An atomic model of the TiO2(100)-1x3 surface has been determined based on Grazing Incidence X-Ray Diffraction (GIXD) data analyzed by direct methods coupled with χ2 minimization. The final model has every surface octahedral site occupied, and resembles the NaCl structure, in contrast to the normal structure of bulk rutile which has half of the sites occupied (Fig. 1). This structure is non-stoichiometric which is consistent with the observed spectroscopic data that indicates that the surface is titanium rich. The structure contains edge and face sharing octahedra, present in other known defect structures in TiO2, and appear consistent with what has been observed in Scanning Tunneling Microscopy.

The data consisted of 19 measured structure factors acquired from GIXD. The phases for the measured beams were calculated using a Sayre equation with Unitary Structure factors and a Genetic Algorithm global search technique. Based upon the calculated phases, charge density maps were constructed and used to determine the positions of titanium atoms within the unit cell (Fig. 2).

Two two-dimensional atomic models were constructed to test the application of Direct Methods for solving surface crystal structures (atomic positions are listed in Table 1). The use of Direct Methods to solve three dimensional structures from X-ray diffraction has been well established for several years. However to the best of our knowledge never before had this technique been applied to 2-D surface structures. A Genetic Algorithm (GA) search routine was implemented for exploring possible solutions.

The models were constructed using electron scattering factors to calculate the intensities and phases for beams out to 1 Å resolution (Fig. 1). Since electron diffraction from surfaces has been shown to be nearly kinematical X-ray and electron scattering can be handled identically. An error term was applied to the generated intensities to simulate realistic sampling errors. The generated phases for each model were removed and later used to monitor the solutions calculated by the direct methods.

Electron crystallography of thin three-dimensional (3D) protein crystals requires very exact determination of tilt angles and spot profiles to obtain correct merging of diffraction spot amplitudes. The reciprocal lattice of 3D microcrystals consists of ellipsoidal spot profiles which are very extended in the direction normal to the crystal face (z*). To extrapolate from the intensity measured in a section to the total spot intensity, two features need to be known very exactly: 1. the orientation of reciprocal lattice relative to the Ewald sphere, 2. the 3D-shape of the spot cloud.

Fig. 1 shows a tilt series of one frozen hydrated catalase crystal, in the order of recording. The third diffraction pattern gives the highest resolution because it is untilted and therefore the electrons have the shortest path length. In the current experimental data taken at 120 keV electron energy inelastic scattering within the crystal leads to a dramatic loss of elastic information in highly tilted patterns.

Silicon carbide (SiC) is a widely used ceramic material, with many structural and electronic applications. It exists in many polytypes, differing from one another only by the stacking sequence of close packed double layers of Si and C atoms. The polytype called 4H has the hexagonal structure shown in figure 1. The double layers here have a stacking sequence of ABACABAC.. The distance z between the Si and the C layers (shown in figure 1) is an adjustable parameter, which is not exactly known. The aim of this work is to determine the atomic positions in the c-direction by quantitative convergent beam electron diffraction (QCBED). The goal is to develop a general refinement approach for structure determination by electron diffraction. Many newly synthesized materials are available in only very small quantities in the single crystal form and/or mixed with other phases, making X-ray diffraction methods difficult. SiC is often full of stacking faults. For these types of materials, the CBED method is ideal because of the small probe that can be used; areas of less than 100 Å can be studied.

Cu2O is a semiconductor with a band gap of 2.17eV, and its crystal structure is simple cubic. Oxygen atoms form a bcc structure and each oxygen atom is tetrahedrally coordinated by copper atoms, which makes copper linearly coordinated by two oxygen atoms. The structure of cuprite is interesting because the charge state of Cu and oxygen. The most common charge states of Cu is 2+, however, in cuprite for each oxygen there are two Cu atoms which suggest a 1+ charge state of Cu. Charge density of Cu and oxygen are also interesting because of the high Tc superconductors based on copper and oxygen planes and chains. The superconductivity is believed due to the different charge states of copper and/or oxygen in the Cu-oxygen planes due to the doping or oxygen deficiency. by studying the naturally occurring charge states of Cu in Cu2O and CuO, a better understanding of the relationship between structure and bonding can be achieved for the important copper oxides.

The phase shift imparted to an incident high-energy electron wave in a TEM is related to the specimen’s electron-refractive properties. These, in turn, are related to the electrostatic potential and, by Fourier transform (1), to the electron scattering factors fei(s) for the various atom species i in the specimen and scattering vectors s. The average refractive index is determined by the mean electrostatic (inner) potential, Φo, and can be modelled as Φo = (C/Ω) Σfei(s0) [equation 1] where C = 47.878 (V-Å2) and the summation runs over all of the atoms in the unit volume Ω (2). Calculated fei(s) data are available from the literature (e.g. 3). These calculations have only been done for neutral atoms and some fully ionized cations and anions. They do not account for electron redistribution due to covalent bonding to which Φo is quite sensitive (4).

This research is making Φo measurements using transmission electron holography. Holograms were collected using a 200keV Philips CM20 FEG TEM equipped with a non-rotatable biprism (5) and a Gatan 794 Multiscan camera.

In solving complex inorganic structures, often the greatest challenge is to obtain good initial atom positions for Rietveld refinement using x-ray diffraction intensities. In this work, a technique is presented which combines direct phasing with HRTEM and exploits dynamical diffraction effects to obtain an image of oxygen atoms. The technique was used in the structure determination of a monoclinic (Ga0.7In0.3)2SnO5 phase to provide accurate first-guess positions for oxygen atoms for subesquent Rietveld refinement.

The solution of the cation positions of the structure relied on HRTEM images. The image in Fig. 1, taken along the crystal’s short b-axis, has a defocus of approximately -50 nm based on the locations of minima in a weak amorphous component in the power spectra of the through-focus image series. Because this is near Scherzer defocus, one expects dark areas of the image to correspond a large projected potential (making the weak phase object approximation). This was used to get successful initial cation positions for a Rietveld refinement of powder x-ray diffraction data. The final refined structure is shown in Fig. 2, including the oxygen positions.

Electron holography can lend crucial insights to understanding the subtle manifestations of electro-magnetism in a wide range of materials. Whereas conventional microscopy is sensitive only to the intensity, holography reveals the phase changes in coherent electron wavefronts making it a unique tool to probe electric and magnetic fields on the nanometer scale. We have employed electron holography to characterize materials for mean inner potential measurements and also their electric and magnetic properties.

Electron holograms were acquired in a 300 kV FE-TEM under two optical conditions: the standard high resolution mode was employed for mean inner potential measurements; to examine the intrinsic electromagnetic states, a lower resolution mode was used whereby the objective lens is turned off and the diffraction lens images the specimen. Digitally acquired holograms were reconstructed with the HolograFREE software package.

Nanophase TiO2 particles generated in a flame burner system were found to have unusual central features. The rutile particles appear to contain faceted voids, raising the question whether the feature is truly a void or a secondary amorphous phase.

The exploration of light for applications in optical integrated devices has received world wide attention during the past decade.Organic compounds are of interest because of their large hyperpolarisabilities β and fast response times as well as offering many diferent types of processing geometry. It has been generally understood in the past that the requirements which must be satisfied by the microscopic system (i.e.the molecule) in order for the macroscopic system (i.e. the crystal ) to display second harmonic generation are the following: (a) the molecule must have a dipole moment μ (b) the molecule must have large components of hyperpolarisability β (c) the molecules must adopt a mutual orientation in the unit cell which avoids centro-symmetry .

In practice, TATB was found to have a very large optical susceptibility,although it did not satisfy conditions (a) and (c). With regard to condition (c), it was suggested that structural defects might explain the appearance of second harmonic generation in a centro-symmetric crystal. For this reason we decided to re-examine the problem using electron microscopy and diffraction.

Atomic structure of hetero-interfaces such as semiconductor/oxide is recently much interested from the viewpoints of the device engineering as well as the interface science. Studies of the interface structures have been performed by transmission electron microscopy (TEM)[1], large-angle convergent beam electron diffraction[2] in the cross-sectional observation mode, and by X-ray diffraction[3], TEM[4] and CBED[5] in the plan-view mode. In the present study we applied a coherent convergent beam electron diffraction (c-CBED) technique to the analysis of the rigid-body shift between vacuum-deposited PbTe and MgO (001) thin crystals in the plan-view mode. The c-CBED technique was originated in nano-diffraction experiments by Cowley[6] and the theoretical analysis by Spence[7]. These results suggested a possibility of direct structure-phase determination from the interference fringes formed at coherent overlaps between the diffraction disks. This technique has been recently revived using TEM on SiC by Vine et al.[8] The present study is the first to be concerned with the analysis of interference fringes in c-CBED patterns from epitaxially grown bi-crystals[9].

VDAC is a voltage-gated ion and metabolite channel that occurs at high density in the mitochondrial outer membrane. Although VDAC is probably related structurally to bacterial porins, small transmembrane voltages cause it to undergo reversible, partial closures that are not seen with the prokaryotic pores. The “closed” states, which are impermeable to ATP, can be induced by effectors, including a synthetic polyanion. There is evidence that closure involves major rearrangements of the pore structure that are difficult to explain in terms of porin-like β-barrels.

The main source of information about the structure of VDAC is electron microscopy of 2D crystals obtained by phospholipase treatment of outer membranes of fungal mitochondria. The unit cell observed after partial lipid hydrolysis (a = 13.3 nm, b = 11.5 nm, γ = 109°) contains six pores which appear to be structurally equivalent at the resolution of correlation averages of crystals embedded in aurothioglucose or vitreous ice (∼1/1.5 nm−1).

We have developed a novel cylindrical mirror analyzer (CMA) for the quantitative Auger electron spectroscopy (AES) to obtain reference electron spectra. This CMA was defined in all respects, thus it yields an absolute electron energy distribution ; the energy resolution of 0.25% for 42.3° ± 6° of incidence angles (theoretically obtainable resolution), the estimated transmission of 6.2% (assuming cosine law), and the signal detection by the combination of Faraday cup and electrometer (SI traceable). The acceleration voltages for the primary electron beam were ranging 1 through 5000 volts and the energy range of the sweep was 0 through 5000 eV in 0.1 eV(min)/step for the whole range and 10 meV or even 0.1 meV/step for the limited narrow range. The energy calibration by using the primary acceleration voltage (SI traceable) and the spectral intensity correction for the relativity have been performed.

The spectra for typical samples have already been published, i.e., C's(soot and graphite), Al(poly), Si(100and 111), Cr(poly), Fe(poly), Ni(poly), Cu(poly), Ga(liquid), Ag(poly), and Au(poly). Among the samples C's and Ga(liquid) seem to be convincing materials.

Recently there has been a renaissance in scanning cathodoluminescence (CL) microscopy and microanalysis primarily brought about by the analytical demands of the semiconductor industry in addition to significant advances in photonics technology (high efficiency PMTs, CCDs and CL collectors). The strength of the CL analysis technique lies in its ability to provide high spatial (lateral and depth) resolution concentration and distribution information about: (i) the chemical state (identity, oxidation state and co-ordination) of trace level impurities and (ii) point / extended structural defects (vacancies and dislocations) in both semiconductors and insulators.

For many materials, interpretation of CL spectra and images measured at low temperature (5 K) is quite straightforward. However CL generation via recombination of electron / hole pairs is a competitive process. Consequently in some specimens the measured CL intensity is not directly proportional to the concentration of the particular luminescent center but relates to the concentration of other radiative or non-radiative centers. In these cases it is difficult to decide whether contrast in the CL image is due to variation in the concentration of the radiative center or the competitive center.

Because of analytical advances, submicron particles as small as 0.2 μm can be characterized for chemical composition, size, and shape using scanning electron microscopy (SEM). Once these characteristics are determined, cluster analysis can be used to group the individual particles into categories based on size, shape, and chemical composition.

Submicron particle analysis is important when characterizing the ash to provide information to help solve ash-related problems in coal combustion and gasification systems. Since a combustion system has an excess of oxygen available, the resulting ash is typically easier to characterize than ash from a gasification system. In a gasification system, the lack of oxygen results in additional categories high in Cl and P.

Adequate dispersion of submicron particles for proper analysis of individual particles is required because of the analysis volume of the SEM beam. Therefore, an aliquot of an aqueous solution with a set sample-to-solvent ratio is drawn while being sonicated. A few drops of solution are placed on a vitreous carbon substrate, allowing for particle dispersion on the surface of a smooth substrate. Next, the particles are analyzed by the fine-particle technique (FPT).

Si(Li) and Ge Energy Dispersive Spectroscopy (EDS) detectors are commonly used for x-ray microanalysis because they are easy to use, inexpensive to operate, and offer both rapid qualitative evaluation of chemical composition and accurate quantitative analysis. Unfortunately, they are limited by energy resolutions on the order of 100 eV, which is insufficient to resolve many important overlapping x-ray peaks in materials of industrial interest, such as the Si Kα and W Mα peak overlap in WSi2. Although WDS spectrometers with excellent energy resolution (typically 2 eV to 10 eV) can resolve most peak overlaps, qualitative WDS analysis is limited by the need to serially scan over the entire energy range using multiple diffraction crystals. There is a need for a new generation of x-ray spectrometers for microanalysis that combines the excellent energy resolution of WDS spectrometers with the ease of use and the parallel energy detection capability of EDS spectrometers.

We are developing a high-energy-resolution x-ray microcalorimeter spectrometer for use in x-ray microanalysis. Our microcalorimeter spectrometer consists of a superconducting transition-edge microcalorimeter cooled to an operating temperature of 100 mK by a compact adiabatic demagnetization refrigerator mounted on a SEM column, read-out SQUID (Superconducting Quantum Interference Device) electronics followed by pulse-shaping amplifiers and pile-up rejection circuitry, and a multichannel analyzer with real-time computer interface.

Although the performance of high-energy-resolution microcalorimeter spectrometers for x-ray microanalysis is encouraging, the future widespread acceptance of these spectrometers as valuable microanalysis instruments depends on improvements in both achievable count rate and geometrical x-ray collection efficiency. While the maximum output count rate of our microcalorimeter (∼160 s−1) is much less than that of conventional EDS detectors operating at their highest energy resolution (∼3000 s−1), we are confident that we can significantly improve the count rate without loss of energy resolution (∼10 eV FWHM over a broad energy range). Increasing the area (and thus solid angle) of the microcalorimeter is a more difficult problem, however, as the best microcalorimeter performance is achieved using small-area (typically 250 μm by 250 μm) absorbers with low heat capacity.

This problem can be solved by using an x-ray lens to increase the collection efficiency of the microcalorimeter spectrometer. A polycapillary optic consisting of tens of thousands of fused capillaries can collect x-rays from a point x-ray source over a large solid angle and focus the x-rays onto the small-area absorber of the microcalorimeter.

The aim of the present study is to discuss the possibility of “counting atoms” by laser ionization technique, which has been applied to post-ionization in sputtered neutral mass spectrometry. The detection of one H2 molecule was tried under an extremely high vacuum (XHV) condition, since the molecule has a higher ionization energy (15.4 eV) than most of elements and the existence probability of it within laser ionization region is less than one under the condition.

Fig.1 shows schematically the experimental setup. The second harmonics (SH) of a picosecond YAG laser was focused with a spherical lens (f=250 mm) into the center of a home-made XHV chamber (Fig. 1a). The focused laser power density (W) is more than 1013W/cm2 for input pulse energy of 30 mJ, and the saturation of H2 ionization can occur under the condition. The repetition rate of the laser was 10 Hz. The XHV chamber can be evacuated to a minimum pressure of 5×10−11n Pa. To the chamber, either an ion counting detector or an ion imaging detector was attached. The ion counting detector (Fig.1b) is composed of ion collecting electrodes, an electron multiplier (EM), and a pulse counter.

The major demand for imaging light element specimens in the TEM is to achieve the appropriate contrast. This is particularly true in biomedical and also in polymer fields of application. The main reason for contrast being challenging is the correlation between the atomic number of the constituting elements and the scattering angle and thus the very low scattering absorption contrast achievable.

Different conventional methods or combinations of them can be used to overcome the problem. 1) Contrast is created chemically by staining with heavy elements. Disadvantages besides the high level of preparational efforts are the artificial unnatural view in general and the high potential of artifacts in particular. 2) Contrast is created physically using small objective apertures and/or long focal length objective lenses. Disadvantages are clearly increased alignment and cleaning efforts for small apertures and the complete loss of high frequency/high resolution information. Moreover, some elastic scattering is still required. 3) Contrast is created physically converting phase contrast into absorption contrast using the phase contrast transfer function of the microscope and high defocus values.

Understanding the contrast formation mechanism in the EM is one of the prerequisites for artefact-free reconstruction of biological structures from images. We found that the normally used correction of contrast formation applied to zero energy loss filtered images corrupted spatial resolution. Therefore the contribution of contrast formed by inelastic electrons was reconsidered, including partial coherence of inelastically scattered electrons and lens aberrations of the microscope. Based on this, a complete description of the zero-loss contrast transfer function (CTF) is now possible.

We used tobacco mosaic virus (TMV), a biological sample known at atomic resolution, for definition of optimum CTF-parameters to reconstruct defocus series from an EFTEM LEO 912. CTF theory as known so far describes image contrast in the weak phase approximation as a linear sum of amplitude and phase contrast. The contribution of amplitude contrast (ratio of amplitude to phase contrast A/P) was determined to be between 7% and 5 % for unfiltered images and 12-14 % for zero-loss filtered images. However, in a filter microscope we remove electrons from the image, so we expect a higher amplitude contrast than in non-filtered images.

The Schelzer resolution is defined as the highest spatial frequency which is transferred into the image with the same phase as all lower frequencies. The resolution of the information limit is, however, determined by the information from the specimen which is equal to the degree of noise. The Schelzer resolution is determined by the wave length and the spherical aberration coefficient Cs of the objective lens. It reached 0.1 nm at 1250 kV. The limit of the resolution has been calculated numerically and it is written as d = 4.65(BsVr)−1/4 (nm), where Bs (in T) is the saturation magnetic flux density of the pole-piece material and Vr the relativistically corrected accelerating voltage. The resolution of the information limit is determined by the axial chromatic aberration coefficient Cc and incoherent effects such as the finite size of the source, beam divergence, energy spread, instabilities of the high voltage and lens current. The limit of the resolution is not clear. Most of the objective lenses of commercial microscopes are designed to optimize Cs rather than Cc. In this investigation, however, we describe the limit of Cc for 200 kV microscopes.