such as texture and grain size.

Metals on oxide systems are important in applications such as catalysts and gas sensors. The Au/ TiO2 system is of particular interest because of its high activity for low temperature oxidation of CO and its good sensitivity as CO gas sensor. In this study, we are reporting results on the effect of substrate temperature on epitaxial orientation relationship of Au particles formed on TiO2 (110).

Two Au films were grown on TiO2 (110) substrates by vapor deposition in a UHV chamber. The first one was produced by evaporation of a 12nm thick Au film at 300K, followed by annealing at 775K for 1 hour. The second one was produced by direct deposition of a 12nm film on TiO2 substrate maintained at 775K. The samples were transferred in air to the Field Emission Scanning Electron microscope (LEO 982 Gemini) for imaging and diffraction. .Electron Backscatter Diffraction (EBSD) patterns were taken from individual Au particles using the Opal system (Oxford Instruments). With this system, a spatial resolution of the order of 80 nm has been achieve for Au.

A typical HRSEM image of the Au film deposited a 300K and annealed at 775K is shown in Fig. 1 revealing a discontinuous Au film consisting of discrete Au particles in the range from 20 to 150 nm.

Crystallographic structure plays an important role in determining the fundamental physical properties of metallic thin films and superlattices, and structural characterization of such films is crucial for furthering the application of these materials [1]. Epitaxial films are usually characterized using a wide range of techniques including x-ray diffraction, REED, neutron diffraction, and highresolution electron microscopy. A complementary approach is electron backscatter patterns (EBSP), which is an SEM technique that can be used to measure crystallographic orientations of single or polycrystals [2].

In this work, EBSPs have been used to characterize the crystallite size and orientation of sputterdeposited Cu films. The EBSPs were formed in a CamScan 44FE SEM and recorded using an ORTEX CCD camera system. The images were analyzed using the Channel+ software package, and orientation maps were plotted using Channel+Ice [3].

Directionally solidified (DS) in-situ composites based on (Nb) and (Nb) silicides, such as Nb5Si3 and Nb3Si, are presently under investigation as high-temperature structural materials [1, 2]. Alloying additions of elements such as Hf, Ti and Mo to these silicides are also being explored. The present paper describes the microstructure of a DS Nb-silicide based composite before and after creep deformation.

Alloys were prepared from high purity elements (>99.9%) using induction levitation melting in a segmented water-cooled copper crucible. The alloys were directionally solidified using the Czochralski method [2]. Creep tests were conducted at 1200°C to 50% deformation. Characterization was performed using scanning electron microscopy, electron microprobe analysis (EMPA), and electron backscatter diffraction pattern analysis (EBSP).

Using the new method known as Orientation Imaging Microscopy (OIM), characterization of mesoscale aspect of the internal structure of polycrystalline materials in the two-dimensional section plane has become routine. However, in some applications which primarily focus either on the characterization of interface types or on their connectivity, OIM is rather inefficient since only the scan points that lie adjacent to the interfaces are used to determine interface character. When the OIM grid spacing is dictated by the precision with which the interfacial in-plane inclination must be determined in the section plane, the efficiency with which conventional OIM can harvest interfacial data is rather poor.

In this paper a new approach and system of microscopy is described, called the Mesoscale Interface Mapping System (MIMS). MIMS overcomes the challenges associated with OIM by directing the beam to the vicinity of interfaces which have been identified by a microstructural contrast image.

Electron diffraction was observed in 1928 by Kikuchi. In electron diffraction, diagrams characteristic of the crystal lattice and the phase orientation are formed. Consequently, Venable, then Dingley recently developed electronic diffraction coupled with scanning electron microscopy. Their approach enables micrometric scale examination of a specimen obtained by simple metallographical preparation (polished surface). Since the 1980’s, this technique has benefited from strong technological and computerized development which have led to the sale of the first systems under the name of Electron Backscatter Diffraction (EBSD). Niobium is used in the cast industry of uranium to refine it (getter function). The strong affinity of niobium for carbon leads indeed to the formation of small precipitates of niobium carbides. Whereas carbon cannot be detected by EDS because of its high absorption by the niobium (absorption coefficient μ=29000), hardness measurement indicates the presence of carbides (HV30 = 2000), but without further information on the nature of those carbides.

We report here our preliminary results on orientation determination and phase identification using the electron backscatter diffraction (EBSD) technique. Using a scanning electron microscope (SEM) equipped with the EBSD attachment. it is now possible to study the correspondence and orientation relationships of parent-phase and martensite variants in shape memory alloys (SMAs). Previously, such an information was obtained from large single crystals studied by micro-beam X-ray Laue diffraction and supplemented by transmission electron microscopy study of thin foils. Figs. 1(a), (b) and (c) are three EBSD patterns taken from neighboring areas in a Cu-12.55Al-4.86Ni (wt%) SMA. Computer simulation reveals that Fig. 1(a) belongs to the parent-phase of D03-structure type, and Figs. 1(b) and (c) belong to variants A and D of 2H martensite, respectively. Corresponding simulated EBSD patterns are shown in Figs. 1(d), (e) and (f). Fig. 1 indicates that the (0 0 2)A basal plane of the martensite variant A is transformed from the (-2-2 0)p plane of the parent-phase and its [0-1 0]A direction from the [0 0 1]P direction. The (0 0 2)D basal plane of the martensite variant D is transformed from the (2-2 0)P plane of the parent-phase and its [0 1 0 ]D direction from the [0 0 1]P direction.

A relationship between the local microstructure (crystallographic orientation) of a sample and an image from a scanning electron microscope (SEM) has been an elusive goal. Being able to show the structure from an exact area and relate it to other data is essential to the understanding of the structure relationship to the entire sample. Conventionally microtextural information of tapes is obtained by using x-ray diffraction and pole figures. But this only provides information about the global texture. The precise location is not well defined. The ability to physically transfer the sample from the x-ray diffractometer to a SEM and accurately state that this is the same area has been impossible. Electron Backscatter Diffraction Patterns (EBSPs) have been obtained in an electron microscope and the exact area in relation to the rest of the sample is determined. Thus larger areas can be defined and put together to form a composite of an entire sample. This, however, may be hard to accomplish in that defined areas for the image scan may be at a different scale than the EBSP scan. A contrast method of imaging in the ESEM using an increased vacuum and the ESD yields a good correlation to mapped areas obtained from EBSPs or also known as backscatter electron Kikuchi diffraction patterns (BEKPs or BKPs).

The Environmental Scanning Electron Microscope, or ESEM, is the only class of SEM that can image in a gaseous environment that will maintain a sample in a fully wet state. The use of the patented Gaseous Secondary Electron Detector, or GSED, which amplifies the secondary electron signal with the gas, has allowed the ESEM to image a multitude of samples with true secondary contrast. Recently, several new modes of imaging in a gas have been developed and will allow further expansion of the capabilities of the ESEM.

To maintain pressures in the ESEM up to 20 Torr (27 mbar), the use of multiple, differentially pumped apertures, is required. This can place a restriction on the low magnification range. In the large field detection mode, all magnification restrictions are removed. Magnifications as low as lOx may be achieved. This is similar to many conventional SEMs.

Charge Contrast Imaging (CCI), is a recently reported technique allowing the imaging of the internal microstructure of non and poorly conductive materials. In previous work, the relationship between controlled growth events in batch-precipitated gibbsite, Al(OH)3 and the appearance of contrasting bands in CCIs has been empirically correlated. The data presented was spatial, based on the number and dimensions of growth rings, however no interpretation of the contrast variation was made. Suggestions that the contrast may be related to charge trapping, which in turn is related to conductivity pathways and impurity concentrations indicates that the technique may provide insight into crystallisation processes. The current work examines the spatial distribution of impurity elements and their effect on the contrast observed within growth layers in charge contrast images of gibbsite.

Analyses for Ca, Fe and Si, the main impurities in the solutions before and after precipitation indicate a preferential partitioning of Ca and to a lesser extent Si and Fe into gibbsite.

Understanding the various interactions between electrons and the specimen chamber gas is essential to using environmental and low-vacuum SEM’s to their full potential. The problem can be divided into two main topics: distortion of the primary beam on its way to the sample surface, and amplification of the various electron signals leaving the sample. These issues are important for all low-vacuum SEM’s, regardless of the system design or signal detection method. All instruments must contend with scattering of the primary beam in the gas, and the subsequent effects on resolution, background signals,.and x-ray generation. Similarly, while some instrument designs rely more heavily than others on signal amplification in the gas, all benefit from the concomitant charge neutralisation.

As the electron probe travels through the gas on its way to the specimen, elastic collisions occur with molecules in the gas.[1] Fortunately, for gas pressures of a few torr, the mean free path is on the order of tens of millimetres, and the vast majority of electrons reach the specimen unscattered.

Uncoated, non-conductive samples can be imaged and analyzed in the environmental scanning electron microscope (ESEM) due to effective charge neutralization at the sample surface by ionized gas molecules. Under some gas pressure and electron dose conditions, ESEM images of uncoated, poorly conductive samples often contain contrast not present in secondary or backscattered electron images of the (coated) samples obtained in conventional SEMs. It has been proposed that the contrast is related to charge trapping at defects and impurities. It has also been suggested that UV cathodoluminescence (CL) may contribute to contrast in the ESEM. In this paper, we present experimental evidence of contrast formation in the ESEM due to charge trapping in Dy doped zircon, electron trapping at oxygen vacancies in sapphire and the absence of signal generation by 360nm UV CL.

The specimens used in this study were (i) cross-sectioned Titanium in-diffusion doped sapphire single crystal, (ii) Dy doped synthetic Zircon7 and (iii) 43 μm epitaxial GaN grown on c-pane sapphire by hydride vapor phase epitaxy.

When a poorly conducting specimen is irradiated with an electron beam in a variable pressure electron microscope, the excess charge on the surface of the specimen can be neutralized by incident gas ions to prevent deflection and retarding of the electron beam. A small fraction (<10∼6) of the incident electrons are trapped at irradiation induced or pre-existing defects within the irradiated micro-volume of specimen. The trapped charge induces an electric field, which may result in the electro-migration and micro-segregation of charged mobile defect species within the irradiated volume of specimen. These charge induced effects are dependent on the density of trapping centers and their capture cross sections. In particular, evidence of these micro-diffusion processes can be directly observed in electron beam irradiated ultra pure silicon dioxide (SiO2) polymorphs using Cathodoluminescence (CL) microanalysis (spectroscopy and imaging). CL microanalysis enables both pre-existing and irradiation induced defects in wide band gap materials (i.e. semiconductors and insulators) to be monitored and characterized with high sensitivity and spatial resolution. Depth resolution is achieved by varying the electron beam energy.

Charge Contrast Imaging (CCI) in the environmental scanning electron microscope (ESEM) has rapidly evolved to an important imaging technique for a range of materials. The investigation of the applicability of CCI to the examination of semiconductor and thin film (MCT) samples and devices has been highly successful. Dopant concentration and depth can be differentiated, as has been noted for variable pressure SEM. Quantification of concentration remains difficult due to the complexity of variables controlling CCI but should be achievable in the near future. The extreme sensitivity to surface contaminants previously noted on polished diamond surfaces is also seen on silicon. Monte Carlo modelling of this contrast indicates, for the examined materials, an information depth of a few nanometres for CCI, whilst using 30kV accelerating voltage. The modelling shows that CCI is capable of routine detection of contaminant layers on silicon at the nanometre scale. These results also provide the first data confirming initial suggestions that CCI is derived from the near-surface regions of the sample.

CCI examination of mercury cadmium telluride (MCT) devices has also provided images with new detail. This detail is currently being interpreted. Initial examination suggests an influence of the substrate on film growth (figure 1). Unusual corner structures are visible around the mask region.

We report the effects of scan rate/electron dose on the secondary electron contrast in liquid systems, studied by Environmental Scanning Electron Microscopy (ESEM). Understanding the mechanisms for SE contrast is essential: previous observations of oil-water emulsions have revealed useful SE contrast between the dispersed and continuous phases, attributed to differences in the electronic structure of these molecular liquids. However, it has been found that large oil droplets for example, have a propensity to give SE contrast that can be inverted (relative to the continuous water phase) as a function of scan rate/electron dose, as shown in Figures 1 & 2. In the ‘grabbed’ image, Figure 1, the scan rate is a relatively rapid 2.1 frames/sec. and results in the expected image of oil droplets with lower SE signal intensity than the surrounding water phase. However, the intensity of the oil droplets changes dramatically when the scan rate is slowed, such that the droplets appear bright. Figure 2 shows the same region of sample as in Figure 1, but with image acquisition over 30 seconds. The same contrast inversion can occur at rapid scan rate if higher magnification is used. Effects such as this could result in ambiguous interpretation of ESEM images. However, useful information may be gathered as to the dielectric nature of the medium — a property related to molecular structure. We note that similar scan rate-dependent effects have been observed in ESEM studies of solid samples, although the mechanisms responsible have yet to be rigorously established.