The role of core level shifts at metallic interfaces has often been ignored in electron energy loss spectroscopy (EELS) even though very small changes in bond length can lead to large core level shifts. However, the popular interpretation of core level shifts as measures of charge transfer is highly problematic. For instance, in binary alloys systems, the core level shifts can be the same sign for both atomic constituents[l]. The simple interpretation would require that both atomic species had lost or gained charge. Further, the signs of the core level shifts can be opposite to those expected from electronegativity arguments[2]. A core level shift (CLS) is still possible, even when no charge transfer occurs. As illustrated in Fig. 1, if the valence band width is increased, the position of the center of the valence band with respect to the Fermi energy will change (as the number of electrons remains unchanged).

Metal/ceramic interfaces play a crucial role in materials science and for various industrial purposes. In technical applications these interfaces are often exposed to high temperatures and different atmospheres. This often results in a change of the mechanical stability via the morphology and electronic structure of the interfaces. We present a comprehensive analytical electron microcopy (AEM) and fracture mechanics study of this connection on a metal/ceramic-interface model system: Cu/ α-Al2O3.

The specimens were produced by UHV diffusion bonding of bulk Cu to (α-Al2O3 single crystals. Two different Cu materials were used, either pure Cu (noted: Cu) or Cu containing 83±12 ppm oxygen (noted: Cu(O)). After bonding the interfaces were annealed in an oxygen partial pressure at 1000°C between 20 and 120 h. Four point bending tests showed an increase of the fracture energy for the Cu(O)/α -Al2O3-interfaces compared to the Cu(O)/α-Al2O3-interfaces by a factor of 5±2.

Interfacial solute-atom segregation is known to have profound effects on macroscopic physical properties (e.g. electronic transport or fracture toughness) in interface-influenced or controlled materials. It has recently been shown that solute-atom segregation also can play an influential role in the dynamics of phase transformations, in particular the oxidation kinetics of alumina-forming alloys such as PtAl. By coupling analytical electron microscopy (AEM) with conventional macroscopic oxidation studies, we have demonstrated that oxidation kinetics are affected by atomic-scale segregation of dopant species to the grain boundaries and metal-oxide boundaries in the system. Furthermore, these studies illustrate the utility of AEM techniques in elucidating the atomic-scale aspects of macroscopic physical phenomena such as phase transformations.

Oxidation studies were carried out on PtAl alloys with and without small additions of Zr to the alloy. While the pure PtAl alloys exhibited oxide scale spallation and very fast oxidation rates, the Zrcontaining alloys maintained a well-adhered oxide scale and significantly lower oxidation rates.

Radiation-induced segregation (RIS) is the spatial redistribution of elements at defect sinks such as grain boundaries and free surfaces during irradiation. This phenomenon has been studied in a wide variety of alloys and has been linked to irradiation-assisted stress corrosion cracking (IASCC) of nuclear reactor core components. However, several recent studies have shown that Cr and Mo can be enriched to significant levels at grain boundaries prior to irradiation as a result of heat treatment. Segregation of this type may delay the onset of radiation-induced Cr depletion at the grain boundary, thus reducing IASCC susceptibility. Unfortunately, existing models of segregation phenomena do not correctly describe the physical processes and therefore are grossly inaccurate in predicting pre-existing segregation and subsequent redistribution during irradiation. Disagreement between existing models and measurement has been linked to potential interactions between the major alloying elements and lighter impurity elements such as S, P, and B.

Experiment and modeling show that there is a general mode of grain boundary dissociation, common in low stacking fault energy FCC metals, that can be well understood in terms of the emission of arrays of stacking faults from the grain boundary plane. Most extensively studied of such dissociated interfaces are the Σ=3 incoherent twin boundaries. Numerous observations now exist of grain boundary dissociation at such interfaces showing a layer that is well described as a narrow, several nanometer wide slab of 9R stacked material. The 9R stacking sequence is equivalent to a close-packed stacking of FCC ﹛111﹜ planes with an intrinsic stacking fault inserted every three planes (i.e., a stacking sequence of ABC/BCA/CBA …). The width of the 9R layer is sensitive to the local stress state. Figure 1 shows results of an atomistic calculation simulating the effect of an applied shear strain parallel to the boundary.

Although grain boundaries typically account for only a small fraction of a material, they can have far reaching effects on the overall bulk scale properties. These effects are usually simply linked to the boundary having a different atomic arrangement to the bulk. A necessary first step in understanding the structure-property relationships is therefore a detailed determination of the boundary structure.

One means of obtaining detailed information on the structure of grain boundaries is through correlated Z-contrast imaging and electron energy loss spectroscopy (EELS). The Z-contrast image generates a map of the grain boundary which can be used to position the probe in defined locations for spectroscopy. In the case of oxides, a structural model of the metal atom positions can be determined directly from the image. Furthermore, using a simple bond-valence sum minimization routine, the oxygen atoms can be placed so that the structure contains atoms that have valences consistent with their expected formal valence state.

Progress towards the large scale integration of active devices within electronics packages has imposed stringent heat dissipation requirements necessitating the development of innovative materials solutions. One possibility being considered is the use of chemically vapor deposited diamond (CVDD) thin films as heat sinks. However, there are technological challenges which must be overcome before these materials become commercially viable. For example, the CVDD substrate must be metallized to provide interconnections between the various devices in the package. Conventional metallizations, such as Au, Cu or Al, display poor adhesion to the diamond causing problems associated with the reliability and stability of metallized diamond packages. A novel solution has been proposed involving the consecutive deposition of a thin layer of Cr and an electrically insulating layer of alumina on the diamond substrate such that the thermal conductivity of the treated substrate is not degraded significantly.

Ceramic matrix composites (CMC's) rely on designed weaknesses in order to provide toughening mechanisms during failure. Specifically, the fiber/matrix interface is well established as the parameter having the most influence on CMC fracture behavior, as first noted by Sambell. With proper design, this interface can allow fracture energy to dissipate by two mechanisms: an increase in crack path tortuosity and frictional energy dissipation through fiber pullout.

One design approach in the area of oxide-based CMC's which has seen considerable success recendy has been in the use of “weakly bonding” oxide coating materials placed between the fiber and the matrix phases. Such CMC systems rely on the formation of a weak interface between the fiber and the coating itself to promote the desired fracture behavior. The most notable example of such a coating material is monazite, LaPO4, as observed by Morgan et al.

This paper focuses on the differences of physical properties, specifically local dielectric function, of biogenic and geological mineral CaCO3. The goal is to assess the role of organism in forming biogenic inorganic materials. Local dielectric functions of biogenic and geological minerals were determined by transmission electron energy loss spectroscopy using a previously developed strategy.

Previous work on microstructural characterization of biological hard tissues, e.g., abalone shells, has shown variations in terms of defects, morphology, crystallography, and organization of nano- and microstructures in two biological polymorphs of calcium carbonate, calcite and aragonite, in the prismatic and nacreous regions, respectively. In the abalone, the outer (1-5 mm thick) region of the shell is composed of calcite crystallites (1-5 μm diameter) with columnar organization perpendicular to shell plane. On the inner region, 1-10 mm thick nacre is composed of aragonite crystallites (0.5 μm thick and 5 -10 μm edge-length) forming flat platelets, parallel to the shell plane.

The analysis of the atomic structure of grain boundaries is often performed through the use of high-resolution transmission electron microscopy (HRTEM). A complication of the HRTEM technique is the inability to analyze directly the experimental images in order to determine projected atomic models of lattice defects. Since contrast features in HRTEM images, in general, do not correspond directly to atomic positions, experimental images are typically evaluated qualitatively through comparison with image simulation. Recently, the interest in quantitatively measuring the atomic structure of internal interfaces for comparison with theoretical calculations has motivated the development of computational methodologies to analyze HRTEM images.123 In this paper, the quantitative analysis of HRTEM images of twin boundaries in semiconductors and metals is described.

AΣ=3 coherent twin boundary in GaP was imaged along the <110> zone in a JEOL-4000EX HRTEM at Sandia National Laboratories, Livermore.

The physical properties of crystalline materials are dependent on the presence and nature of planar defects such as grain boundaries within the crystalline lattice. A comprehensive understanding of a material's properties requires a detailed knowledge of the atomic structure and chemistry of such defects. In general, the crystal orientation relationship at grain boundaries and related planar defects are described in geometric terms, such as the coincidence-site lattice (CSL) model. However, geometric models do not take account of the atomic bonding across grain boundaries whose characteristics differ from that in the bulk. In general, defect structures may relax from the exact CSL orientation in order to minimize the energy of the interface. Lattice relaxations can manifest as a local rearrangement of the atoms close to the boundary and/or rigid-body lattice translations which perturb the relative positions of the grains far from the interface.

A1N is an attractive material because of its mechanical, thermal and electronic properties. However, these properties are often adversely altered by extended defects such as inversion domain boundaries (IDBs), and through impurities associated with the defects, especially oxygen. Thus to improve the properties of the material it is important to understand the structures of the defects. A number of studies have been carried out to characterize these defects using fringe contrast, weak-beam, convergent-beam electron diffraction and conventional high-resolution phase contrast imaging. These techniques all require the use of model structures for data interpretation. In this paper, we report a direct determination of the structures of IDBs and dislocation cores in A1N by highresolution Z-contrast imaging using a 300 kV VG HB603U scanning transmission electron microscope, with a probe size of 1.3 Å.

Although grain boundaries are known to dominate the bulk properties of many technologically important materials, in most cases there is no fundamental atomic scale understanding of why they should have such an effect. One of the problems in developing this understanding is that conventional structure determination techniques, such as phase contrast imaging in TEM or Z-contrast imaging in STEM, produce only a 2-dimensional projection of the crystal structure. Atomic scale compositional changes must be simulated and a unique interpretation is clouded by boundary reconstructions and strain effects. Furthermore, neither technique provides any information on the local changes in the electronic structure that are critical for both the electrical and mechanical properties of the boundary.

EELS provides a means to quantify local changes in both composition and electronic structure. However, without a knowledge of the structure, interpretation of any observed changes at grain boundaries is extremely difficult.

The electric transport of high-temperature superconductors, such as YBa2Cu307-x (YBCO), can be strongly restricted by the presence of high-angle grain boundaries (GB). This weak-link behavior is governed by the macroscopic GB geometry and the microscopic grain boundary structure and composition at the atomic level. Whereas grain boundaries present a considerable impediment to high current applications of high Tc materials, there is considerable commercial interest in exploiting the weak-link-nature of grain boundaries for the design of microelectronic devices, such as superconducting quantum interference devices (SQUIDs). The Josephson junctions which form the basis of this technology can also be formed by introducing artifical barriers into the superconductor. We have examined both types of Josephson junctions by EM techniques in an effort to understand the connection between microstructure/chemistry and electrical transport properties. This knowledge is a valuable resource for the design and production of improved devices.

High electron mobility structures have been built for several years now using strained silicon layers grown on SixGe(1-x) with x in the 25-40% range. In these structures, a thin layer of silicon is grown between layers of unstrained GeSi alloy. Matching of the two lattices in the plane of growth produces a bi-axial strain in the silicon, splitting the conduction band and providing light electron levels for enhanced mobility. If the silicon channel becomes too thick, strain relaxation can occur by injection of misfit dislocations at the growth interface between the silicon and GeSi alloy. The strain field of these dislocations then gives rise to a local potential variation that limits electron mobility in the strained Si channel. This study seeks to verify this mechanism by measuring the absolute conduction band shifts which track the local potential near the misfit dislocations.

Electrical phenomena such as positive temperature coefficient of resistance (PTCR), grain boundary layer capacitance, and varistors are controlled by Schottky barriers to current transport formed in grain boundary regions. These grain boundary potential barriers are a result of charged defect accumulation, either through equilibrium segregation or kinetic diffusion processes. Recent work on Nb and Mn doped BaTiO3 highlights the importance of substitutional defects which activate grain boundary trap states and give rise to the PTCR effect. The dynamic interaction of charge carriers with such trap states also produces the rapid electrical breakdown characteristic of varistors. This study explores the issues of defect species, space charge, and electrical transport in donor doped SrTi03 by electron holography and electron energy loss spectroscopy.

The specific defect species and mechanisms of current transport through donor doped SrTiO3 remain unresolved because of the difficulty in connecting defect distributions (structural information) and resultant Schottky barrier effects (chemical information).

There is much current interest in the magnetotransport properties of systems consisting of two (or more) magnetic metal layers separated by thin insulating layers. Traditional magnetic tunnel junctions (MTJs) are comprised of simple ferromagnet-insulator-ferromagnet trilayer structures and the conductance depends on the relative alignments of the magnetizations in the two ferromagnets. In the case of discontinuous metal/insulator multilayers, negative magnetoresistance (MR) results from spindependent tunneling. In both types of systems, the tunneling phenomena are strongly influenced by the microstructure of the films, particularly the metal/insulator interfaces and the nature and uniformity of the thin oxide barrier layers. In this study, we have used cross-sectional HREM to characterize a variety of magnetic tunnel junctions and discontinuous multilayers.

The MTJs were prepared by rf and dc magnetron sputtering onto thermally oxidized (100) silicon wafers at room temperature. The magnetic layers consisted of thin films of Co, Fe and/or CoFe alloys with thicknesses ∼ 30-50nm, and the barriers included MgO, HfO2, CoO, SiO2 as well as Al2O3 (thicknesses in the range 2-10nm).

Three aspects of radiation damage are of concern to electron microscopists: changes in crystallographic or molecular structure, mass loss and change in chemical composition. Structural change can be monitored from the fading of diffraction patterns or from loss of fine structure in an energy-loss spectrum. Total mass loss, in the form of a reduction in inelastic-scattering power, can be observed from the low-loss spectrum. Mass loss can also be monitored from energy-loss ionization edges, with the advantage that the loss of particular elements can be studied separately. It is possible to assign a characteristic dose De for the disappearance of a particular element.

At room temperature, the amount of damage usually depends on the accumulated dose (exposure) but not on the dose rate (current density). However, cooling the specimen tends to reduce mass loss, probably because of the reduced diffusion coefficients.

1. Introduction

The definition of resolution as introduced by Lord Rayleigh [1] is related to the width of the point spread function of the imaging device. In this definition, noise has not been taken into account. Another definition of resolution has been introduced by Rose [2] in the field of radar and TV. Here the resolution is defined in terms of the dose (D) (i.e. number of imaging particles per unit area) and the signal to noise ratio SNR (i.e. the minimal contrast) A third definition of resolution is based on the idea that the microscope is a communication channel between the object and the observer. The resolution can then be rephrased as the amount of information that is transmitted by the channel in the sense as defined by Shannon [3] as a number of bits per unit area. This definition however does not describe how this information can be deduced and what its precision is.

A good understanding of the effect of electron irradiation on polymers is necessary in order to optimally utilize the spectroscopic information and resolution of spatially-resolved electron energy-loss spectroscopy (EELS). This investigation studies the effect of electron irradiation on the low-loss spectroscopic signal and spatial resolution obtainable from polystyrene (PS) homopolymer. Because of the conjugated valence electron distribution associated with its pendant phenyl ring, polystyrene is relatively stable under electron irradiation and has well characterized spectroscopic fingerprints including a notable π—π* transition circa 7eV (1). In addition, polystyrene is used as a positive photoresist because it can cross-link effectively when exposed to an electron irradiation (2).

The critical dose characterizing degradation of aromatic polymers is of the order 1-10 C/cm2 (3). In practice, the dose delivered to a specimen is determined both by electron probe size and probe current.