Core Level Peak Research Articles

Band line-up determination at p- and n-type Al/4H-SiC Schottky interfaces using photoemission spectroscopy

The band lineup of p- and n-type 4H–SiC/Al interfaces was determined using x-ray photoemission spectroscopy (XPS). Al was deposited in situ on ex situ cleaned SiC substrates in several steps starting at 1.2 Å up to 238 Å nominal film thickness. Before growth and after each growth step, the sample surface was characterized in situ by XPS. The analysis of the spectral shifts indicated that during the initial deposition stages the Al films react with the ambient surface contamination layer present on the samples after insertion into vacuum. At higher coverage metallic Al clusters are formed. The band lineups were determined from the analysis of the core level peak shifts and the positions of the valence bands maxima (VBM) depending on the Al overlayer thickness. Shifts of the Si 2p and C 1s XPS core levels occurred to higher (lower) binding energy for the p-(n-)type substrates, which was attributed to the occurrence of band bending due to Fermi-level equilibration at the interface. The hole injection barrier at the p-type interface was determined to be 1.83±0.1 eV, while the n-type interface revealed an electron injection barrier of 0.98±0.1 eV. Due to the weak features in the SiC valence bands measured by XPS, the VBM positions were determined using the Si 2p peak positions. This procedure required the determination of the Si 2p-to-VBM binding energy difference (99.34 eV), which was obtained from additional measurements.

XPS studies of chemically etched surfaces of (La,Sr)(Al,Ta)O 3 single crystals

The electronic structure measurements of three LSAT single crystals, annealed in hydrogen atmosphere at 900–1400 °C for 1 or 4 h, slowly or rapidly cooled were performed. The existence of core level peak satellites at the lower binding energy site were found. These satellites may be related to the defect states. The crystal annealed at 1400 °C and slowly cooled has the smallest low binding energy satellites. For La 3d, besides the satellites at lower binding energy (S II), the satellites (S I) at higher binding energy to the main peak (M) were detected. It was found that the area peak ratio I M/ I SI of the La 3d states is not sensitive to the annealing, polishing or treatment with HCl and HNO 3:HF of the wafer surfaces. Contrary to that, the area peak ratio I M/ I SII depends on such treatment. The XPS results show that the treatment of the surface with HNO 3:HF increases the roughness of the surface which confirmed the AFM results and showed that the surface chemical composition changes with processing.

Electronic structure of electrochemically Li-inserted TiO2 studied with synchrotron radiation electron spectroscopies

The electronic properties of TiO2 and electrochemically Li-inserted TiO2 have been studied using synchrotron radiation photoelectron spectroscopy and x-ray absorption spectroscopy (XAS) in conjunction with resonant photoelectron spectroscopy. Core level (Ti 2p) and valence level spectra show the presence of Ti3+ states in LixTiO2. The x values determined from core level peak intensities were found to be directly correlated to the inserted amount of Li+ determined electrochemically. The x-dependent width of the Ti 2p peaks is consistent with a two-phase regime at intermediate x values. Resonant photoelectron spectroscopy at the Ti 2p edge was performed for TiO2 and Li0.5TiO2 to delineate the Ti4+ and Ti3+ contributions to the XAS spectrum.

Study of the early stages of Cu/6H–SiC(0 0 0 -1) interface formation

The early stages of the Cu/SiC interface formation at room temperature and the influence of annealing were investigated on the carbon-face of the single crystal n-type 6H–SiC, by X-ray photoelectron and auger electron spectroscopy (XPS/XAES), low energy electron diffraction and work function (WF) measurements. Upon stepwise copper evaporation in UHV up to a thickness of 5–10 monolayers (ML) at RT, the binding energy of the XPS Cu2p 3/2 core level peak shifted from 933.6±0.1 eV, below 0.1 ML coverage, to 932.7 eV for the final metallic Cu deposit. The WF, following an initial steep decrease from the clean SiC substrate value of 4.4 down to 4.0 eV up to 0.1 ML, increased then gradually to the metallic Cu value of ∼4.75 eV after approximately 5 ML of Cu. The growth of the film was initially via 2D-cluster formation, and exhibited a 3D character beyond 1–2 ML of Cu. The height of the Schottky barrier for the Cu/6H–SiC(0 0 0 -1) contact was found by XPS to be 1.5±0.1 eV. Annealing of the contact up to 520 K caused small changes in the Cu and SiC XPS peak intensities accompanied by a 0.3 eV increase of the WF, indicating small structural changes within the film and gave rise to a slight change of the Schottky barrier height to 1.6±0.1 eV. The results do not indicate any significant silicide formation at 520 K.

Photoemission study of the initial stage of Er/Si(100) interface formation

The initial stage of Er/Si(100) interface formation has been investigated by usingsynchrotron radiation photoelectron spectroscopy combined with low-energyelectron diffraction. Both the valence band and the core level peaks of the Siphotoemission spectra shift rigidly with increasing Er coverage in thesubmonolayer region. Upon depositing 0.6 monolayers of Er on the Si(100) surfaceat room temperature, the surface Fermi level is ultimately pinned at 0.29 eVabove its initial value, which is equivalent to a Schottky barrier height of 0.67 eV.No evidence is found for the formation of Er silicides at the as-depositedsurfaces. Annealing of the Er-covered Si(100) surfaces at 600ºC results in theappearance of a new peak located 1.2 eV below the Si 2p peak, indicatingthe presence of some sort of Er silicide. Meanwhile, the Er 4f spectrummeasured for samples upon annealing exhibits a well-resolved fine structure,implying that only monospecies of Er silicide may exist on the surface.

Surface evolution of Ni–V transparent oxide films upon Li insertion reactions

AbstractA detailed picture is reported of the nature of an Li insertion Ni–V mixed‐oxide electrode and of its evolution upon prolonged intercalation/deintercalation cycles. A combination of bulk and surface‐sensitive techniques is used, including electrochemical analysis (cyclic voltammetry, impedance spectroscopy), atomic force microscopy (AFM) and XPS. The Ni–V oxide film, obtained by r.f. sputtering as a thin film on conductive glass for use as a transparent electrode in electrochromic windows, is an Li insertion material with high charge capacity, very good reversibility and a surface morphology characterized by a low initial roughness. Its surface composition shows an initial Ni enrichment. The nominal Ni–V stoichiometry is recovered already after ten cycles. Extended Li charge–discharge cycles induce progressive changes in the electrode reversibility (capacity fading) and in the reaction kinetics (increase of impedance). The surface evolution consists of the progressive accumulation of an Li carbonate layer, with thickness and roughness increasing with the number of cycles. The constancy of the core level and Auger peaks of Ni and V after 100 cycles excludes major changes in the electronic structure of the host oxide. However, electrochemical and AFM experiments performed after 1000 cycles indicate that the carbonate deposition on the electrode progresses further, therefore the surface layer is unable to yield a permanent electrode passivation and stabilization. Copyright © 2002 John Wiley & Sons, Ltd.

Interface electronic structure of Ta2O5–Al2O3 alloys for Si-field-effect transistor gate dielectric applications

Interfacial electronic structure is important for a fundamental understanding of the properties of field-effect transistor (FET) device structure systems. Previous studies using soft x-ray photoelectron spectroscopy (SXPS) have demonstrated well-defined interface states that appear at binding energies between the peaks due to the substrate Si and the oxide SiO2. Recently we have shown that significant interface changes with annealing for the SiO2/Si system commonly used for FET gate dielectrics illustrating the effectiveness of SXPS for the current study. The present article presents SXPS studies using synchrotron radiation of the electronic structure at the interface between Ta2O5–Al2O3 alloys and Si(111). This system is typical of FET alternative-gate-oxide films and has an electronic structure characteristic of high-k gate dielectrics. Thin films (<20 Å) of Ta2O5–Al2O3 alloys were carefully prepared using remote plasma enhanced chemical vapor deposition on Si(111). For this study, SXPS spectra were taken for the Si 2p, Al 2p, and Ta 4f core levels. The Si 2p SXPS data confirm that SiO2 is formed at the Si interface during deposition for all alloys (in agreement with previously published results) and that the Si interface is qualitatively similar to SiO2/Si samples. The binding energy of the Si4+ Si 2p3/2 core level varies significantly for our alloy samples and this is interpreted as a screening effect rather than being due to Ta or Al silicate formation. The Al 2p and Ta 4f SXPS core level peak positions reveal screening effects. For these data, line shape analysis also supports alloy homogeneity as independently determined by Auger electron spectroscopy and Fourier transform infrared spectroscopy.

Chemical nature of nanostructures of La 0.6Sr 0.4MnO 3 on SrTiO 3(1 0 0)

In order to elucidate the dynamics of coalescence of La 0.6Sr 0.4MnO 3 nanostructures on SrTiO 3(1 0 0), we have investigated the relationship between the electronic states of La 0.6Sr 0.4MnO 3 and the surface morphology of nanostructures, by using photoelectron spectroscopy and local probe microscopy. When the La 0.6Sr 0.4MnO 3 nanodots are heated from 700 to 800 °C, the average nanodot size increases, and the core level peaks of Mn 2p and La 3d do not change. In addition, the observed nanodot height is always 0.4 nm. Therefore, when the nanodots coalesce, they are laterally grown, maintaining the perovskite manganese structure. On the other hand, when nanowire structures are formed along the steps by heating the sample above 1000 °C, the Mn 2p peaks vanish, although the La 3d spectrum does not change. Therefore, upon this process, Mn atoms of the island dissolve into the bulk of SrTiO 3(1 0 0), and consequently, the remaining species migrate to steps on the surface, resulting in the formation of step-decorated nanowire perovskite structures.

N- GaN surface treatments for metal contacts studied via x-ray photoemission spectroscopy

The surface chemistry and electronic properties of n-GaN surfaces were studied via x-ray photoemission spectroscopy before and after wet chemical treatments. Shifts of the surface Fermi level were measured with the change in position of the Ga 3d core level peak. HCl treatment of n-GaN led to a 0.9 eV shift of the surface Fermi level toward the conduction band minimum, while KOH treatment led to a 0.3 eV shift of the surface Fermi level toward the valance band maximum. These shifts lead to a reduction in the surface barrier for HCl-treated n-GaN and for KOH-treated p-GaN, potentially improving contact resistance. The changes in surface chemistry indicate that a N (or Ga) deficiency with HCl(KOH) treatment alters the surface state density through the formation of donor (acceptor)-like states.

Surface reactivity of the alkali-metal-induced Ge( [formula omitted])-3×1 surface under ultrahigh vacuum

We have investigated the surface reactivity of the Li- and Na-induced Ge(1 1 1)-3×1 surfaces in ultrahigh vacuum (mid-10 −10 Torr range) using core-level photoemission. The core-level peaks from the Li/Ge(1 1 1)-3×1 surface showed significant changes over about 15 min, while those from the Na/Ge(1 1 1)-3×1 surface remained unchanged. We identified that the changes in the Li/Ge(1 1 1)-3×1 surface were due to the contamination by residual oxygen-containing gases in the chamber. The contamination rate of the Li/Ge(1 1 1)-3×1 surface was higher than that of the Ge(1 1 1)-c(2×8) surface, indicating that its surface reactivity is enhanced (in contrast to the passivation of the Na/Ge(1 1 1)-3×1 surface). The Li/Ge(1 1 1)-3×1 case is against the notion that the common 3×1 reconstruction induced by alkali-metal adsorbates passivates the surface as shown for the Na-induced Si(1 1 1)-3×1 and Ge(1 1 1)-3×1 surfaces.

Photoelectron Spectroscopy Study of the Electronic Structures of Al/MgF2/tris-(8-hydroxyquinoline)aluminum Interfaces

We have studied the electronic structures of Al/MgF2/tris-(8-hydroxyquinoline)aluminum (Alq3) interface using UV and X-ray photoelectron spectroscopy (UPS & XPS). The UPS revealed that the valence peak shift occurred with MgF2 deposition before Al was deposited and was independent of the gap state formation. The XPS core level peaks indicated that the MgF2 strongly interacted with Al and O atoms in Alq3 even before Al was deposited, and the deposition of Al caused slight change to the N 1s core level peak. These results indicate that the interaction mechanism in Al/MgF2/Alq3 is different from those found in Al/LiF/Alq3 and other metal/Alq3.

Cluster Assembled Sb Films: Studies On Structure, Surface And Optical Properties

A low energy cluster beam deposition (LECBD) technique has been used to prepare the Sb cluster films on different substrates and are characterized using a variety of probes. Proton induced X-ray emission (PIXE) analysis shows the absence of any foreign trace elemental impurity even at ppm level. Glancing angle X-ray diffraction (GXRD) and transmission electron diffraction (TED) studies reveal the presence of single crystalline feature of Sb with hexagonal symmetry along with Sb-oxides. The transmission electron micrograph (TEM) of the cluster films of thickness 20 Å and 100 Å show size distribution which is more for the 100 Å film compared to that of the 20 Å one. The photoluminescence (PL) studies at 300 K show red shifted peaks along with the one possibly due to HOMO-LUMO transition around 2.4 eV. X-ray photoelectron spectroscopy (XPS) of Sb cluster films show shoulders to the core level peaks of the Sb corresponding to 3d 3/2 and 3d 5/2 which are shifted by 1.0 to 2.0 eV suggesting the formation of Sb-oxide. Raman scattering studies show the shift of A 1g and E g vibrational modes from their bulk value.

Differential photoelectron holography

Atoms near a selected surface atomic site can be imaged in three dimensions by photoelectron holography. The experiment involves acquiring hundreds or thousands of angle-resolved photoemission spectra from a core level of the selected atomic site over a wide range of photon energy and for a large number of emission directions. For good image quality, the precision of the intensity measurement must be better than ∼1% over the entire data set, but this is difficult to achieve. Two methods based on differential measurements that allow intensity self-normalization are presented here as a solution. One method is based on measurements of intensity branching ratios between spin–orbit-split core level peaks, and the other is based on measurements of logarithmic derivatives of the intensity function. These methods are illustrated by two examples, As-terminated Si(1 1 1) and Bi-terminated Si(1 1 1). The As/Si(1 1 1) results are in excellent agreement with the accepted structure of the system, thus verifying the efficacy of the methods. The Bi/Si(1 1 1) results resolve a long-standing controversy regarding the correct structure of this system.

Time-resolved study of beryllium surface reactions using electron momentum spectroscopy of the core-level

We have measured the binding energy of the Be 1s core-level in metallic beryllium using an electron-impact ionisation technique––electron momentum spectroscopy (EMS). The value we obtain, 111.7 ± 0.1 eV, is in good agreement with previous results. We have also determined the chemical shift of the core level in an oxygen environment to be 2.68 ± 0.06 eV. Using the same technique, we have followed the evolution of the core-level peak as Be undergoes surface reaction induced by background gases present in the vacuum (mainly water vapour and nitrogen). The core-level peak intensities as a function of time clearly show that reaction of the Be surface progresses in two distinct steps. The initial relatively rapid stage progresses up to a coverage of around 18 L of H 2O and N 2 after which the onset of a slower reaction process is observed. These results demonstrate the ability of EMS to provide time-resolved electronic structure measurements as a solid undergoes chemical modification.

Electronic structure of NaNbO 3–Mn single crystals

The XPS spectra of the NaNbO 3 single crystals doped with manganese ( x=0.1, 0.5, 1.0, 3.0, 5.0 and 10.0%) were measured using the PHI 5700/660 physical electronics photoelectron spectrometer at room temperature within the antiferroelectric phase. From the spectra the atomic concentration was calculated and compared to the nominal composition. The approximate solubility limit of the Mn ions, built into the sodium niobate crystal lattice, was determined as 1.0% Mn. The dependence of the core level peak position and valence band on manganese dopant was determined. The influence of sputtering with the Ar + ions on the Nb core line and valence band shape was observed. This effect was ascribed to the change in niobium valence due to the oxygen defects produced during the sputtering process.

Photoelectron spectroscopy study of the electronic structures of Al/MgF2/tris-(8-hydroxyquinoline) aluminum interfaces

We have studied the electronic structures of Al/MgF2/tris-(8-hydroxyquinoline)aluminum (Alq3) interface using UV and x-ray photoelectron spectroscopy [(UPS) and (XPS), respectively]. The UPS revealed that the valence peak shift occurred with MgF2 deposition before Al was deposited and was independent of the gap state formation. The XPS core level peaks indicated that the MgF2 strongly interacted with Al and O atoms in Alq3 even before Al was deposited, and the deposition of Al caused a slight change to the N 1s core level peak. These results indicate that the interaction mechanism in Al/MgF2/Alq3 is different from those found in Al/LiF/Alq3 and other metal/Alq3.

Stability of cerium oxide on silicon studied by x-ray photoelectron spectroscopy

The silicon–cerium oxide interface is studied using x-ray photoelectron spectroscopy. The oxidation and reduction of species at the interface are examined as a function of annealing temperature both in vacuum and oxygen ambient, in order to determine their relative stabilities. By depositing a very thin CeO2 film (∼30 Å), the cerium and silicon core level peaks can be monitored simultaneously. The presence of characteristic chemical shifts of the Si 2p peak gives information about any SiOx layer that may form at the interface. The oxidation state of the cerium can be probed from three different areas of the spectrum. From this information we can infer the oxidation state of both the silicon and the cerium. For the first time a complete picture of the interface is obtained. The implications of these findings on the utility of CeO2 in device applications are discussed.

Angle-resolved XPS measurements on copper phthalocyanine thin films

Copper phthalocyanine films about 100 nm in thickness were produced by vacuum sublimation. Different heat treatments transformed the layer from the starting α- to the β-structure. The chemical states of the layer were characterized by XPS during this transformation. The measurements from core level peaks and the valence band spectra showed systematic changes. At temperatures above 300°C the Cu detached itself from the chemical bonds in the molecule, the aromatic share of the carbon decreased and new CN bonds were formed.

In situ characterization of the nitridation of AIII–BV semiconductor surfaces by means of X-ray photoelectron spectroscopy

The effect of low-energy N 2 + ion beam bombardment on InAs, InP, and InSb surfaces has been studied using in situ X-ray photoelectron spectroscopy. The formation of a nitrided surface layer has been observed for all investigated AIII–BV semiconductors. Beside the emerging InN bonds also PN bonds were detected for InP, whereas for InAs and InSb, there is no evidence for the formation of AsN and SbN bonds, respectively. An analysis of the N 1s core level peak also reveals the build-in of interstitial nitrogen. A change of the ion beam incidence angle has great influence on the composition of the surface layer as demonstrated for InP. The amount of nitrogen in PN bonds and of interstitial nitrogen decreases for changing the ion incidence to grazing angles. A detailed XPS analysis provides information on the temporal evolution of the process of nitridation and the thickness of the nitrided surface layer.

A photoemission study of the SO2 adsorption on TiO2 (110) surfaces

We have investigated the adsorption of the SO2 molecule on a TiO2(110) single crystal at a substrate temperature of 120 K by means of core level synchrotron radiation photoemission. By studying the S 2p and O 2s core level peaks, we show that from the first adsorption stages up to the saturation coverage, SO2 adsorbs molecularly on the TiO2(110)-(1×1) surface forming a sulphite–sulphate-like mixed and disordered adsorbed layer. The bridge oxygen atoms of the clean surface are involved in the oxidation of the molecule.