Oxygen-exposed Surface Research Articles

Gallium vacancy formation in oxygen annealed β-Ga2O3

In this study, the formation and character of gallium vacancies (VGa) and their complexes in near surface and bulk regions of single crystal β-Ga2O3 were explored using unintentionally doped single crystals grown by the Czochralski method. As-grown and O2 annealed (up to 1550 °C) samples were investigated using positron annihilation spectroscopy (PAS) to study the top 0.05–6 μm, and also current–voltage measurements and infrared (IR) spectroscopy, with hydrogenated samples to probe VGa, to study the bulk. After annealing in O2 > 1000 °C, the β-Ga2O3 resistivity begins increasing, up to ∼109 Ω cm for 1550 °C treatment, with the top 0.5 mm being many orders of magnitude more resistive. PAS measurements of the top 6 μm (S values) and very near surface 200 nm (diffusion length, L) indicate differential behavior as a function of peak annealing temperature. At least four temperature regimes of behavior are described. VGa are present in the bulk after growth, but considerable changes occur upon annealing at a temperature ≈1000 °C, where L and S decrease simultaneously, suggesting an increasing defect concentration (L) but a decreasing defect volume (S). Annealing at a temperature ≈1400 °C increases S again, showing an increasing volume concentration of VGa, with IR absorption showing a large signature of VGa-2H, indicative of increased VGa formation that was not present when annealing at a temperature ≈1000 °C. These results suggest that defect changes from annealing in oxygen are depth dependent, and that VGa configuration may not be the same near the oxygen-exposed surface of the sample and in the bulk.

INFLUENCE OF SPECIFIC THERMAL TREATMENT ON YOUNG'S MODULUS CHANGES IN A NATURAL RUBBER–BASED COMPOUND: STUDY BY AFM NANOINDENTATION

ABSTRACT The determination and study of Young's modulus at the nanometric scale in an efficient and fast way is carried out using atomic force microscopy (AFM) in contact mode. The aim of this study is to evaluate the impact of a specific thermal treatment on Young's modulus changes from the surface toward the bulk of a cured natural rubber (NR)–based compound. The AFM nanoindentation experiments were conducted on a cross-linked NR sample. Following a depth profiling analysis method, AFM deflection distance curves were measured every 40 μm from the oxygen-exposed surface toward the bulk, and then the Hertz's model was applied to force-indentation curves for the calculation of the Young's modulus. Investigations on the thermal-treated samples revealed an overall increase of the Young's modulus along the samples after heating. Moreover, a decrease of Young's modulus values from the surface exposed to oxygen during heating toward the bulk was observed, indicating a stiffening of the sample at the surface.

Photoemission study of sulfur and oxygen adsorption on GaN( [formula omitted

The surface electronic structure of thin film n-type GaN( 0 0 0 1 ¯ ) has been studied using high resolution angle resolved photoemission spectroscopy. Clean surfaces, sulfur-exposed surfaces, and oxygen-exposed surfaces were investigated. Spectra were recorded for the Ga 3d shallow core states and for the valence band states. The sulfur covered surface was found to be inert with respect to subsequent oxygen exposure. For the clean Ga-adlayer terminated ( 0 0 0 1 ¯ ) surface, the Fermi level is pinned by surface states at least 1.4 eV below the bottom of the conduction band. Sulfur and oxygen covered surfaces exhibit smaller values for surface band bending, with the Fermi edge lying 0.7–1.0 eV below the conduction band minimum. Finally, we discuss a possible new interpretation for the asymmetry in the shape of Ga 3d photoemission feature.

Photodesorption and Trapping of Molecular Oxygen at the TiO2(110)−Water Ice Interface

By means of temperature programmed desorption (TPD), static secondary ion mass spectroscopy (SSIMS), and electron energy loss spectroscopy (EELS), we have investigated further the states of oxygen adsorbed on rutile TiO2. Previous work has shown that annealing the (110) surface in a vacuum produces isolated bridging oxygen vacancies, and that these vacancies are intimately connected with molecular and dissociative oxygen adsorption channels. We find that at 115 K illumination of the oxygen-exposed surface with 4.1 eV photons results in photodesorption of the oxygen associated with a TPD state centered on 410 K, in contrast to the remaining oxygen destined for a dissociative channel. An unusual effect of water overlayers on the photochemical properties of the O/TiO2(110) system is explored. For thick overlayers, it is possible to generate via UV irradiation a previously unobserved oxygen TPD state. Evidence is presented for this new O2 TPD state originating from the photolysis of an O2−water adduct.

Scanning tunnelling microscopy investigation of the oxygen-induced faceting and “nano-faceting” of a vicinal copper surface

Scanning tunnelling microscopy (STM) has been applied to investigate the well-known tendency of Cu(100) vicinals to facet to Cu(410) in the presence of chemisorbed oxygen. Room-temperature imaging was performed on a nominally (610) face of Cu following different oxygen exposures and annealing treatments. The well-annealed oxygen-exposed surface shows large (410) facets but a rather small tendency to produce comparably sized (100) facets; these only occur in conjunction with (401) facets which arise due to a miscut of the crystal surface by a few degrees out of the [001] zone relative to a pure (n10) face. Instead, the additional (100) surface needed to make up the average surface orientation occurs in the form of regions of wider (100) terraces between groups of terraces of (410) width, leading to nanofacet regions with modulated step-edge spacings. Oxygen dosing around room temperature produces small (410) facets elongated in the [001] direction, with pile-up of the kinks resulting from the azimuthal twist leading to 〈100〉 step-edge facets as a precursor to the formation of (100) and (401) two-dimensional facets which form on annealing. Explicit evidence is found for significant adatom diffusion at room temperature, even over the oxygen-induced (410) facets. Atomic-scale images of nanofacets, which can be regarded (4n10) regions with n an integer, provide some further information on specific atomic models of oxygen-covered Cu(100) vicinals in this zone.

Very-low-energy electron-diffraction analysis of oxygen on Cu(001).

We have developed a method for adding the effect of the surface barrier to the Van Hove--Tong suite of low-energy electron-diffraction programs enabling us to analyze very-low-energy electron-diffraction (LEED) spectra from complex reconstructed systems. The method models the effect of the surface-potential barrier on the intensities and accurately replicates the fine-structure features found at low energies on many surfaces. It is free from many of the simplifying assumptions made by others attempting to model similar systems. We find that the fine-structure peak positions depend on the inelastic scattering in the barrier region as well as the shape of the real part of the potential barrier. This limits the accuracy of measurement of the shape of the barrier, but useful shape information can be derived from these analyses. We have studied the effect of oxygen overlayers on Cu(001) as well as the clean Cu(001) surface. These surfaces are well described by the model for a range of incident-beam directions. On the oxygen-exposed surface, it is necessary to assume that a c(2\ifmmode\times\else\texttimes\fi{}2) structure exists as a precursor to the (\ensuremath{\surd}2 \ifmmode\times\else\texttimes\fi{}2 \ensuremath{\surd}2 )R45\ifmmode^\circ\else\textdegree\fi{} in order to explain all the data. Further, it is found that the surface-potential barrier moves away from the surface with establishment of the precursor c(2\ifmmode\times\else\texttimes\fi{}2) structure, but then moves back towards the surface as the final (\ensuremath{\surd}2 \ifmmode\times\else\texttimes\fi{}2 \ensuremath{\surd}2 )R45\ifmmode^\circ\else\textdegree\fi{} forms with additional exposure to oxygen. This accounts for the changes in work function with exposure to oxygen found on this surface.

Optical response of Si(111)-7 × 7

The optical properties of the Si(111)-7 × 7 surface are studied by means of a surface differential reflectometer. The optical reflectivity of the clean surface is first compared to that of the native oxide-covered Si surface, then to that of the oxygen-exposed surface. The exact surface optical response function of the Si(111)-7 × 7 surface and its changes upon adsorption of oxygen are determined by use of an electromagnetic surface model. A main peak is observed at 3.9 eV and is interpreted as due to electronic transitions from the back-bond states of the Si adatoms to a bulk conduction band. Secondary maxima at lower energies, due to transitions involving the dangling bond states, are also observed. Indications on the oxygen chemisorption mode are finally deduced.

Autoionization and energy-loss structures in Nb and Mo clean and oxygen-exposed surfaces

Autoionization features have been observed in Nb and Mo surfaces above the N 23VV Auger lines. Similarly to 3d transition metals, they are interpreted as due to a coherent decay of the excited 4p electron and to a subsequent interference effect with the direct photoexcitation channel from the valence band. If the radial dependence of the 4d wavefunction does not vary strongly across the band, the line shape of the autoionization electron emission is expected to be similar to the core-energy loss result from 4p core levels. The correspondence between secondary electron emission and energy loss backscattering yield has been studied for both the clean and oxygen-exposed surfaces. A close similarity is found between the electron-induced phenomena observed above the N 23VV Auger lines and the N 23 core absorption.

Autoionization and energy-loss structures in Nb and Mo clean and oxygen-exposed surfaces

Autoionization features have been observed in Nb and Mo surfaces above the N23VV Auger lines. Similarly to 3d transition metals, they are interpreted as due to a coherent decay of the excited 4p electron and to a subsequent interference effect with the direct photoexcitation channel from the valence band. If the radial dependence of the 4d wavefunction does not vary strongly across the band, the line shape of the autoionization electron emission is expected to be similar to the core-energy loss result from 4p core levels. The correspondence between secondary electron emission and energy loss backscattering yield has been studied for both the clean and oxygen-exposed surfaces. A close similarity is found between the electron-induced phenomena observed above the N23VV Auger lines and the N23 core absorption.

Projectile dependent sputter-erosion of oxygen-exposed molybdenum at different temperatures

The sputtering yield of molybdenum and angular distribution of sputtered molybdenum under oxygen exposure have been measured for D +- and He +-irradiation in the energy region from 2.5 keV to 6 keV at temperatures ranging from ambient room temperature to 640°C. The components of the sputtered particles from oxygen-exposed molybdenum have also been measured for 8 keV Ar +-irradiation at temperatures of 25°C and 485°C. The yields were determined from the weight loss of the targets and by the aid of the collector method which was applied to determine the angular emission distribution. Components of the sputtered particles were determined by means of the mass analysis of emitted neutral and charged particles. It was found that the oxygen exposure dependence of the yield strongly depends on the incident ions and on the temperature. The results are interpreted in terms of the chemical reactivity of the incident ions with oxygen, a dilution of metal atoms in the oxygen-exposed surface and a beam-induced desorption of Mo-oxide molecules.

Oxygen coverage dependent emission of sputtered neutrals and secondary ions

Energy spectra of sputtered neutrals and secondary ions emitted from molybdenum and indium have been measured for different oxygen surface coverages. In the case of the Mo the energy distribution of the neutrals exhibits a distinct broadening and a deficit in the low-energy flux at high oxygen coverages. This can be attributed to the strong increase of Mo + secondary ions. the latter then constituting 20–30% of the total emitted particles. By contrast, the energy spectra of neutral In atoms are less sensitive to the O 2 pressure, exhibiting a small (~ 2 eV) shift to lower energies and a less pronounced broadening. The energy distributions of In + and Mo − secondary ions also become broader with increasing O 2 coverage. The evolution of neutral and ion signals during sputtering of previously oxygen-exposed surfaces has also been measured. For both metals the intensities of neutrals and ions increase and decrease, respectively, reaching the steady-state level after the impact of some (2–3) × 10 15 Ar-ions/cm 2.

Autoionization and Auger features of clean and oxygen-exposed surfaces of Fe, Co and Ni.

The ${M}_{2}$,3VV Auger and 3p core loss spectra for clean and oxygen-exposed surfaces of the transition metals Fe, Co, and Ni have been studied. All the peaks observed in the Auger spectra are accounted for in terms of both atomic and bandlike ${M}_{2}$,3${\mathrm{M}}_{4}$,5${\mathrm{M}}_{4}$,5 transitions and an autoionization emission involving the decay of a localized excited state. For the clean surfaces the localized excitation and subsequent deexcitation by autoionization emission is considered, within a quasiatomic model, to be 3${p}^{6}$3${d}^{n}$\ensuremath{\rightarrow}3${p}^{5}$3${d}^{n}$${d}^{\mathrm{*}}$\ensuremath{\rightarrow}3p $^{6}$3${d}^{n\mathrm{\ensuremath{-}}1}$+ef. However, upon oxidation of these surfaces, the splitting observed in the 3p core loss spectra and the energy of the observed autoionization peaks indicate changes in the nature of the excitation due to the chemical environment. This is contrary to the accepted notion for such quasiatomic processes, which are considered essentially unaffected by the chemical environment. Further, for these oxidized surfaces, the observed autoionization features are shown to be associated with the excitation of 3p electrons to a localized state at the bottom of the conduction band as opposed to the empty d states at the top of the valence band.

A study of oxidized Nb–Al alloys using Auger electron spectroscopy and photoemission spectroscopy

The oxidation properties of Nb–Al alloys have been studied using Auger electron spectroscopy and x-ray photoelectron spectroscopy. Al in the alloy is found to be preferentially sputtered during argon-ion bombardment and to be preferentially oxidized in Nb–Al alloys. Oxidation increases the Al concentration in the oxide layer, which is seen even in the initial stages of oxidation with photoemission spectroscopy using synchrotron radiation. Auger peak shifts of 5×1011 L oxygen-exposed surfaces of Al, Nb, and Nb–Al alloys are found to be due to Auger transitions involving chemically shifted core levels caused by an alloying effect.

5105. Relations between the orientations of ion bombarded single crystals, resulting surface structures and sputtered spot patterns: V Alexander et alRad Effects,56 ( [formula omitted]), 1981, 241–250

The sputtering yield of molybdenum and angular distribution of sputtered molybdenum under oxygen exposure have been measured for D+- and He+-irradiation in the energy region from 2.5 keV to 6 keV at temperatures ranging from ambient room temperature to 640°C. The components of the sputtered particles from oxygen-exposed molybdenum have also been measured for 8 keV Ar+-irradiation at temperatures of 25°C and 485°C. The yields were determined from the weight loss of the targets and by the aid of the collector method which was applied to determine the angular emission distribution. Components of the sputtered particles were determined by means of the mass analysis of emitted neutral and charged particles. It was found that the oxygen exposure dependence of the yield strongly depends on the incident ions and on the temperature. The results are interpreted in terms of the chemical reactivity of the incident ions with oxygen, a dilution of metal atoms in the oxygen-exposed surface and a beam-induced desorption of Mo-oxide molecules.

5102. A molecular dynamics simulation study of the influence of the ionatom potential function upon sputtering: Don E Harrison, JrJ Appl Phys,52 (3), 1981, 1499–1508

The sputtering yield of molybdenum and angular distribution of sputtered molybdenum under oxygen exposure have been measured for D+- and He+-irradiation in the energy region from 2.5 keV to 6 keV at temperatures ranging from ambient room temperature to 640°C. The components of the sputtered particles from oxygen-exposed molybdenum have also been measured for 8 keV Ar+-irradiation at temperatures of 25°C and 485°C. The yields were determined from the weight loss of the targets and by the aid of the collector method which was applied to determine the angular emission distribution. Components of the sputtered particles were determined by means of the mass analysis of emitted neutral and charged particles. It was found that the oxygen exposure dependence of the yield strongly depends on the incident ions and on the temperature. The results are interpreted in terms of the chemical reactivity of the incident ions with oxygen, a dilution of metal atoms in the oxygen-exposed surface and a beam-induced desorption of Mo-oxide molecules.

Electron-Paramagnetic-Resonance Study of Clean and Oxygen-Exposed Surfaces of GaAs, AlSb, and Other III-V Compounds

Clean surfaces of $p$-type GaP, GaAs, GaSb, and InAs and $n$-type GaAs, AlSb, and GaSb have been prepared by crushing in ultrahigh vacuum (${10}^{\ensuremath{-}9}$ Torr) and measured by the electron-paramagnetic-resonance technique at room temperature and 77\ifmmode^\circ\else\textdegree\fi{}K. A small clean-surface signal was found. When oxygen was adsorbed at 77\ifmmode^\circ\else\textdegree\fi{}K, a new signal was found, due to $\mathrm{O}_{2}^{\ensuremath{-}}$ ions. In the case of AlSb, hyperfine structure was resolved. A complete analysis was carried out without the usual restriction of setting the $\stackrel{\ensuremath{\leftrightarrow}}{\mathrm{A}}$ (hyperfine) and $\stackrel{\ensuremath{\leftrightarrow}}{\mathrm{g}}$ tensors parallel. Parameters were checked by computer simulations of the spectra. About 5% of the wave function of the unpaired electron on the $\mathrm{O}_{2}^{\ensuremath{-}}$ molecule is localized on the Al atom. The unfilled (dangling) orbital of this surface atom is found to be over 90% $p$ type. This provides, apparently, the first experimental determination of a clean-semiconductor-surface wave function. A model of the (110) surface of a III-V compound semiconductor is proposed. The group-V atom dangling bond contains an electron pair in a $p$ orbital, while the group-III atom dangling bond is a largely empty $p$ orbital. In-surface bonds are largely $s{p}^{2}$. Parameters for the clean-surface signal on GaAs and GaSb are $g=2.0038\ifmmode\pm\else\textpm\fi{}0.0004$, width 10 \ifmmode\pm\else\textpm\fi{} 1 G, corresponding to about 4 \ifmmode\times\else\texttimes\fi{} ${10}^{11}$ spins ${\mathrm{cm}}^{\ensuremath{-}2}$, reduced by about 30% after exposure to air. For the $\mathrm{O}_{2}^{\ensuremath{-}}$ signal on GaAs ${g}_{\ensuremath{\parallel}}=2.036$, ${g}_{\ensuremath{\perp}}=2.007$, and on AlSb ${g}_{\ensuremath{\parallel}}=2.041$, ${g}_{\ensuremath{\perp}}=2.005$.