Energy Auger Electron Research Articles

Surface analytical studies of metal dusting of iron in CH4-H2-H2S mixtures

The metastable iron carbide, cementite (Fe 3 C), occurs as an intermediate phase during the high temperature corrosion process called metal dusting. The kinetics and thermodynamics of metal dusting of iron have been studied by Grabke et al. [1-5] using CO-H 2 -H 2 O and CH 4 -H 2 gas mixtures. H 2 S additions to carburising atmospheres impede the carbon transfer and retard the onset of metal dusting [6-14], thus allowing to study the early stages of the process. In this work the metal dusting process was studied in CH 4 -H 2 -H 2 S atmospheres at 700°C. Segregation experiments and surface analyses showed that S segregates on iron surfaces as well as on cementite surfaces. By means of Auger electron spectroscopy (AES), scanning Auger electron microscopy (SAM) and energy dispersive X-ray analysis (EDX) it was shown that coke contains graphite, cementite and iron particles with adsorbed sulphur.

Magnetic moments and anisotropies in ultrathin epitaxial Fe films on ZnSe(001)

The morphology, atomic magnetic moments, and in-plane magnetic anisotropies of ultrathin bcc Fe(001) films deposited by molecular beam epitaxy on ZnSe epilayers grown on GaAs(001) single crystal are reported. The growth mode and structure have been determined in situ by means of reflection high energy electron diffraction and Auger electron spectroscopy. The magnetic properties were characterized ex situ by an alternating gradient magnetometer, superconducting quantum interference device (SQUID) magnetometry, and conversion electron Mössbauer spectroscopy (CEMS). The Fe growth is epitaxial and occurs by three dimensional nucleation at the beginning. The coalescence of the islands is observed around 7 monolayers (ML). In agreement with SQUID results, CEMS measurements indicate no reduction of the Fe magnetic moment compared to the bulk value even for the first Fe monolayers. Determination of the in-plane anisotropy constants as function of the Fe thickness shows a strong interface-induced uniaxial in-plane magnetic anisotropy, which leads to a continuous evolution from a pure uniaxial anisotropy with easy axis along [110] direction for thickness below 10 ML to the pure bulk cubic Fe anisotropy above 40 ML.

Epitaxial growth and surface structure of cuprous halide thin films

The surface atomic and electronic structures of cuprous halide CuX (X=Cl, Br, and I) films, which were grown on various GaAs faces, have been studied by reflection high-energy electron diffraction, Auger electron spectroscopy, and electron energy loss spectroscopy. Epitaxial growth in a layer by layer fashion was observed commonly on all GaAs faces, except on a GaAs (111) As 2×2 face. The growth mode at the initial stage was affected by the lattice misfit of CuX to GaAs. Auger electron spectra and electron energy loss spectra of CuX depend on the crystal face of the substrate. These results indicate that CuX have both ionic and covalent features in the film growth process.

Effect of carbon coating on electron field emission from polysilicon

Carbon films deposited at relatively low temperatures were studied as a coating for polysilicon substrates to provide enhancement of the electron field emission. Undoped and phosphor doped polysilicon substrates were coated by carbon films grown by a very high frequency chemical vapor deposition process. Critical process parameters include substrate temperature of 225 °C, pressure of 56 mTorr, loaded power of 2 W/cm2, and discharge frequency of 56 MHz. The substrates were pretreated prior to film deposition, and bias enhanced nucleation was performed in hydrogen/C6H14 plasma. A carbon layer was grown to a thickness of 700 nm. Field emission measurements were performed in an ultrahigh vacuum chamber equipped with an electron gun, a carousel for five samples, and a Faraday cup. A multigrid quasispheric energy analyzer was used for Auger electron spectroscopy and electron energy loss spectroscopy characterization. Additionally, before and after field emission measurements, the secondary emission coefficient σ was measured by a scanning electron beam, which provided a two-dimensional σ map of the sample surface. The field emission current was measured in a diode configuration using a spacer thickness of 45–55 μm through a current collection window 2×5 mm2. Emission current–voltage curves were measured from three different regions of every sample by shifting the current collection window over the sample. A surface conditioning effect was observed, resulting in stabilization and better reproducibility of the emission current–voltage characteristics. The most significant changes in emission characteristics were observed in undoped polysilicon substrates with carbon coatings. Low deposition temperatures make this coating process very attractive because of its compatibility with addressing circuitry based on either silicon or poly-, microcrystalline, or amorphous silicon.

Determination of the inelastic mean free path of electrons in polyaniline samples by elastic peak electron spectroscopy

Application of x-ray photoelectron spectroscopy, Auger electron spectroscopy and electron energy loss spectroscopy for quantitative characterisation of surfaces requires knowledge of the electron inelastic mean free path and its energy dependence. Inelastic mean free path values have been determined for a wide variety of solids using theoretical and experimental methods. However, the number of published inelastic mean free path values for polymers is limited. In principle, these values for polymers can be estimated using predictive formulae such as TPP-2M and the G1 formula of Gries. An alternative method for inelastic mean free path determination in polymers is elastic peak electron spectroscopy. In the present work, the surface chemical composition of polyaniline samples (undoped and doped with palladium) was analysed quantitatively using x-ray photoelectron spectroscopy. The inelastic mean free path values were determined by elastic peak electron spectroscopy using a high-energy-resolution hemispherical analyser in the energy range 500– 5000 eV. The resulting inelastic mean free path values evaluated using Ni and Ag standards are smaller than the values resulting from the G1 predictive formula of Gries and the TPP-2M predictive formula of Tanuma et al. The discrepancies in the inelastic mean free path values are discussed extensively. Copyright © 2000 John Wiley & Sons, Ltd.

Defect analysis of Cl2/HBr/He/O2 etching process by imaging time-of-flight secondary ion mass spectrometry

Defects appeared during Cl2/HBr/He/O2 etching process were investigated by high laterally resolved imaging time-of-flight secondary ion mass spectrometry (TOF-SIMS). The defects with size of one to several microns appeared after the etching process. These defects are electron-beam sensitive during Auger electron spectroscopy and energy dispersive spectrometry analysis. Imaging TOF-SIMS clearly shows that the defect consists of bromine and H2O. Relatively weak oxygen and silicon signals were detected from the defect site. A negative secondary ion mass spectrum from region of interest (ROI) analysis of the defect area shows a very strong Br− signal. A positive secondary ion mass spectrum from ROI analysis has a very strong H2O+ signal from the defect. These results amply demonstrate that imaging TOF-SIMS enables chemical analysis of small defect at outmost surface with high lateral resolution and submonolayer sputtering damage under static SIMS analytical conditions.

The surface characterization of the early stages of the oxidation of titanium

In order to study the early stages of oxidation of pure titanium surface nanoscale characterisation has been carried out using different surface sensitive techniques. Metal surfaces have been prepared so as to study the effects of different surface finishes such as buffing (BP), chemical (CP), electrical (EP) and mechano-chemical (MCP) treatments. Characterisation of the oxide layers has been carried out using scanning electron microscopy (SEM), Auger electron spectroscopy (AES) and X-ray photoelectron spectroscopy (XPS). These measurements have shown that the BP and MCP surfaces have thick modified layers whereas only very thin oxide layers are formed on the CP and EP surfaces. The influence of hydrogen on the titanium oxidation during the surface treatments has been estimated by obtaining the depth profiles of hydrogen, as measured using elastic recoil detection analysis (ERDA). High concentrations of hydrogen have been found in the surface regions especially for MCP and BP treatments. An α-titanium single crystal has been used to investigate the initial stages of the oxidation processes using grazing incidence reflection high energy electron diffraction and Auger electron spectroscopy (RHEED-AES). This method has demonstrated the time dependence of the titanium oxidation during the early stages with the RHEED intensities from the α-titanium single crystal surface, TiO being observed in the oxide layer. We discuss the characteristics of the oxidised surface on a sub-micro scale in relation to parameters such as the thickness of the oxide layer, crystalline state, chemical state, compactness of the oxide layer and hydrides of titanium.

Depth resolution in sputter depth profiling-characterization of a third batch of tantalum pentoxide on tantalum certified reference material by AES and SIMS

Depth profiles have been made for a new batch of the certified reference material, BCR 261, of ∼30 nm and 100 nm of anodic tantalum pentoxide layers on tantalum foil. Atomic force microscopy studies show that the preparation method traditionally used provides an excellent substrate root-mean-square (rms) smoothness of 1.5 A. This goes part of the way to explain the exceptionally good interface resolutions readily obtained with this material using AES. Further AES measurements, in the pulse counting mode, show that the same excellent depth resolution is obtained as previously (1.6 nm at 100 nm depth). Interestingly, the background of inelastic primary electrons under the oxygen peak increased with a characteristic length of only 4.0 nm at the interface for a 5 keV electron beam. Measurement of the ion-induced emission at the oxygen Auger electron energy, or in a broad range around this energy, shows that this signal can be used to monitor the ion beam stability and the presence of the oxide without the electron beam, although the signal levels are smaller than for AES. This emission is attributed to O− ions being detected through the electron spectrometer. Measurements using time-of-flight SIMS with a 12 keV Ga+ beam for analysis and 2 keV Ar+ or SF5+ beams for depth profiling show that the material may be used to characterize aspects of SIMS instrumental performance, in addition to AES and XPS, at a very high level of quality. The optimum interface resolutions obtained are better than 1.0 nm, with trailing edge decay lengths down to 0.33 nm. The use of molecular ions gives excellent depth resolution, comparable to the 0.8 nm depth resolution of AES at its best. Copyright © 2000 John Wiley & Sons, Ltd.

BREAKDOWN OF THE FORWARD SCATTERING MODEL IN MgO(001)

Typically, enhanced intensities are observed along low-index crystallographic directions of the crystal when Auger or photoelectrons with a kinetic energy of several hundred eV or more are emitted from atoms in a solid. This effect is a result of strong forward scattering of emitted electrons by overlying lattice atoms. Contrary to this image, a strong dip along the [110] direction has been found in the angular distribution curve of O-KLL Auger electrons of the MgO(001) surface, even though the kinetic energy of O-KLL Auger electrons is about 500 eV. The angular dependence of Auger (O-KLL, Mg-KLL) and photoelectrons (O1s) in MgO(001) is discussed and explained within a multiple scattering cluster model, respectively.

Chromium interaction with TiO2 dispersoids in oxide dispersion strengthened ferritic steel

Auger electron spectroscopy and energy filtering TEM were used to study alloying element distributions in the vicinity of matrix–TiO2 dispersoid interphase boundaries in an Fe–Cr ferritic steel strengthened by dispersed TiO2 particles. Pronounced interfacial Cr segregation was found to occur in the course of annealing at 700°C. The Cr segregation also leads to Cr diffusion into the dispersoids and to the formation of oxide phases with different Cr/Ti ratios. The observed Cr segregation is described in terms of diffusion into the field of elastic stresses.

Tribological properties of automotive disc brakes with solid lubricants

In automotive brake systems, high temperatures and pressures are generated at the contacting surfaces. This affects the pad and disc materials, creating a friction film between the surfaces, which consists of wear particles and volatile reactants from the pad and disc. To acquire suitable tribological properties, a pad matrix contains up to 20 different ingredients, mainly selected from experience. In the present report the effect of solid lubricants has been studied in three different pad matrices with relatively few components as compared to commercial brake pads. The components are Cu 2S, PbS and Sb 2S 3, which are known to modify and stabilise the friction coefficient. The friction coefficient and wear rates of the pads are examined on a dynamometer, which simulates series of real-life car brake events. Two different energy levels and two ambient temperatures are included. After these tests, the brake disc surfaces are analysed with energy dispersive X-ray (EDX) and Auger electron spectroscopy (AES) in combination with argon ion sputtering to study the microscopic lateral and in-depth distribution of elements on and below the surface, looking for traces of the friction film. These experiments are used to discuss the correlation between the tribological properties and the external variables — braking temperature, solid lubricants and pad matrix.

Excess silicon at the silicon nitride/thermal oxide interface in oxide–nitride–oxide structures

The chemical composition and structure of Si3N4/thermal (native and wet) SiO2 interface in oxide–nitride–oxide structures are studied by using secondary ion mass spectroscopy, electron energy loss spectroscopy (EELS) and Auger electron spectroscopy (AES) measurements. EELS and AES experiments show the existence of excess silicon at the Si3N4/thermal SiO2 interface. Excess silicon (Si–Si bonds) at Si3N4/SiO2 interface exists in the form of Si-rich silicon oxynitride. Numerical simulation of the Si–Si bond’s electronic structure by using semiempirical quantum-chemical method (MINDO/3) shows that Si–Si defects act as either electron or hole traps. This result explains the abnormally large electron and hole capturing at this interface reported earlier.

Surface analytical investigation of the electrochemical and corrosion behaviour of nanocrystalline FeAl8

From the higher fraction of grain boundaries in nanocrystalline substances a different corrosion behaviour in comparison to the conventional polycrystalline material can be expected, which may be utilised for the development of new corrosion resistant alloys. Therefore, the oxidation behaviour of these two different crystallisation states of FeAl8 was compared by means of electrochemical and surface analytical experiments. The oxide films formed after electrochemical passivation were investigated by Auger Electron Spectroscopy. The application of inelastic peak shape analysis by the method of Tougaard showed, that for both materials the oxide layer may be described by a model of a (below the contamination) buried layer with a thickness of only a few nanometers depending on the preparation conditions. Factor Analysis was applied for the evaluation of the differentiated low energy Auger electron spectra (20–100 eV) as a function of depth profiling sputtering time. For both, the nanocrystalline and the polycrystalline material, the inner part of the oxide layer was enriched in Al, whereas the very outer part (surface region) was enriched in Fe. No differences concerning the sputtering time for removal of the oxide layers were found for the two alloys.

The decomposition of nitrobenzene on a Cu(110) surface

The decomposition of nitrobenzene on Cu(110) has been studied using Auger and high resolution electron energy loss spectroscopies. The Auger spectrum of the decomposed species indicates the presence of carbon and oxygen but no nitrogen. A direct comparison of the energy loss spectrum to that of phenol on Cu(110) excludes the decomposition of nitrobenzene to phenoxy. The difference between the on-specular HREELS spectra along the 〈001〉 and 〈110〉 azimuths and the appearance of the B 1 mode in RAIRS indicates a C 1 symmetry for the surface species with the molecular plane aligned close to the 〈110〉 azimuth. From ab initio calculations, typical non-crossing potential energy surfaces have been found for decomposition of the nitrobenzene adsorbed on a Cu cluster. By increasing the size of the Cu clusters, the repulsive state is effectively stablised, the intersection distance between this PES and the attractive ground state PES is shortened and the dissociation barrier is reduced.

Electrodeposition of CuIn xGa 1− xSe 2 thin films

CuIn x Ga 1− x Se 2 (CIGS) polycrystalline thin films with various Ga to In ratios were grown using a new two-step electrodeposition process. This process involves the electrodeposition of a Cu–Ga precursor film onto a molybdenum substrate, followed by the electrodeposition of a Cu–In–Se thin film. The resulting CuGa/CuInSe bilayer is then annealed at 600°C for 60 min in flowing Argon to form a CIGS thin film. The individual precursor films and subsequent CIGS films were characterized by X-ray diffraction, scanning electron microscopy, energy dispersive spectroscopy and Auger electron spectroscopy. The as-deposited precursor films were found to be crystalline with a crystal structure matching that of CuGa 2. The annealed bi-layers were found to have the same basic chalcopyrite structure of CuInSe 2, but with peak shifts due to the Ga incorporation. Energy dispersive spectroscopy results show that the observed shifts correlate to the composition of the films.

In situ EC-AFM observation of Cd electrodeposition on Au(100)

Nowadays, epitaxial growth is one of the most important technologies to obtain uniform and defect-free thin films, which are of utmost importance in the fabrication of certain devices. For this process, the selection of the substrate on which crystal grows is of major importance. Since fundamental information about Cd electrodeposition on a single crystal are needed to form CdX (X 5 S, Se, Te) compound semiconductor thin films, there are many studies of cadmium electrodeposition on various substrates, e.g. Bi, Sn, Pb (1), Ag (2,3,7–9), Au (3,4,7,10), Pt (5,10), Cu (3,10), Ti (6), which were performed using different surface (structure) sensitive methods such as electrochemical techniques, optical secondharmonic generation (SHG), low energy electron diffraction (LEED) and Auger electron spectroscopy (AES). However, scanning probe microscope images of underpotentially deposited (UPD) Cd are restricted to work examining Cd on Au(111) (11) and onto UPD adlayers of S, Se and Te (12–14). The results presented here are part of a research project devoted to the study of Cd electrodeposition on Au(hkl) and include in situ electrochemical atomic force microscopy (EC-AFM) observations of the atomic structures and the growth morphology of electrodeposited Cd films on Au(100) electrode in H2SO4 electrolyte.

Generation and Reactions of CH2 and C2H5 Species on Mo2C/Mo(111) Surface

The adsorption and dissociation of CH2I2 and C2H5I on Mo2C/Mo(111) surface have been investigated with the purpose of producing adsorbed CH2 and C2H5 species. Methods used include high resolution electron energy loss, X-ray photoelectron, Auger electron, and temperature programmed desorption spectroscopies. Independently of the coverage, CH2I2 adsorbs molecularly at 90–100 K. The dissociation of an adsorbed layer starts around 180–190 K. The primary products of thermal dissociation are adsorbed CH2 and I. The species CH2 undergoes self-hydrogenation to CH4 at Tp=300 K and dimerization into C2H4 at and above 222–280 K. Ethylene formed desorbs above 400 K. C2H5I also adsorbs molecularly on Mo2C at 90–100 K and dissociates to C2H5 and I above 150 K. The reaction of C2H5 on Mo2C/Mo(111) surface yielding C2H6 and C2H4 proceeds at a much lower temperature, above 180 K, than that of CH2. Neither the cleavage of the C–C bond nor the coupling of C2 compounds occurred to detectable extent under the reaction conditions. The ethylene formed in the reactions of both CHx species exhibited the same features as observed following C2H4 adsorption on Mo2C: the stable di-σ-bonded ethylene is transformed into ethylidyne at higher temperature. The results are discussed in relevance to the conversion of methane into benzene on Mo2C deposited on ZSM-5.

Electronic and vibrational electron energy loss studies of the formation of SiO 2 and carbonaceous layers on Si(111)7×7 by thermal and ion-assisted deposition techniques

Electronic and vibrational electron energy loss spectroscopy (EELS) and Auger electron spectroscopy (AES) were used to characterize the formation of SiO 2 and SiC on Si(111)7×7. The differences between thermal deposition and ion-assisted deposition of these films were investigated. Thermally formed SiO 2 layers obtained by annealing the silicon sample in O 2 at 700 K can be characterized by the electronic transitions near 5 and 7 eV and by the blue shifts in its vibrational features from that of atomic oxygen adsorption. The formation of SiO 2 could also be achieved by ion irradiation at room temperature (RT). Carbonaceous layers formed at different ethylene dosages and sample temperatures were characterized by EELS and AES. Like the oxide formation, a higher efficiency in producing the carbonaceous layer was obtained by using ion-assisted deposition than thermal deposition. In particular, a SiC layer could be obtained by cycles of ethylene ion irradiation at RT and annealing to 800 K, whereas a very large dosage of ethylene ions at RT was found to produce a graphite layer.

Stable surface reconstructions on 6H–SiC(000 [formula omitted

We used scanning tunneling microscopy (STM), low energy electron diffraction (LEED) and Auger electron spectroscopy (AES) to identify and investigate stable surface reconstructions on 6H–SiC(000 1 ̄ ). Starting from ex situ prepared (1×1) ordered surfaces heating of the sample in ultra-high vacuum leads to (3×3), (2×2) C and (1×1) graphitic phases with increasing temperature. The stoichiometry of the phases was determined by AES by comparison to the initial, bulk like (1×1) phase: the (3×3) structure is carbon enriched, while the (2×2) C structure reaches the bulk like stoichiometry again. The two superstructures were investigated in detail by STM revealing significant differences in structural complexity. Additional Si flux during heating produces a silicon enriched surface phase, which also exhibits (2×2) periodicity. However, this new phase, denoted as (2×2) Si, has a different surface structure than the (2×2) C phase obtained by mere heating. Apart from the different stoichiometry this is revealed by significantly different LEED intensities.

Adsorption and desorption of butanethiol on Au{100}-(5×20)

The adsorption and desorption of butanethiol in ultra-high vacuum on Au{100}-(5×20) has been investigated using temperature programmed desorption/reaction (TPD/R), low energy electron diffraction and Auger electron spectroscopy. The TPD/R results clearly show that butanethiol adsorbs into both a chemisorbed and a physisorbed state. Desorption of physisorbed butanethiol occurs molecularly at ∼138 K. By contrast, desorption of the chemisorbed butanethiolate species occurs with decomposition at ∼500 K to yield primarily 1-butene; the thiol sulfur remains adsorbed on the surface and either desorbs or possibly dissolves into the bulk of the gold sample at above 700 K. The substrate temperature dependence of the chemisorption process suggests a precursor mechanism for the chemisorption kinetics. The TPD/R results also show that chemisorption does not occur on a very clean and ordered Au{100}-(5×20) surface at 100 K, and that low coverages of pre-adsorbed sulfur atoms facilitate the chemisorption process, suggesting a defect-mediated precursor mechanism.