One of the ultimate objectives for energy-dispersive X-ray spectrometry (EDS) in the analytical electron microscope (AEM) is single-atom detection in thin specimens, as well as quantitative microanalysis with high accuracy approaching ±1% relative. In order to realize the single-atom analysis, the design of the AEM has to be optimized with respect to improvements in spatial resolution and detectability limits. The detectability limit, as defined by the minimum mass fraction (MMF), is given by:

where P is the peak intensity of interest, (P/B) is the peak-to-background ratio for that peak, and r is the acquisition time. To improve the sensitivity for analysis, any or all of the variables P, (P/B), and τ should be increased. Intermediate-voltage analytical electron microscopes combined with high brightness field-emission gun (FEG) are expected to improve the MMF, while maintaining high spatial resolution. Additionally, the MMF should also be improved by maximizing the solid angle of X-ray collection.

A study of reactive phase formation in thin films requires the understanding of the thermodynamics and kinetics of the reactions. Previous investigations of phase formation in Nb/Al multilayers using calorimetry and x-ray diffraction have observed that despite the presence of rather large driving forces for phase formation, a pronounced nucleation event occurs. This behavior has been reported in other systems including Ti/Al and Ni/Al. The initial nature of the interface can affect the nucleation, therefore careful studies of the structure and composition of the initial interfacial region may explain the reason for a kinetically separated nucleation and lateral growth step preceding normal phase growth. In addition, examination of the interface may also lend insight into the effects of different deposition processes and can provide evidence for the formation of solid solutions or compounds that can lower driving forces available for phase formation. Finally, the cross-sectional multilayer sample provides a means of characterizing spatial resolution and isolating specimen preparation artifacts.

The distribution of Nb and Fe in Zr-2.5wt%Nb pressure tube alloys has been of interest, in particular the effects of neutron irradiation. Both Fe in the β-phase and Nb in the α-phase are present in levels below lwt% and fluctuations of these elements in different parts of a specimen or between different specimens are of interest. In EDS quantification the signal-to-noise ratio and hence the detection limit is determined by the signals but depends also on the data analysis, in particular the background subtraction procedure. Here we compare the noise as given by the commercial analysis software to a simple estimate based on counting statistics. We used a VG HB 501 UX equipped with a thin window HP Ge detector, digital pulse processor and Voyager 3105 data acquisition and analysis system (Noran).In the past artefacts of Fe, Ni and Cu had been observed in EDS spectra, even for materials which did not contain any impurities of these elements.

A spectrum image can be acquired with a scanning transmission electron microscope (STEM) by translating a focused probe over a two-dimensional area of a specimen and acquiring a spectrum at an array of positions of the probe. Analogously, a rocking-beam spectrum image is formed when the incident electron beam is tilted relative to the specimen and a series of spectra is acquired from a single volume of the specimen. The orientation-dependence of such a series of spectra can be used to extract the atomic site-distributions of an ordered alloy or compound. One of the drawbacks of spectrum image acquisition is that the raw images are typically files of many megabytes, whereas the actual information of interest may be reducible to a fraction of this size. The dimension of the data set may be reduced by a number of methods. For example, Rossouw and coworkers have acquired rocking-beam X-ray maps for the characterization of atomic site-distributions in titanium aluminide intermetallic alloys.

Over the last ten years TEMs have been developed that are capable of HREM, EDX, PEELS and diffraction using a single objective pole piece. More recently these TEMs have been equipped with the capability of energy filtering the electron beam after it has passed through the sample so that energy filtered images and electron diffraction patterns can be obtained. In this work the use of a Topcon 002B TEM equipped with a GATAN PEELS imaging filter (GIF) to generate zero-loss energy filtered zone axis CBED patterns and elemental images from inelastically scattered electrons will be described. An analysis of this energy filtered data indicates that elemental imaging using the GIF is an informative, but semiquantitative technique, whereas zero-loss energy filtered zone axis CBED patterns can be accurately quantified so that the two lowest-angle x-ray form factors of cubic elements can be measured with errors of the order of 0.1% or less.

Alloy X-750 is a γ -strengthened, nickel-base alloy used in demanding high-temperature applications. The HTH commercial heat treatment condition provides excellent strength and good corrosion resistance. However, the resultant precipitate structure is complex with fine (˜20 nm) intragranular γ' phase and copious intergranular precipitation of at least four phases (γ', M23C6, M23B6, and TiN). The intergranular precipitation causes localized grain boundary migration and results in a convoluted grain boundary structure. Such complex grain boundary microstructures increase the difficulty of phase identification and interfacial composition measurements by traditional analytical electron microscopy methods. Elemental mapping by EFTEM is a useful additional or alternative technique for characterizing such structures. A Gatan Imaging Filter (GIF) on a Philips CM30 (LaB6) was used in the current investigation. Experimental details have been summarized elsewhere.

Elemental maps of Cr, Ti, and Ni (net L23 intensities) are presented in Fig. la-c for a typical grain boundary region of a HTH heat treated specimen.

In order to obtain high spatial resolution microchemical data from beam-sensitive materials, alternative techniques which avoid the use of fine probes, such as energy-filtered imaging techniques must be used. Robust quantification procedures are currently under development. The quantification procedure which yields images whose intensities are proportional to the elemental concentrations (atoms per unit volume) has already been determined. The net core-loss intensity images, Si, can be corrected for diffraction contrast by normalization with low loss images, whereas thickness variations are corrected by normalization with t/λ, maps. This procedure has been shown to give the correct value for the concentration ratio between two phases for a single elemental map.1 Determining concentration ratios between two elements in the same phase can actually be simpler, in theory, than the above procedure since it only requires dividing the net core-loss intensity images of two elements and correcting for the difference in cross-section (Cx/Cy = [Sx/Sy][(σy/(σx])- However, there are a few practical factors that must be considered in practice.

ESI (electron spectroscopic imaging)-detection methods are mainly based on the core edges above 100 eV (see, as an example, Ref. 1). However, it would be of major interest to use the high intensity values in the low-loss region below 100 eV. There are at least two elements of general interest for biological applications showing core edges in this region: Au (O2,3 ) and Fe (M 2,3 )• Since colloidal gold and ferritin are used to label targets in biological samples at high spatial resolution we started our study aiming at iron-detection (iron-maximum at 62 eV comparable to gold-maximum at 65 eV).

Ferritin (Ref. 2) contains about 200 to 2000 iron atoms in a densely packed core representing an ideal prerequisite for ESI-detection. Furthermore, ferritin used to be a common marker in conventional transmission electron microscopy in the past. The contrast produced by ferritin in a conventional transmission electron microscope (CTEM) is not specific for iron but only for a dense region in the specimen.

Though stainless steels are important technologically for a wide range of applications, they are not generally known for their very high strength. We have rapid-solidification-processed many stainless steels by gas atomization and achieved strength improvements of over 50% relative to conventionally-processed stainless steels with concomitant improvement in corrosion and oxidation behavior. These strength improvements are most pronounced after aging treatments when elevated concentrations of oxygen and vanadium are present in the stainless steel. An austenitic (FCC) stainless steel (Fe-16%Ni-9%Cr-0.5%Mn-0.2%V-0.0137%N-0.008%O by weight) was prepared by gas atomization and consolidated by hot extrusion at 900°C. These specimens were heat treated for 1 hour at 1000°C and aged at 600°C for 500 hours.

The microstructure of each alloy composition was observed in TEM with bright field imaging. After aging, most alloys showed the same precipitate morphology as before aging. An obvious change, however, was found only in the alloy with highest oxygen content.

It has long been the bane of analytical electron microscopy (AEM) that the use of focused probes during microanalysis of specimens increases the local rate of hydrocarbon contamination. This is most succinctly observed by the formation of contamination deposits (fig.l) during focused probe work typical of AEM studies. While serving to indicate the location of the electron probe, the contamination obliterates the area of the specimen being analyzed and adversely affects all quantitative microanalysis methodologies. A variety of methods including: UV, electron beam flooding, heating and/or cooling can decrease the rate of contamination, however, none of these methods directly attack the source of specimen borne contamination . Research has shown that reactive gas plasmas may be used to clean both the specimen and stage for AEM, in this study we report on quantitative measurements of the reduction in contamination rates in an AEM as a function of operating conditions and plasma gases.

The issue of specimen contamination becomes more important at a rate proportional to the use of high-brightness electron source Transmission Electron Microscopes (TEM). These TEMs combine smaller electron probes with increased current, allowing high resolution specimen imaging and enhanced analytical data collection. Small, high current electron probes tend to increase the detrimental effect of hydrocarbon presence on the specimen's surface. The subsequent formation of carbon deposits caused by the focused probe often times obstructs imaging and precludes acceptable analytical results. by plasma cleaning the specimen, contamination is removed and the results obtained by high resolution electron microscopy (HREM), scanning transmission electron microscopy (STEM) and analytical electron microscopy (AEM) using EDS or electron energy loss spectroscopy (EELS) are greatly enhanced. Recent instrumentation developments have resulted in the application of a high frequency, low energy, reactive gas plasma generated in an oil-free vacuum that chemically removes hydrocarbon contamination from both the TEM specimen holder and the specimen without altering its properties.

Whereas efficient red (ZnS:Mn) and green (ZnS:Tb) phosphors are available for full-color flat-panel display technology, efficient blue phosphors are still under development. SrS:Ce is being investigated as a suitable material. As part of a larger study, annealed SrS:Ce films produced from sputter targets incorporating Ga2S3 were found to be five times brighter than films produced from targets containing no Ga2S3. Consequently, the significance of added gallium sulfide to the morphology of SrS:Ce films during annealing is being investigated.

Following deposition, plan view specimens of films were prepared for transmission electron microscopy by mechanical grinding, dimpling, and Ar+ milling. Films were examined in a Philips CM 12, and a JEOL 200CX. Additionally, EDS line scans were obtained in the scanning-transmission mode of a Philips CM200FEG, integrated with an EMiSPEC Vision acquisition system. The EDS line scans were defined as a series of 40 points along a line, spaced approximately 3.6 nm apart.

Electron energy loss spectroscopy (EELS) is ideally suited for the study of oxide materials. The large cross section for the oxygen K-absorption edge generates enough signal for fine-structure analysis even when the probe is small enough for atomic resolution imaging (∼2Å). Therefore, by using the Z-contrast imaging technique in the STEM to locate defects and position the probe with atomic resolution, detailed spectroscopic information can be obtained from a known atomic location. by selecting the appropriate acquisition conditions, these spectra can have the same atomic resolution and incoherent features as the Z-contrast image. The primary benefit of this methodology is that the image provides a structural model for interpretation of the spectrum through theoretical modeling.

One means of interpreting this spectral fine-structure is through multiple scattering analysis. In the multiple scattering formulation, the fine-structure of the absorption edge is considered to arise from interference effects resulting from the scattering of an excited photoelectron from neighboring atoms.

Martensitic transformation is the first-order diffusionless phase transformation in solids, and is responsible for unique phenomena such as shape memory effect and superelasticity. Recently, many researchers tried to explain the origin of martensitic transformations from a viewpoint of electronic structure. It is strongly required to investigate electronic state changes associated with a martensitic transformation by an accurate experiment. EELS (Electron Energy-Loss Spectroscopy) was successfully applied to study electronic state in several alloys and oxides, e.g. a composition dependence of density-of-state (DOS) in Cu-3d band was clearly observed for Cu1-xAlx alloys by a core-loss measurement. However, this method has been less applied to studies of phase transformations to date. The purpose of the present work is to investigate electronic state changes associated with a martensitic transformation in a Ti50Ni48Fe2 alloy by EELS.

A Ti50Ni48Fe2 alloy was prepared by induction-melting method. Thin-foiled specimens were solution-treated in Ar atmosphere at 1173K for 1hr.,

The metal-ceramic interface between copper and alumina has been studied by TEM, Electron Energy-Loss Spectroscopy (EELS), and Energy Dispersive X-ray spectroscopy (EDX). of partic-ular interest is the potential role of silicon which is the major impurity in commercial alumina. This investigation focuses on the presence and distribution of silicon within the bulk alumina and at the diffusion bonded CU-AI2O3 interface using a Topcon 002B TEM equipped with an EDAX EDX detector and a Gatan Imaging Filter. The interface was created under vacuum by diffusion bonding of 100/xm copper foils (99.999% purity) pressed between polished alumina substrates (∼99.5% purity) for several hours at ∼90% of the melting temperature of the metal.

An initial investigation of the bulk AI2O3 indicated that the majority of the Si occurred at the triple points. A typical example is shown in Fig. la. EDX spectra acquired with a 6nm probe from the center of the junction (Fig. lbi) and at the grain boundary indicate that the Si wets the grain boundary to a limited extent.

The intermetallic compound TiAl is a promising material for applications at high stress and high temperature. It has good balance between lightness, strength and oxidation resistance, but suffers of an intrinsic brittleness at lower temperature. Small, ternary additions of V, Cr or Mn are known to increase the ductility of this material. In the present and a recent study we have used electron energy loss spectroscopy as a sensitive test to investigate the electronic structure in these materials. The near edge structure (NES) may give complementary information to e.g. convergent beam electron diffraction (CBED) about the bonding character in these materials.

EELS spectra were collected using a Gatan 666 spectrometer with parallel recording fitted to a Philips CM 30 transmission electron microscope operating at a nominal voltage of 150 kV. The energy resolution (measured as full width at half maximum) was typically around 1 eV. Spectra from areas of similar thickness, as measured relative to the total inelastic mean free path of the 150-keV electrons were obtained.

ELFS spectroscopy (energy loss fine structure) is used to obtain local atomic structure information It can outperform XAFS (x-ray absorption fine structure) not only because of its low Z element sensitivity, but also because of its high spatial resolution and the capability of combining other high resolution TEM measurements. Although TEM continues to gain importance as an indispensable and unique tool to study nanoscale phenomena by providing simultaneous imaging, diffraction, and spectroscopy information, direct observation and quantitative measurements of physical phenomena are also desirable. This paper gives a first-time demonstration of such a measurement, namely local temperature determination in a TEM sample by ELFS.

The principle is simple and as follows. One can measure, with ELFS, the atomic distances up to ±0.01 Å accuracy for the fist shell (typically around 2 Å ) and the second shell (around 3-4 Å). Atomic distances in a sample will change when its temperature changes, the phenomena that are coupled by the macroscopic temperature dependent lattice expansion.

In an EFTEM the full range of signals generated by interaction of the primary electron beam with the specimen can be detected. Thus operating such a system and generating combined digital information usually is a rather complex issue. The demands of the users on the other hand are to achieve results fast, easily and reproducibly. Moreover it should be possible to tailor the integral system according to dedicated needs. In general it should be no problem to use modern digital equipment and just let an integral computer control everything. However, in order to make such digital settings really useful, there should be no D/A conversion in between the data paths of the microscope because any analogue system tends to drift and the changes of lens parameters between different modes of operation should be minimised to overcome hysteresis. Fully digitised in-column EFTEMs like the LEO EFTEMs with OMEGA filter and Koehler illumination including also multiple parallel and serial remote capabilities provide optimum preconditions to fulfil the demands.

Although crystal lattices can be determined reasonably well using high-resolution electron microscopy, determination of local chemistry at high spatial-resolution remains a challenge. An energy-filtering system has made it possible to perform chemically sensitive imaging in a transmission electron microscope (TEM). This type of imaging usually relies on the signal of the inner shell ionization edge, the intensity of which is affected by the threshold energy-loss and the ionization cross-section of the edge. Therefore, the spatial resolution of a core loss image image is strongly affected by the signal-to-noise ratio. In this respect, lower loss electrons from the plasmon or valence region of the energy loss spectrum (10-100 eV) are favorable for chemically sensitive imaging due to the much higher signal intensity, provided any delocalization effects are small in comparison to the required spatial resolution. Compositionally sensitive imaging using the aluminum plasmon energy-loss electrons has been shown to produce ∼2nm resolution for an atomically sharp Al/Ti interface.

The physical properties of diamond are some of the most interesting of natural materials. Electron energy-loss spectroscopy studies of a single crystal diamond wedge allow some of the basic electron matter interactions to be studied. The electron energy-loss spectrum has been examined with particular reference to the plasmon excitations in the crystal. The mean-free path of the electrons through the diamond crystal and the nature of the observed features in the energy-loss spectrum have been examined.

A single crystal of diamond was cut and polished to produce a diamond wedge with a 33° angle making it suitable for observation in a transmission electron microscope (Fig. 1). The determination of the geometrical parameters of the crystal wedge involved the analysis of pendellosung fringes (Fig. 2) and comparisons to bloch wave calculations. The gradient analysis plot shown in Fig. 3 allowed the wedge thickness to be determined at any distance from the edge.