The nano-scale dimensions of next generation VLSI and ULSI devices will drive the development of a variety of new processing requirements. Currently device processing conditions from substrate cleaning to thin film deposition require temperatures in the range of 600°C to 1200°C. In order to realize a Si device circuit architecture which integrates Si/Ge structures or the needed super abrupt junctions of buried channel CMOS, low temperature processes must replace those in current production lines. For these processes to be successfully developed and implemented, proper characterization techniques must be used. In the case of epitaxy, cross-sectional TEM is the tool of choice. We will discuss the prominent role that TEM has played in the development of a new Si epitaxy technology. Recently, at the National Renewable Energy Laboratory (NREL), we have shown low temperature, 195°C to 400°C, Si epitaxy via hot-wire chemical vapor deposition- HWCVD. In the past HWCVD has been used to produce amorphous, micro-crystalline, and polycrystalline Si thin films.

Nanosized semiconductors have been highly desirable as optoelectronic materials since the discovery of photoluminescence from porous nanostructured silicon. There are many investigations about visible light emission from nanostructured semiconductors, and several studies showed the size dependence of the photoluminescent and electronic property [1], [2]. In order to investigate the size effect of those optical properties of nanoparticles, the size-selection technique is important.

Size-selection of silicon nanoclusters has been carried out by crossing, an Ar molecular beam perpendicularly to the silicon cluster beam. The laser ablation in inert gas method was used for fabricating nanometer-sized silicon. Large size silicon clusters are produced by Nd:YAG laser irradiation of a silicon wafer which was rotating in ablation chamber under He-gas flow. Produced nanoparticles were extracted through a nozzle ( ϕ = 0.6 mm) to a differentially pumped chamber and then skimmed into a high vacuum chamber for size selection and deposition.

Recently there has been increasing interest in nickel induced lateral recrystallization of amorphous silicon because of its potential to improve device performance and to lower the thermal budget during processing. The hypothesis is that the formation of nickel silicide provides a low energy nucleus for the recrystallization of amorphous silicon. The silicide, moving into a-Si, leaves crystalline silicon behind.1 The grains formed, therefore, tend to elongated. In this paper, we attempt to use TEM to investigate in detail the nickel assisted lateral crystallization of amorphous silicon. The sample was prepared by first depositing a 1000A thick low temperature, oxide layer, LTO, on Corning 1737 glass. A 1000A thick amorphous silicon layer, a-Si, and 1000A thick a-Si were deposited subsequently. The sample was pattern and etched with hydrofluoric acid to form lOum x lOum holes in the oxide layer. Next, 200A of nickel was evaporated onto the sample, followed by a 600°C, 6 hours anneal to induce lateral recrystallization.

Zirconolite (CaZrTi207) is an important phase proposed for high level nuclear waste immobilization. Zirconolite was irradiated by 1 MeV Kr+ at various temperatures. At room temperature, zirconolite became amorphous after a dose of 7x1014 ions/cm2.1 Amorphization dose increased with temperature due to thermal annealing. The critical temperature, above which amorphization does not occur, was estimated to be 654 K. During the low temperature irradiation (<654 K), concurrent with amorphization, zirconolite transformed from a monoclinic structure to the cubic pyrochlore structure and then to the fluorite substructure. The structural change is due to the disordering between cations and between oxygen and oxygen vacancies.

After an irradiation at 673 K to a dose of 3.6x1015 ions/cm, the zirconolite samples remained crystalline. The diffraction pattern consists of strong maxima from the fluorite structure and diffuse maxima surrounding the Bragg positions of the pyrochlore superlattice (FIG. 1). Diffuse scattering patterns have been reported in other phases, and were generally attributed to the shortrange- order (SRO) domains.

Artificial defects generated by ion irradiation have been considered an efficient method to enhance the critical current density in superconducting materials. The mechanism of producing defects as flux pining centers is still an important issue since the efficiency of irradiation-induced defects in flux pinning strongly depends on their microstructures. Different types of defects have been found in heavy ion irradiation. However, there are few results that show light ion irradiation due to the target material selected, the type of light ion and energy, and the incident ion angle. Another factor is the difficulty of cross-sectional sample preparation. In this paper, a single crystal Bi2Sr2CaCu2O7-x with 11 MeV B5+ ion irradiation was observed by transmission electron microscope (TEM) from both plan and cross-sectional view.

The Bi2Sr2CaCu2O7-x single crystals used for ion irradiation were prepared using the floating-zone melting method. The crystals were cleaved into thin sheets of about 20 μm thickness along the a-b plane and cut to about 2mmx2mm size.

Irradiation-assisted stress corrosion cracking (IASCC) of irradiated austenitic stainless steels has been attributed to both microchemical (radiation-induced segregation (RIS)) and microstructural (radiation hardening) effects. The flux of radiation-induced point defects to grain boundaries results in the depletion of Cr and Mo and the enrichment of Ni, Si, and P at the boundaries. Similar to the association of stress corrosion cracking with the depletion of Cr and Mo in thermally sensitized stainless steels, IASCC is attributed in part to similar depletion by RIS. However, in specific heats of irradiated stainless steel, “W-shaped” Cr profiles have been observed with localized enrichment of Cr, Mo and P at grain boundaries. It has been show that such profiles arise from pre-existing segregation associated with intermediate rate cooling from elevated temperatures. However, the exact mechanism responsible for the pre-existing segregation has not been identified.

Two commercial heats of stainless steel (304CP and 316CP) were forced air cooled from elevated temperatures (∽1100°C) to produce pre-existing segregation.

TEM analyses of Si implanted with immiscible elements have shown that the implanted species characteristically segregate to form nanoscale inclusions in the amorphous Si matrix. Subsequen annealing induced recrystallization was found to lead to simultaneous removal of nearly all of the implanted species, explained in terms of stress-induced trapping of the implanted atoms at the amc phous/crystalline interface as it progresses towards the surface.

The solubility of Pb in Si is negligible, and it has previously been shown-that elemental nanoscale inclusions can be formed in Si either by ion implantation at elevated temperature or by rapid solidi cation. In the present work single crystals of <110> Si have been implanted with 80 keV Pb+ k at 300K, 625K and 925K respectively. The implanted fluence was 8-1019 m−2 corresponding to an average concentration of about 3 at.%. Combined TEM and RBS/channeling analysis showed that implantations at 300 K and 625 K induced amorphization of the implanted layer with dense distributions of nanoscale lead inclusions embedded in the amorphous silicon matrix (fig. 1).

When a noble gas element such as Xe is implanted in an fee metal matrix such as Al at room temperature, a fine dispersion of precipitates forms. The precipitates are elementary fee crystals up to diameters of several nanometers (for Xe in Al, 8-10 nm), above which they are non-crystalline. The precipitates exhibit a cube-on-cube orientation relation with the matrices and have lattice parameters which are much larger than those of the matrices (axe ≃ 1-5aAl). Thus the interphase interfaces are incommemsurate though the lattices are isotactic. The precipitates assume the shape of matrix cavities; for an Al matrix, at equilibrium this is a cuboctahedron, a {111} octahedron truncated at the corners on {100}. Fig. 1 is a sketch of a dispersion of such cuboctahedra, viewed approximately along a <110>.

For this study specimens were prepared in the HVEM-Tandem Facility at Argonne National Laboratory by implanting 35 keV Xe to a dose of 4x1019 m−2 into well-annealed 5N Al discs which had been thinned by jet electropolishing.

The Optical Solar Reflector (OSR) of Earth Observing System-PM (EOS-PM) is made of a flexible function layer of silvered Teflon/Inconel/Kapton laminate bonded to a honeycomb panel. After having been thermal cycled in vacuum, the laminate/panel was irradiated in air with UV at 2.2 equivalent suns intensity for 500 hours. Severe darkening was observed in areas exposed to UV but little or no darkening occurred in adjacent but shaded areas. Drastic increase of absorptance was observed-due to the darkening.

In order to investigate the cause of the degradation, test samples were irradiated at different configurations for 1000 equivalent sun hours (ESH). These samples were freestanding laminates and irradiated in vacuum and air, respectively. No darkening was observed on the sample that was irradiated in vacuum. Severe darkening was observed on the sample that was irradiated in the air. A laminate sample was irradiated in dry nitrogen for 1000 ESR and found no darkening either.

We report x-ray absorption near edge structure (XANES) and extended x-ray fine structure analysis (EXAFS) spectra from the plutonium Llll and cerium Lm edges in prototype titanate ceramic hosts for disposal of surplus fissile materials. These spectra were obtained using the MRCAT beamline at the Advanced Photon Source (APS). The XANES and EXAFS results are compared with electron loss spectra (EELS) determination of oxidation state from the plutonium MlV,V and cerium MlV,V edges [1,2]. The titanate ceramics studied are based upon the hafniumpyrochlore and zirconolite mineral structures and will serve as an immobilization host, containing as much as 10 weight % fissile plutonium, and 20 weight % (natural or depleted) uranium. Similar formulations were composed using cerium as a “surrogate” element, replacing both plutonium and uranium in the ceramic matrix. We find the plutonium to be present almost entirely as Pu (IV), while the cerium is clearly in a mixed III-IV oxidation state in the surrogate ceramic.

As the size of Si integrated circuit structures is continually reduced, interest in semiconductor-oninsulator (SOI) structures has heightened. SOI structures have already been developed for Si using oxygen ion implantation. However, the application of Si devices is limited due to the physical properties of Si. As an alternative to Si, SiC is a potentially important semiconductor for high-power, high-speed, and high-temperature electronic devices. Therefore, this material is a candidate for expanding the capabilities of Si-based technology. In this study, we performed oxygen ion implantation into bulk SiC to produce SiC-on-insulator structures. We examined the microstructures and compositional distributions in implanted specimens using transmission electron microscopy and a scanning transmission electron microscope equipped with an energy-dispersive X-ray spectrometer (STEM-EDX).

Figures 1(a) and 2(a) show bright-field images of 6H-SiC implanted with 180 keV oxygen ions at 650 °C to fluences of 7xl017 and 1.4xl018 cm−2, respectively. Three regions with distinct image contrast are apparent in Figs. 1(a) and 2(a), as indicated by A, B, and C.

The Al alloy composites reinforced with Si3N4 or SiC have been reported to exhibit superplasticity at high strain rate of faster than 1x10−2 s−1. It has been shown in many aluminum alloy composites that the optimum superplastic temperature coincides with an incipient melting temperature. The coincidence suggests a contribution of the liquid phase to the superplasticity mechanism. This paper shows a direct evidence of partial melting along matrix grain boundaries and matrix-reinforcement interfaces. Based on the obtained results, the role of the liquid phase in the high-strain-rate superplasticity is discussed.

The sample was Al-Mg (5052) alloy reinforced with 20vol% Si3,N4 particles, fabricated by a powder metallurgy process. The sample showed an excellent superplasticity under the conditions given in Table 1. Partial melting was confirmed to occur at 821 K by differentail scanning calorimetry. The microstructural changes during heating were observed in situ by TEM using a heating stage. The structure of interfaces and grain boundaries was observed by HREM. Chemical analysis was performed with EDS attached to VG-STEM.

Characteristics of the interfacial region between a polymer and substrate are critical to the durability of adhesive-adherend system performance. The High Speed Civil Transport (HSCT) requires structural laminates to be light weight, non-corrosive, environmentally safe, and able to withstand extreme service conditions (Mach 2.4 with in-service temperatures at 177°C, projected life to 120,000 hours). The present work characterizes multi-layer laminates comprised of a Titanium substrate (6A1-4V and 15V-3Cr-3Al-3Sn alloy), a silicate/zirconate sol-gel coating with aminopropyl functionality (nominally 1000 Angstroms), a NASA developed PETI-5 psuedothermoplastic polyimide primer (nominally 1000 Angstroms), and a high toughness, thermo-oxide stable PETI-5 adhesive. Previous studies report the physical effect of aging; therefore, this work aims to measure the nano-mechanical properties of the interface utilizing nano-indentation and AFM, focusing on the accelerated time-dependent changes due to extreme environmental exposure.

Samples were prepared by sectioning each laminate (No Environmental Exposure, 1000 and 2000 Hours at 200°C) in two locations, one section externally exposed to environmental test conditions and the other from the middle of the laminate.

Nanoindentation allows the hardness of thin coatings and synthetic multilayer structures to be measured, since indentation depths can be as little as a few 10s of nm. In combination with the cross-sectional transmission electron microscopy (TEM) analysis described here it is possible to observe the deformation structure under an indent, and potentially to understand deformation mechanisms on a nm scale in a wide variety of materials. Synthetic multilayers are a particularly interesting system to investigate. Variations in hardness with the multilayer compositional repeat distance (A) have been reported for several systems. The highest hardnesses, which are in excess of what a simple “rule of mixtures” would predict, occur in nitride multilayers at A ∼5nm. Here we present some preliminary results showing the deformation structure in both a monolithic NbN film and a TiN/NbN multilayer in which both components have the rQck salt structure with lattice parameters 0.424nm (TiN) and 0.439nm (NbN).

Lath-like silicon oxynitride crystals have often been observed in the microstructure of silicon nitride based ceramics after processing. They are usually located in glassy regions which are siliceous solidified sintering aid liquid, and usually contain a small (∼100nm) a-Si3N4 crystal. These nitride crystals are considered to be seeds, incompletely dissolved in the melt, that are heterogeneous nucleation sites for the oxynitride crystals. We present here the first observations of morphological and crystallographic habits between the seed nanocrystals and the host oxynitride laths.

Fig. 1 shows a typical oxynitride lath containing a nitride seed crystal. The lath is surrounded by glass and ß-Si3N4 particles, and a small cristobalite particle (a minor constituent). This microstructure is from an Si02-Si3N4 ceramic processed with Al2O3 sintering aid. The same oxynitride lath/seed structures were observed when other sintering aids (eg. Y2O3, MgO, ZrO2) were used, so they are independent of sintering aid.

Ferroelectrics are used in many applications such as non-volatile memory, infrared detectors, phased array radar and optical switches. While substantial progress has been made in the development of commercial ferroelectric devices, we still do not possess a good understanding of the fundamental processes in ferroelectrics. Trapped charge and oxygen vacancies are believed to strongly influence domain motion. In order to study these issues, we have designed an in situ TEM holder that can subject ferroelectric crystals to voltage, heat and UV irradiation.

Bulk BaTi03 crystals, grown by the Remeika method, were mechanically polished and thinned in hot ophosphoric acid. The resulting thin flakes were attached to copper rings with conductive carbon paint and electrical wires were glued to the copper rings, which act as electrodes. The sample was placed in the cradle of the in situ TEM holder and examined in a Hitachi H9000NAR TEM.

A series of homologous compounds InMO3(ZnO)m (M=In, Fe, Ga and Al, m=integer) has been synthesized successfully by the present authors. From x-ray diffraction, the strutcure is considered to be a layered structure, consisting of InO2−1-(In-O layer) interleaved with M/ZnmOm−l1+ (M/Zn-O layer). The space group is assigned to be R3m when m is an odd number, while it is P63/mmc when m is an even number. The distribution of M atoms within the M/Zn-O layers is considered to be random. However, from electron microscopy study, the existence of the modulated structures showing non-integral periodicities has been observed. The high spatial resolution EDS analysis using 0.5 nm electron probe diameter under a 300kV FE-TEM showed that the sinusoidal modulated structure is caused by the ordering of M atoms within the M/Zn-O layers. In the present study, modulated structures of InMO3(ZnO)13, where M atoms consists of two metals such as In and Fe, Ga, Al and Fe are studied by combined use of HRTEM and high spatial resolution EDS analysis using a 300kV FE-TEM.

The fiber/matrix interfaces play a critical role in the mechanical performance of fiber-reinforced ceramic-matrix composites (CMCs). In this vein, the placement of an interphase material between the fiber and matrix phases has received considerable attention. The main purpose of the interphase coating is to enhance toughness of the composite by influencing crack trajectories at the mechanically critical (weak) fiber/coating interface.

This interest in coated-fiber composites has been accompanied by the need for accurate microanalysis of the CMC systems, owing to issues of both morphological and chemical incompatibilities between the dissimilar fiber and coating materials. Consequently, microanalysis has been driven below the micrometer scale, into the length scale of nanometers. Thus, the examination of the fiber/coating interfaces in CMCs falls squarely into the realm of transmission electron microscope (TEM) and scanning transmission electron microscopy (STEM) analyses.

However, the success of TEM/STEM-based microanalysis of CMCs hinges critically on the quality of TEM specimen.

The establishment of ALCHEMI (atom location by channeling-enhanced microanalysis) as a viable microanalytical technique owes much to its early success in determining cation distributions in crystals having the structure of the mineral spinel, MgAl2O4. The first application of ALCHEMI was to a synthetic chromite spinel, and geological applications to natural mineral spinels containing multiple 3d transition elements provided many of the early applications of the technique. Spinel has three distinct regular lattice sites, two cation sites and one anion site, occupying the 8a, 16d and 32e Wyckoff positions of the Fd3m space group. In projection along (800), tetrahedrally coordinated (8a) cations occupy one set of alternate planes, whereas octahedrally coordinated (16d) cations and, to a good approximation, the anions occupy the other set. In principle, therefore, ALCHEMI performed at an 800 planar channeling condition could be used to analyze site distributions in spinel, using oxygen as a marker for the planes containing the octahedral cations, as was early noted.

The most widely spread application of energy filtering transmission electron microscopy (EFTEM) is the determination of elemental distributions in analytical investigations. The new EFTEM techniques are based on data acquisition by electron spectroscopic imaging (ESI), which offers the advantage of fast two-dimensional recording of the inner-shell loss intensities. In a more general framework, the new ESI methods and the standard PEELS technique both aim at exploring the same three dimensional data space. This data space is represented by the spatial coordinates x, y and the energy loss ΔE and the resulting intensity distribution I(x,y,ΔE) is the quantity which is evaluated. It is intuitively clear, that acquiring more than just three consecutive ESI images gives us the possibility to get quantitative information about the spectra and the shape of the ionization edge. The information on the EEL spectrum can be extracted from a series of ESI images for each individual pixel or, by integrating over corresponding pixels, for a given image area and can be quantified by applying standard EELS quantification techniques.