Inelastic electron scattering observation using energy filtered transmission electron microscopy for silicon -- germanium nanostructures imaging

This paper presents a new technique using energy filtered TEM (EFTEM) for inelastic electron scattering contrast imaging of Germanium distribution in Si-SiGe nanostructures. Comparing electron energy loss spectra (EELS) obtained in both SiGe and Si single crystals, we found a spectrum area strongly sensitive to the presence of Ge in the range [50-100 eV]. In this energy loss window, EELS spectrum shows a smooth steeply shaped background strongly depending on Ge concentration. Germanium mapping inside SiGe can thus be performed through imaging of the EELS background slope variation, obtained by processing the ratio of two energy filtered TEM images, respectively, acquired at 90 and 60 eV. This technique gives contrasted images strongly similar to those obtained using STEM Z-contrast, but presenting some advantages: elastic interaction (diffraction) is eliminated, and contrast is insensitive to polycrystalline grains orientation or specimen thickness. Moreover, since the extracted signal is a spectral signature (inelastic energy loss) we demonstrate that it can be used for observation and quantification of Ge concentration depth profile of SiGe buried layers.     

The effect of modulation amplitude on electron-excited auger data from titanium

The effects of the distortion due to large modulation amplitudes on the electron-excited Auger spectra of titanium have been studied using a cylindrical mirror analyzer. It is found that the normal derivative Auger data are very sensitive, both in magnitude and shape, to the value of the amplitude. The peak-to-peak signal strengths behave in a very complex manner and there is no simple way of relating the various values. Since the background is essentially eliminated in the derivative data, the Auger spectrum, itself, is obtained in the electron energy distribution by integrating the differential spectrum. It is shown that the magnitude and shape of these spectra are better behaved as a function of the modulation amplitude. The strengths of these features vary almost linearly with amplitude and the effect of distortion is simply to broaden the features. Auger area values obtained by double integration over the Auger structure in the derivative spectrum are shown to follow the predicted linear relation with the modulation amplitude. This simple relationship allows relative elemental surface concentration measurements to be made with enormous improvements in signal-to-noise ratio while maintaining the ability to exactly correct for potential modulation distortion.     

Application of auger ewlectron depth profile analysis to thin film interdiffusion studies

Because of its good depth resolution, the Auger electron depth profile analysis allows to investigate diffusion phenomena in thin films directly. Complicated calibration procedures, however, are needed to correct for the matrix effects inherent in the Auger method, particularly artefacts due to the sputtering process. In this paper, two types of thin-film systems are presented in order to determine diffusion coefficients from depth profiles: double-layer and periodic multi-layer film structures. Compared to the double-layer films, the multi-layer structure has the advantage of less stringent requirements on depth resolution and allows to detect smaller diffusion effects. Finally, it is shown how grain boundary and bulk diffusion data can be extracted separately from the composition profiles.     

Special techniques for the Auger analysis of microelectronic devices

Microelectronic devices are becoming more complex and device features are getting smaller as the level of integration continues to increase. Although scanning Auger microscopy has been applied extensively to the analysis of microelectronic devices with a great deal of success, the analysis of current and future devices is presenting new challenges. The major limitations are (1) features of interest in microelectronic circuits are often comparable in size to the beam diameter of commercial Auger microprobes, and (2) the electron beam tends to drift about on the specimen surface because of mechanical instability and differential thermal expansion of the apparatus. In this paper, we present two different techniques developed to overcome these limitations. In the specimen modulation technique, the modulating signal is applied to the electrically isolatable regions of a device instead of to the electron energy analyzer. This method of modulation permits the detection of only the Auger electrons that are emitted from the modulated region. Spurious contributions from adjacent areas inadvertently illuminated by the analyzing beam are suppressed. In the position modulation technique, the analyzing beam is scanned repetitively across the feature to be analyzed and the Auger signal is synchronously detected at the scan frequency. The resulting Auger signal magnitude is shown to be unaffected by beam drift. This method of signal detection eliminates the error and uncertainty caused by beam instability during long-term depth profiling, but is applicable only to specimens with certain geometries.     

SEM investigation of the dependence of magnetic domain structure on the thickness of cobalt monocrystals

The magnetic domain structure of cobalt monocrystal is observed by means of a scanning electron microscope (SEM). It is revealed by the so-called type-I magnetic contrast [1]. The dependence of magnetic domain width on the specimen is thickness is investigated and discussed. Digital image processing (image restoration, enhancement and analysis) is used on the images obtained directly from the SEM. The main reasons for the application of digital image processing are: poor resolution of type-I magnetic contrast due to the diffuseness of the leakage magnetic fields above the specimen surface, and complex character of magnetic domains. The resolution limit of type-I magnetic contrast in cobalt monocrystal is evaluated. Statistical distributions of magnetic domain width are also calculated and presented.     

Investigation of the compositional changes of a Y-Ba-Cu-O HTSC target under ion sputtering

An Auger electron spectroscopy study is reported of the elemental depth profile of Y-Ba-Cu-O HTSC targets subjected to ion-plasma sputtering in a magnetron deposition system and ion-beam sputtering in the Auger spectrometer chamber. It has been established that the process consists in all cases of predominant copper sputtering accompanied by the formation of a modified surface layer and of a copper-depleted region. This region is assumed to originate from intense copper diffusion from the bulk to the modified surface layer driven by a concentration gradient.     

Luminance contrast, a new illumination technique in light microscopy: Optical basics, practical evaluations, further developments

Luminance contrast is a new illumination technique in light microscopy recently developed by the author, which leads to extraordinary contrast effects, enhanced focal depth and supramicroscopic resolution when transparent stained or unstained specimens are examined.This method is characterized by a small illuminating light beam running centrically to the specimen, so that the specimen is illuminated very homogenously. Within the objective, the illuminating beam is blocked by a small light stop. When all parts of the illuminating light are totally blocked, the background is completely dark; when small parts of the illuminating light pass the light stop, the background is moderately brightened.The microscopic image is a result of scattered light components, which are bent or reflected by the specimen. When the illuminating light is not completely blocked by the light stop, the transmitted illuminating components interfere with the scattered light components coming from the specimen.Additional three-dimensional effects are achievable when the illuminating light passes the specimen obliquely.The resulting variants of illumination are similar to dark field, negative phase contrast and interference contrast. In all modes of luminance contrast the specimen is illuminated very homogenously so that it appears as a self-luminous, fluorescent body – similar to fluorescence microscopy.Luminance contrast can be carried out when mirror objectives are used or common objectives containing glass lenses are equipped with individually adjusted light stops.Further aspects of technical developments are discussed in full details.     

Enhancement of the ESDIAD method for imaging bond directionality in chemisorbed species

The ESDIAD method for imaging adsorbate bond directions by photographic observation of positive ion angular distributions during electron stimulated desorption suffers from inherent low contrast due to background effects. The use of a digital acquisition system designed to overcome this difficulty in ESDIAD measurement is presented. Measurements on a Ni(110) single crystal substrate show the presence of a significant background signal due to soft X-ray generation by electron impact. By subtraction of the background signal, a significant enhancement of positive ion signal-to-noise ratio is achieved in ESDIAD, converting the ESDIAD method into a high contrast, high resolution surface measurement technique. Quantitative studies of the soft X-ray background have shown it to be linearly dependent on electron current density and electron energy, with no change in angular shape. These properties permit an accurate background subtraction procedure to be employed to significant;y enhance the capability of the ESDIAD method.     

SEM investigation of the temperature dependence of magnetic domain structure of cobalt monocrystals

The magnetic domain structure of a cobalt monocrystal is observed by means of a scanning electron microscope (SEM). It is revealed by the so-called type-I magnetic contrast [1]. The dependence of the magnetic domain structure on temperature up to about 700 K is investigated and discussed. Digital image processing (image restoration, enhancement and analysis) is used on the images obtained directly from the SEM. The main reasons for the application of digital image processing are: poor resolution of type-I magnetic contrast due to the diffuseness of the leakage magnetic fields above the specimen surface, and the complex character of the magnetic domain structure. Statistical distributions of magnetic domain width are also calculated and presented.     

Photoelectron microscopy — Applications to biological surfaces

Photoelectron microscopy utilizes incident light to eject electrons which are then accelerated and used to form an image of the specimen surface. Photoelectron microscopy is now being developed to the point where it can begin to contribute to the analysis of biological structures including cell surfaces and the organization of cellular organelles. PEM is the electron optics analog of fluorescence microscopy and shares with it many advantages and areas of application. As in fluorescence microscopy, the strategy is not an elemental analysis of components present since proteins and other functional macromolecules usually do not differ significantly in elemental composition, but rather the combination with site-specific antibodies to identify the organization and structure-function relationships in cells. Material contrast is based on photoelectron yield, which depends on the valence electron properties of the surface materials rather than on atomic number differences, as with most surface analytical techniques. Topographic contrast makes very small objects visible on relatively flat surfaces because the low energy electrons emerging from the specimen surface are easily deflected by the accompanying field variations. The surface sensitivity is also high because of the short escape depths of the ejected photoelectrons. These and other characteristics combined with a lateral resolution of about 10 nm makes photoelectron microscopy a promising technique for studies of organic and biological surfaces.     

Laser-Induced Fluorescence Bronchoscopy for Detection and Localization of Early Lung Cancer

Experimenatal results on the development of a Laser-Induced Fluorescence Bronchoscopy(LIFB) for the detection and localization of early lung cancer are reported in this paper. The system utilizes fluorescence of photosensitizer drug to provide real time video imaging for the examined lung tissue. Color filters are used to differentiate signal from background and a computer image processing technique is also applied to subtract the background. Moreover, a pseudocolor contrast enhancement method was developed to enhance the fluorescence image displayed on the vidio monitor. Suspicious areas are identified by pseudocolor image to guide biopsy, and several clinical trials show that sensitivity and contrast capability of the system should permit the detection and localization of early lung cancer.     

Laser-Induced Fluorescence Bronchoscopy for Detection and Localization of Early Lung Cancer

Experimenatal results on the development of a Laser-Induced Fluorescence Bronchoscopy(LIFB) for the detection and localization of early lung cancer are reported in this paper. The system utilizes fluorescence of photosensitizer drug to provide real time video imaging for the examined lung tissue. Color filters are used to differentiate signal from background and a computer image processing technique is also applied to subtract the background. Moreover, a pseudocolor contrast enhancement method was developed to enhance the fluorescence image displayed on the vidio monitor. Suspicious areas are identified by pseudocolor image to guide biopsy, and several clinical trials show that sensitivity and contrast capability of the system should permit the detection and localization of early lung cancer.     

Analysis of material modifications caused by nanosecond pulsed UV laser processing of SiC and GaN

The effects of direct UV laser processing on single crystal SiC in ambient air were investigated by cross-sectional transmission electron microscopy, Auger electron spectroscopy, and measurements of the electrical resistance using the transfer length method (TLM). Scanning electron microscopy was applied to study the morphology and dimensions of the laser-treated regions. After laser processing using a nanosecond pulsed solid-state laser the debris consisting of silicon oxide was removed by etching in buffered hydrofluoric acid. A layer of resolidified material remains at the surface indicating the thermal impact of the laser process. The Si/C ratio is significantly disturbed at the surface of the resolidified layer and approaches unity in a depth of several tens of nanometers. A privileged oxidation of carbon leaves elementary resolidified silicon at the surface, where nanocrystalline silicon was detected. Oxygen and nitrogen were detected near the surface down to a depth of some tens of nanometers. A conductive surface film is formed, which is attributed to the thermal impact causing the formation of the silicon–rich surface layer and the incorporation of nitrogen as dopant. No indications for microcrack or defect formation were found beneath the layer of resolidified material.     

Semi-empirical elemental sensitivities for quantitative auger electron spectroscopic analysis

A semi-empirical equation is presented which may be used to calculate the relative sensitivities of the elements for Auger electron spectroscopic analysis. The equation involves the product of a large number of theoretically or empirically derived terms, including ionization cross section. Coster—Kronig transitions, backscattered electron contribution, individual Auger transition probabilities, atomic density, electron mean free path, analyser transmission, modulation effects, detector efficiency, and peak asymmetry. The calculated sensitivities are compared with experimental relative elemental sensitivities from the Handbook of Auger Electron Spectroscopy. The close agreement obtained with experiment suggests that the semi-empirical theory is now capable of providing elemental sensitivities for analysis, in a predictive way, where reliable empirical sensitivities are missing.     

CHARACTERIZATION OF POLYCRYSTALLINE GRADIENT THIN FILM BY X-RAY DIFFRACTION METHOD

A direct method is proposed to quantitatively characterize the structural depth profiles emerged in the polycrystalline thin films based on the information obtained by X-ray diffraction (XRD) with various incident angles and treated by a numerical procedure known as the constrained linear inversion. It should be noted that the proposed method was neither sensitive to the random noise appearing in experiment nor to the error originated from the measured thickness of the specimen. To testify the validity of the method, XRD measurements were carried out on a specially designed Pd/Ag bilayer sample, which was annealed at 490℃ for 20 min, and the depth profiles were accordingly calculated through resolving the obtained XRD patterns. The elemental concentration depth profile of the Pd/Ag bilayer sample was in turn calculated from the resolved patterns, which was in good agreement with those obtained by Auger electron analysis on the annealed sample.     

Comparison of detectability limits for elemental mapping by EF-TEM and STEM-XEDS

The analytical sensitivity in terms of the signal-to-noise ratio (SNR) was investigated for elemental mapping by a transmission electron microscope equipped with an energy filter (EF-TEM) and a scanning transmission electron microscope with an X-ray energy dispersive spectrometer (STEM-XEDS). To compare the detectability limits of the elemental maps by the two techniques, homogeneous Cu-0.98+/-0.34 wt% Mn and Cu-4.93+/-0.49 wt% Mn thin specimens were used. Elemental maps can be considered as either an image or a spectrum. Therefore, the detectability limits of the elemental maps were characterized by the spectral SNR. To evaluate the detectability limits of the elemental maps with statistical confidence limits such as 1 sigma, 2 sigma and 3 sigma, the SNR values were reviewed from the statistical point of view. In STEM-XEDS mapping, the spectral SNR values improve as the specimen thickness increases since the signal intensity increases. Conversely, the spectral SNR in EF-TEM mapping is maximized at a certain thickness and then reduces as the thickness increases. To compare the two mapping techniques with regard to the analytical sensitivity, a method to estimate the minimum mass fraction (MMF) from measured signal and background intensities was developed. In this experimental approach, the MMF value can be evaluated by selecting the appropriate SNR value corresponding to the statistical confidence limits. In comparing the estimated MMF values from the two mapping approaches, EF-TEM mapping can be more sensitive than STEM-XEDS mapping up to specimen thicknesses <20-30 nm in the 1 sigma confidence limit and < approximately 50 nm in the 3 sigma limits. However, as the specimen thickness increases, the XEDS maps provide better detectability limits in the Cu-Mn dilute alloy specimens.     

Application of factor analysis to elemental detection limits in sputter depth profiling

Factor analysis of Auger spectra acquired during sputter depth profiling is superior to the conventional peak-to-peak amplitude method for determining elemental compositions, especially when the Auger signal strength is near the detection limit. The reason for the improvement is that factor analysis utilizes information from all the data channels in the Auger spectrum while the peak-to-peak amplitude method uses information from only two data channels. In addition, factor analysis can separate much of the spectral noise from the signal during processing, while peak-to-peak amplitude additively measures signal plus the range of the spectral noise. Procedures can also easily account for interfering species, even when their spectra are not known. In one example, at least a factor of five improvement in the minimum detection limit was achieved. In application to secondary ion mass spectrometry, only a marginal improvement in detection and precision was achieved. This is because our existing procedure (peak area measurement) already utilizes spectral information content fairly efficiently. However, factor analysis is capable of handling spectral interferences that the peak area method cannot.     

DSP图像处理技术在空间瞬态光辐射定位探测中的应用

在分析空间瞬态光辐射信号成像特点的基础上,提出了一种在空间瞬态光辐射信号定位系统中,用高速浮点DSP芯片实现图像处理的方案.图像处理过程为：首先通过阈值分割滤除背景和弱噪音;然后根据目标信号的时空连续性进行目标识别以去除强噪音;最后对识别出的目标像点进行重心位置的提取.实验表明,该硬件图像处理方案具有高的实时处理性和定位准确性,可以很好的满足系统要求.     

Influence of the matrix on boron detection by auger electron spectroscopy (AES)

The difficulties of boron detection by AES in catalysts containing low concentrations of boron can be partially associated with elastic scattering of Auger electrons. In the present work, the Monte Carlo approach is used to estimate the effects of Auger electron collisions in different matrices. Elastic scattering events are described within the partial wave expansion method. The scattering centres are approximated by the Thomas-Fermi-Dirac potentials. It is found that elastic collisions decrease the boron KLL Auger electron signal by up to 10%. In addition, the sampling depth of AES is strongly decreased. This decrease reaches 35% for medium atomic number elements. The values of parameters correcting for elastic scattering effects are determined for boron embedded in 14 elements. An attempt is also made to interpolate the correcting parameters for matrices not considered in the present work.     

Quantitative analysis of iron nitrides (γ′-Fe4 and ϵ-Fe2−3N) using auger electron spectroscopy

The investigation and development of tribologically important transition metal nitride surface layers rely on the accurate quantitative analysis of light elements used in conjunction with a suitable depth profiling technique. This paper describes the quantitative AES analysis of iron nitrides (γ′-Fe₄N and ϵ-Fe₂₋₃N) by comparison with reference materials and by applying correction factors for ionization by backscattered primary electrons, atom number density and inelastic mean free paths of the Auger electrons in the matrix. It is shown that the nitrogen contents of these iron nitrides may be determined to within 10% accuracy, and it is also shown that this accuracy depends firstly on the correct calculation of the abovementioned correction factors, and secondly on the correct determination of the Auger electron yield. The Auger electron yield should be determined by measuring the area below the Auger peak on the N(E)E spectrum, after removal of the background electron spectrum to ensure that details of the relatively small nitrogen peaks are not overlooked.