Monochromator versus deconvolution for XPS studies using a kratos ES300 system

Deconvolution has been used by several workers to reduce or eliminate X-ray and analyzer-induced broadenings of XPS spectra, but there is a 30% limit on the amount of improvement in the FWHM of XPS peaks. Deconvolution algorithms generally used also increase the noise in the spectrum. Monochromatic X-ray sources can give high resolution data directly, but the count rates are generally low. Given these limitations, a comparison has been made to see which method of high resolution data acquisition. can achieve data of a certain quality in the least amount of time, for a given system, namely a Kratos ES300 photoelectron spectrometer. Signal-to-noise ratios are used as a means of comparing spectral quality at a given resolution. Results for the Kratos ES300 show that if the desired resolution could be achieved without deconvolution, an order of magnitude less counting time was needed for the Mg Kα standard source. If deconvolution is necessary, the counting times for both sources to achieve a given S/N ratio were found to depend mainly upon the signal-to-background ratios of the peak measured with each source. These results are illustrated for the S 2p doublet from MoS₂ and the Ni 2p peaks from oxidized stainless steel.     

Surface contributions to the XPS spectra of nanostructured NiO deposited on HOPG

In this work we present an in situ X-ray photoelectron spectroscopy (XPS) study of the growth of NiO on highly ordered pyrolitic graphite (HOPG). The XPS spectra were measured as a function of the equivalent NiO coverage. Also, ex-situ atomic force microscopy (AFM) images were taken for some of these stages in order to follow the morphology of the NiO deposits. For low coverages the lineshapes of the Ni 2p spectra differ strongly from those of bulk NiO. This has been related to the large surface contribution. The O 1s XPS spectra also show a surface related structure which follows the same trend observed in the Ni 2p spectra.     

The application of deconvolution methods in electron spectroscopy — a review

In this review of the application of deconvolution in electron spectroscopy the operations of convolution and deconvolution are first defined. The theoretical background to direct methods, Fourier Transform techniques and iterative approaches to deconvolution is given. The use of smoothing techniques, the removal of secondary electron background and the determination of broadening functions are then considered. Examples of deconvolution in electron spectroscopy are discussed. These include valence and core levels in XPS, gas-phase UPS studies and resolution enhancement in Auger spectroscopy. Deconvolution of Auger spectra to yield density of states curves is examined. The likely future status of deconvolution in electron spectroscopy is considered.     

Iterative deconvolution of X-ray photoelectron spectra generated by an achromatic source

Three cases are presented where X-ray photoelectron spectra generated by an achromatic X-ray source have been deconvoluted by the iterative, Van Cittert method. These are (i the gold 4 f doublet, (ii) the gold valence band, and (iii) the aluminium 2 p spectrum of an oxidized aluminium foil. All prominent features in the deconvolute are shown to be real by comparison with high resolution studies, and the noise levels are low. Resolution after deconvolution is similar to that obtained with a monochromatized X-ray source. The spectra are also similar to those obtained by Fourier transform deconvolution.     

Comparison of site-specific valence band densities of states determined from Auger spectra and XPS-determined valence band spectra in GeS (001) and GeSe (001)

Auger lineshapes of the Ge M₁M_4,5V and M₃M_4,5V and Se _M1M_4,5V transitions in GeS (001) and GeSe (001) are measured and compared to XPS valence band spectra. Distortions in both types of spectra due to inelastic scattering, analyzer and source broadening, and core level lifetime broadening are removed by deconvolution techniques. The valence band consists of three main peaks at ?2 eV, ?8 eV, and ?13 eV. There is excellent agreement of peak positions in AES and XPS spectra. The Auger lineshapes can be interpreted in terms of site-specific densities of states. They indicate that the states at ～?8 eV and at ～?13 eV are associated with the cation and anion sites respectively. The bonding p-like states at the top of the valence band have both cation and anion character. The Auger lineshapes indicate that the states closest to the valence band maximum are preferentially associated with Ge.     

The XPS spectra of the metathesis catalyst tungsten oxide on silica gel

XPS spectra of supported and unsupported tungsten oxides before and after use in the metathesis reaction of propene are reported. It is shown that the broad peaks, usually measured for supported materials, are due to line broadening caused by differential charging. A new sample preparation technique is described which leads to a considerable reduction of the line broadening. Both XPS and X-ray diffraction show that in fresh catalysts a well-defined phase Of WO₃ is present. The reduction of the supported trioxide by propene to W₂₀O₅₈, as concluded from X-ray diffraction measurements, results in the formation of W^v - and/or W^IV -species in used catalysts. For unsupported WO₃ these lower valencies have been observed in the valence-band spectra.     

The effect of vacancies on the electronic structure of Nbo

The crystal structure of NbO is considered as a NaCl structure with 25% ordered vacancies in each sublattice. Self-consistent energy band structure calculations using the augmented plane wave (APW) method have been carried out for a hypothetical NbO with the full NaCl structure and for the structure with the vacancies. We find that the introduction of the vacancies leads to (1) a lowering of the Nb-d like bands with respect to the O-p bands, (2) an almost completely filled “vacancy band”, and (3) a splitting and broadening of the O-p like band. These effects are shown to have a significant influence on both the theoretically calculated X-ray photoemission spectra (XPS) and the X-ray emission spectra (XES). All theoretical spectra obtained with the vacancy-structure are found to be in good agreement with experiments in contrast to those spectra calculated for hypothetical NbO in the NaCl structure.     

Layer structure variations of ultra-thin HfO2 films induced by post-deposition annealing

To meet challenges for a smaller transistor feature size, ultra-thin HfO₂ high-k dielectric has been used to replace SiO₂ for the gate dielectric. In order to accurately analyze the ultra-thin HfO₂ films by grazing incidence X-ray reflectivity (GIXRR), an appropriate material model with a proper layer structure is required. However, the accurate model is difficult to obtain, since the interfaces between layers of the ultra-thin HfO₂ films are not easily identified, especially when post-deposition annealing process is applied. In this paper, 3.0 nm HfO₂ films were prepared by atomic layer deposition on p-type silicon wafer, and annealed in Ar environment with temperatures up to 1000 °C. The layer structures and the role of the interfacial layer of the films in the post-deposition annealing processes were evaluated by X-ray diffraction and X-ray photoelectron spectroscopy (XPS). The experimental results and analysis showed that layer thicknesses, crystal phases and chemical structures of the ultra-thin HfO₂ films were significantly dependent on annealing temperatures. The binding energy shifts of Hf 4f, O 1s, and Si 2p elements revealed the formation of Hf silicate (Hf-O-Si bonding) with increasing annealing temperatures. Due to the silicate formation and increasing silicon oxide formation, the interface broadening is highly expected. The structure analysis of the GIXRR spectra using the modified material structure model from the XPS analysis confirmed the interfacial broadening induced by the post-deposition annealing.     

X-ray photoemission of valence electrons in cuprous halides,and lead and cadmium iodides.

The XPS spectra of valence bands in CuCl, CuBr, and CuI show the existence of two separate bands after appropriate deconvolution. The evaluation of the average p-d mixing rates on the basis of the tight-binding approximation with a few simplifying assumptions indicates, in accordance with optical data, that the upper valence band arises mainly from the 3d state of copper. The XPS spectra coincide qualitatively but not quantitatively with the spectra obtained by excitation with monochromatized synchrotron radiation at 40, 61, and 76 eV. A major difference is that the ratio of integrated intensity of the upper band to that of the lower band is larger in the case of excitation at 40, 61, or 76 eV. The XPS spectra of PbI₂ and CdI₂ have also been measured. In both materials, the valence band spectra have composite structures and give two well-defined peaks after deconvolution. The profile of the spectrum of PbI₂ appears to have some deviation from the reported energy band structure.     

On line shape analysis in X-ray photoelectron spectroscopy

Any solid state X-ray photoelectron spectrum (XPS) contains contributions due to multiple inelastic scattering in the bulk, surface excitations, energy losses originating from the screening of the final state hole (intrinsic losses), and, for non-monochromatized incident radiation, ghost lines originating from the X-ray satellites. In the present paper it is shown how all these contributions can be consecutively removed from an experimental spectrum employing a single general deconvolution procedure. Application of this method is possible whenever the contributions mentioned above are uncorrelated. It is shown that this is usually true in XPS to a good approximation. The method is illustrated on experimental non-monochromatized MgK spectra of Au acquired at different detection angles but for the same angle of incidence of the X-rays.     

Disorder broadening of alloy Auger spectra

Auger spectroscopy promises the means to separate initial and final state contributions to the disorder broadening of core XPS spectra in disordered alloys. Auger disorder broadening, deduced from recent ab initio results, is predicted to be greater than XPS disorder broadening for Cu₅₀Pd₅₀ and Ag₅₀Pd₅₀ alloys. Simulations are used to assess whether this effect is observable experimentally despite the greater lifetime broadening of Auger spectra. A number of cases where narrow core–core–core Auger transitions should allow clear experimental identification of this effect are identified. The prospects for determining environment-resolved Auger spectra using APECS have been investigated.     

Study of YBa2Cu3O7−δ ceramics/Al interface

Properties of oxide layers at the interface between Al metal and YBa₂Cu₃O_7−δ ceramics were investigated using X-ray diffraction (XRD), X-ray photoelectron spectra (XPS) and impedance measurements. There have been found the oriented grains having their c-plane parallel to the surface of YBa₂Cu₃O_7−δ. When Al metal was evaporated onto the ceramics sample, the aluminum oxide layer was produced at the interface between Al and YBa₂Cu₃O_7−δ because Al metal oxidizes more easily. The oxygen-deficiency was observed at the ceramics side of the interface as examined by X-ray photoelectron spectra. This oxygen-deficiency can be partly replenished by post annealing whereas it can be fully replenished for the sample to which the mechanical polishing is applied beforehand in order to remove the oriented grains. The thickness of aluminum oxide layer was evaluated by means of the impedance measurements using the alternating current three-terminal method.     

Resolving surface chemical states in XPS analysis of first row transition metals, oxides and hydroxides: Cr, Mn, Fe, Co and Ni

Chemical state X-ray photoelectron spectroscopic analysis of first row transition metals and their oxides and hydroxides is challenging due to the complexity of their 2p spectra resulting from peak asymmetries, complex multiplet splitting, shake-up and plasmon loss structure, and uncertain, overlapping binding energies. Our previous paper [M.C. Biesinger et al., Appl. Surf. Sci. 257 (2010) 887-898.] in which we examined Sc, Ti, V, Cu and Zn species, has shown that all the values of the spectral fitting parameters for each specific species, i.e. binding energy (eV), full wide at half maximum (FWHM) value (eV) for each pass energy, spin-orbit splitting values and asymmetric peak shape fitting parameters, are not all normally provided in the literature and data bases, and are necessary for reproducible, quantitative chemical state analysis. A more consistent, practical and effective approach to curve fitting was developed based on a combination of (1) standard spectra from quality reference samples, (2) a survey of appropriate literature databases and/or a compilation of literature references and (3) specific literature references where fitting procedures are available. This paper extends this approach to the chemical states of Cr, Mn, Fe, Co and Ni metals, and various oxides and hydroxides where intense, complex multiplet splitting in many of the chemical states of these elements poses unique difficulties for chemical state analysis. The curve fitting procedures proposed use the same criteria as proposed previously but with the additional complexity of fitting of multiplet split spectra which has been done based on spectra of numerous reference materials and theoretical XPS modeling of these transition metal species. Binding energies, FWHM values, asymmetric peak shape fitting parameters, multiplet peak separation and peak area percentages are presented. The procedures developed can be utilized to remove uncertainties in the analysis of surface states in nano-particles, corrosion, catalysis and surface-engineered materials.     

Photoelectron Energy Loss in Al(002) Revisited: Retrieval of the Single Plasmon Loss Energy Distribution by a Fourier Transform Method

A Fourier transform (FT) algorithm is proposed to retrieve the energy loss function (ELF) of solid surfaces from experimental X-ray photoelectron spectra. The intensity measured over a broad energy range towards lower kinetic energies results from convolution of four spectral distributions: photoemission line shape, multiple plasmon loss probability, X-ray source line structure and Gaussian broadening of the photoelectron analyzer. The FT of the measured XPS spectrum, including the zero-loss peak and all inelastic scattering mechanisms, being a mathematical function of the respective FT of X-ray source, photoemission line shape, multiple plasmon loss function, and Gaussian broadening of the photoelectron analyzer, the proposed algorithm gives straightforward access to the bulk ELF and effective dielectric function of the solid, assuming identical ELF for intrinsic and extrinsic plasmon excitations. This method is applied to aluminum single crystal Al(002) where the photoemission line shape has been computed accurately beyond the Doniach–Sunjic approximation using the Mahan–Wertheim–Citrin approach which takes into account the density of states near the Fermi level; the only adjustable parameters are the singularity index and the broadening energy Г (inverse hole lifetime). After correction for surface plasmon excitations, the q-averaged bulk loss function, <Im[??1 / ε(E, q)]> _q , of Al(002) differs from the optical value Im[??1 / ε(E, q?=?0)] and is well described by the Lindhard–Mermin dispersion relation. A quality criterion of the inversion algorithm is given by the capability of observing weak interband transitions close to the zero-loss peak, namely at 0.65 and 1.65 eV in ε(E, q) as found in optical spectra and ab initio calculations of aluminum.     

The effect of atomic ordering on the photoelectron spectra of V3Au

The X-ray photoelectron (XPS) spectra of the valence bands are measured on 3 different modifications of the compound V₃Au, as obtained from the same starting sample. Two of these modifications are of the same crystal structure type (A 15), but have a different LRO parameter, S. An increase of the XPS V3d density of states, in agreement with the observed increase of the electronic specific heat coefficient, γ, and the superconducting transition temperature, T_c.     

Inner shell excitation,valence excitation and core ionization in NF3 studied by electron energy-loss and X-ray photoelectron spectroscopies

Electron energy-loss Spectroscopy (EELS) at impact energies of 2.5–3 keV has been used to obtain the electron excitation spectra for the N 1s (K-shell), F 1s (K-shell) and valence shell regions of NF₃. The inner shell spectra were recorded using small angle scattering (?1° ) while the valence shell spectrum was obtained at zero degree scattering angle. The inner shell excitation spectra show a strongly enhanced 1s→ δ^* type transition and continuum features which are typical for molecules with highly electronegative ligands. One of the peaks in an earlier published photoabsorption study of the N 1s region has been shown to be due to a N₂ impurity. The valence shell electron energy-loss spectrum shows a number of transitions which are considered to be mainly due to valence-valence type transitions, with also some evidence of Rydberg structure.The X-ray photoelectron spectra (XPS) of the N 1s and F 1s electrons along with their associated satellite structures have also been recorded using Al Kα (1486.58 eV) radiation. The vertical ionization potentials for the N 1s and F 1s electrons were found to be 414.36 (10) eV and 693.24 (10) eV, respectively. Both spectra exhibit a rich and different satellite structure. These “shake-up” features in the satellite XPS spectra are compared with continuum features of the inner shell electron energy-loss spectra and also with the valence shell spectrum.     

Origin of the surface recombination centers in ZnO nanorods arrays by X-ray photoelectron spectroscopy

The surface composition of as-grown and annealed ZnO nanorods arrays (ZNAs) grown by a two-step chemical bath deposition method has been investigated by X-ray photoelectron spectroscopy (XPS). XPS confirms the presence of OH bonds and specific chemisorbed oxygen on the surface of ZNAs, as well as H bonds on surfaces which has been first time observed in the XPS spectra. The experimental results indicated that the OH and H bonds play the dominant role in facilitating surface recombination but specific chemisorbed oxygen also likely affect the surface recombination. Annealing can largely remove the OH and H bonds and transform the composition of the other chemisorbed oxygen at the surface to more closely resemble that of high temperature grown ZNAs, all of which suppresses surface recombination according to time-resolved photoluminescence measurements.     

Magnetic,electronic and structural properties of ZnxFe3−xO4

A series of samples Zn_xFe_3−xO₄ have been prepared by the chemical coprecipitation technique and characterized by X-ray diffraction (XRD), vibrating sample magnetometry (VSM) and X-ray photoelectron spectroscopy (XPS). XRD demonstrates all the samples of Zn_xFe_3−xO₄ have a spinel structure same as Fe₃O₄. The magnetic hysteresis loops of Zn_xFe_3−xO₄ obtained from VSM indicate that the saturation magnetization has a maximum when x is ∼1/3. The chemical states of Fe atoms and Zn atoms in zinc ferrites have been measured using XPS and Auger electron spectroscopy (AES). The Fe 2p core-level XPS spectra and Zn L₃M₄₅M₄₅ Auger peaks have been analyzed and the results have been discussed in correlation with the samples’ magnetic properties. These results suggest most of Zn atoms occupy the tetrahedral sites and a small amount of them occupy the octahedral sites.     

Polymeric like carbon films prepared from liquid gas and the effect of nitrogen

Polymeric like carbon (PLC) films are grown by a capacitance coupled RF-PECVD on the grounded electrode at room temperature from liquid gas (40% propane and 60% butane) in two regimes with nitrogen and without nitrogen gas. Films are characterized by X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM), Fourier transform infrared (FTIR) absorption and Raman spectroscopy. The result of FTIR analyses indicates that more than 90% of hydrogen atoms are bonded to carbon with sp³ hybridization. The abundance of CH₃ is more than that of CH₂ and this one is more than that of CH for carbon with sp³ hybridization in these films. The C 1s line of the XPS spectra is deconvoluted to several peaks that are attributed to the CH₃, CH₂ and CH terminations. The result of this deconvolution is consistent with FTIR results. AFM images show that the mean nanoparticle size is reduced from about 100 nm for films without nitrogen to less than 80 nm for films with nitrogen. This is in agreement with our Raman results. By addition of nitrogen to the feed gas, no variation in the C-H stretching vibration mode is observed. The effect of N-H bonds is observable in both FTIR and XPS spectra and a very small trace of N-C bonds is present only in deconvolution of N 1s line of XPS spectra. These results indicate that by addition of nitrogen to feed gas, internal structure of a-C:H nanoparticles is not changed but particle size is decreased. We suggest that the internal stress reduction due to nitrogen addition in the feed gas for PLC films can be related to decreasing of the a-C:H particle size.     

Molecular structural analysis of 55mol% CuI-45mol% Ag2MoO4 solid electrolyte using XPS and laser raman techniques

The solid electrolyte 55 mol% CuI - 45 mol% Ag₂MoO₄ has been analyzed structurally by employing X-ray photoelectron spectroscopic (XPS) and laser Raman studies. The core level Ag 3d, Cu 2p, Mo 3d, I 3d and O1s XPS spectra of the constituent elements were recorded. The presence of both bridging (BO) and non-bridging (NBO) oxygen atoms in the oxyanion framework has been identified through the deconvolution of the O 1s spectrum. Laser Raman spectroscopic studies tend to reveal that the oxyanion framework is constructed with [MoO ₄ ²⁻ ]-tetrahedral and [MoO₆]-octahedral units as in the case of CuI - Cu₂MoO₄ and AgI - Ag₂O - MoO₃ glasses.