Soft X-ray appearance potential spectroscopy (SXAPS) study of TiO2(1 1 0)-1 × 2 and (0 0 1)-1 × 1 surfaces

A soft X-ray appearance potential spectroscopy (SXAPS) apparatus with high sensitivity was built to measure non-derivative spectra. SXAPS spectra (non-derivative) of Ti 2p and O 1s for TiO₂(1 1 0)-1 × 2 and (0 0 1)-1 × 1 surfaces have been measured using low incident currents (about 10 μA/cm²) and a photon counting mode. Density of empty states on Ti and O sites are deduced by self-deconvoluting the spectra. The self-deconvoluted SXAPS spectra are qualitatively similar to those measured by X-ray absorption spectroscopy (XAS). The Ti 2p_3/2 spectrum shows two strong peaks which correspond to t_2g and e_g states. For the O 1s spectrum two strong peaks near the threshold are also found which can be ascribed to O 2pπ and O 2pσ states. These results suggest that the spectra almost obey the dipole selection rule, so-called the “approximate dipole selection rule”. The SXAPS spectra of Ti 2p and O 1s for the (1 1 0) and (0 0 1) surfaces resemble qualitatively, which is consistent with the XAS results. The spectra measured on the (1 1 0)-1 × 2 surface at an incident angle of 45° off normal to the surface and on the (1 1 0) surface sputtered by Ar ions indicate that SXAPS is very sensitive to the surface electronic states.     

Composition and structure of chemically prepared GaAs(1 1 1)A and (1 1 1)B surfaces

The (1 1 1)A and (1 1 1)B surfaces of GaAs chemically treated in HCl-isopropanol solution (HCl-iPA) and annealed in vacuum were studied by means of X-ray photoelectron spectroscopy (XPS), low-energy electron diffraction (LEED) and electron energy loss spectroscopy (EELS). To avoid uncontrolled contamination, chemical treatment and sample transfer into UHV were performed under pure nitrogen atmosphere. The HCl-iPA treatment removes gallium and arsenic oxides, with about 0.5-3 ML of elemental arsenic being left on the surface, depending on the crystallographic orientation. With the increase of the annealing temperature, a sequence of reconstructions were identified by LEED: (1 × 1) and (2 × 2) on the (1 1 1)A surface and (1 × 1), (2 × 2), (1 × 1), (3 × 3), (√19 × √19) on the (1 1 1)B surface. These sequences of reconstructions correspond to the decrease of surface As concentration. The structural properties of chemically prepared GaAs(1 1 1) surfaces were found to be similar to those obtained by decapping of As-capped epitaxial layers.     

Surface structure of K/Pd(1 0 0) and the effect of H coadsorption: Density functional theory calculations

We have performed density-functional theory calculations to study the atomic structure of the K/Pd(1 0 0)-p(2 × 2) and -c(2 × 2) surfaces formed at 0.25 ML and 0.5 ML, respectively. We find that K atoms prefer the hollow site with the K adsorption height 2.44 Å for p(2 × 2) and 2.50 Å c(2 × 2). The first interlayer spacing (d₁₂) of the Pd(1 0 0) substrate appears slightly contracted from the bulk value as Δd₁₂ = −0.8% and −0.3% for p(2 × 2) and c(2 × 2), respectively. The calculated contraction Δd₁₂ = −0.3% for c(2 × 2) is not in accord with the expansion Δd₁₂ = +1.3% reported by a low-energy electron diffraction (LEED) study. As the origin of this difference, a possibility of hydrogen contamination of the surface sample used in the LEED study is suggested: Our calculations show that the d₁₂ of K/Pd(1 0 0)-c(2 × 2) increases linearly with the coverage of H coadsorption, which leads to an estimation for the H coverage of the surface sample as 0.1-0.4 ML.     

Anisotropy of electron structure at InAs(1 1 1) surfaces by laser pump-and-probe photoemission spectroscopy

The electronic structure and the electron dynamics of the clean InAs(1 1 1)A 2 × 2 and the InAs(1 1 1)B 1 × 1 surfaces have been studied by laser pump-and-probe photoemission spectroscopy. Normally unpopulated electron states above the valence band maximum (VBM) are filled on the InAs(1 1 1)A surface due to the conduction band pinning above the Fermi level (E_F). Accompanied by the downward band banding alignment, a charge accumulation layer is confined to the surface region creating a two dimensional electron gas (2DEG). The decay of the photoexcited carriers above the conduction band minimum (CBM) is originated by bulk states affected by the presence of the surface. No occupied states were found on the InAs(1 1 1)B 1 × 1 surface. This fact is suggested to be due to the surface stabilisation by the charge removal from the surface into the bulk. The weak photoemission intensity above the VBM on the (1 1 1)B surface is attributed to electron states trapped by surface defects. The fast decay of the photoexcited electron states on the (1 1 1)A and the (1 1 1)B surfaces was found to be τ_1 1 1 A ? 5 ps and τ_1 1 1 B ?  4 ps, respectively. We suggest the diffusion of the hot electrons into the bulk is the decay mechanism.     

On the preparation and electronic properties of clean W(1 1 0) surfaces

We have studied the influence of oxygen pressure during the cyclic annealing used for the cleaning of W(1 1 0) surfaces. For this purpose the surface morphology and electronic properties are measured by means of scanning tunneling microscopy (STM) and spectroscopy (STS), respectively. It is found that the surfaces with impurity atom densities as low as 2 × 10⁻³ can be obtained by gradually reducing the oxygen pressure between subsequent annealing cycles down to about 2 × 10⁻⁸ mbar in the final cycle. Only on the clean surface a bias-dependent spatial modulation of the local density of states (LDOS) is observed at step edges and around impurity sites by STS. In addition, we find a pronounced peak in the occupied states. In combination with density functional theory calculations these features can be traced back to a dispersive p_z-d_xz-type surface resonance band and the lower band edge of a surface state, respectively.     

Theoretical studies of electronic states and phonon modes on the TiC(0 0 1)(1 × 1) surface

We present a comprehensive picture of structural and electronic properties of the TiC(0 0 1)(1 × 1) surface. Our investigations are based on first-principles calculations within the local-density approximation of the density-functional theory. Good agreement has been observed between our calculation and experimental data for the atomic geometry of the surface. In particular, the calculated bond lengths between the first-layer C and the second-layer Ti (d_1C-2Ti = 2.188 Å) and between the first-layer Ti and the second-layer C (d_1Ti-2C = 2.031 Å) are in good agreement with the corresponding experimental values of 2.25 Å and 2.14 Å, respectively. We have also identified surface electronic states and provided clear support for previously available photoemission measurements. We have further calculated surface phonon modes at the zone centre and at the zone-edge point X using a linear response scheme based on the ab initio pseudopotential method. Our calculated surface phonon results are in excellent agreement with electron energy loss spectroscopy results.     

Chemically prepared well-ordered InP(0 0 1) surfaces

In the present work HCl-isopropanol treated and vacuum annealed InP(0 0 1) surfaces were studied by means of low-energy electron diffraction (LEED), soft X-ray photoemission (SXPS), and reflectance anisotropy (RAS) spectroscopies. The treatment removes the natural oxide and leaves on the surface a physisorbed overlayer containing InCl_x and phosphorus. Annealing at 230 °C induces desorption of InCl_x overlayer and reveals a P-rich (2 × 1) surface. Subsequent annealing at higher temperature induces In-rich (2 × 4) surface. The structural properties of chemically prepared InP(0 0 1) surfaces were found to be similar to those obtained by decapping of As/P-capped epitaxial layers.     

CO2 adsorption on Cr(1 1 0) and Cr2O3(0 0 0 1)/Cr(1 1 0)

We attempt to correlate qualitatively the surface structure with the chemical activity for a metal surface, Cr(1 1 0), and one of its surface oxides, Cr₂O₃(0 0 0 1)/Cr(1 1 0). The kinetics and dynamics of CO₂ adsorption have been studied by low energy electron diffraction (LEED), Aug er electron spectroscopy (AES), and thermal desorption spectroscopy (TDS), as well as adsorption probability measurements conducted for impact energies of E_i = 0.1-1.1 eV and adsorption temperatures of T_s = 92-135 K. The Cr(1 1 0) surface is characterized by a square shaped LEED pattern, contamination free Cr AES, and a single dominant TDS peak (binding energy E_d = 33.3 kJ/mol, first order pre-exponential 1 × 10¹³ s⁻¹). The oxide exhibits a hexagonal shaped LEED pattern, Cr AES with an additional O-line, and two TDS peaks (E_d = 39.5 and 30.5 kJ/mol). The initial adsorption probability, S₀, is independent of T_s for both systems and decreases exponentially from 0.69 to 0.22 for Cr(1 1 0) with increasing E_i, with S₀ smaller by ∼0.15 for the surface oxide. The coverage dependence of the adsorption probability, S(Θ), at low E_i is approx. independent of coverage (Kisliuk-shape) and increases initially at large E_i with coverage (adsorbate-assisted adsorption). CO₂ physisorbs on both systems and the adsorption is non-activated and precursor mediated. Monte Carlo simulations (MCS) have been used to parameterize the beam scattering data. The coverage dependence of E_d has been obtained by means of a Redhead analysis of the TDS curves.     

A valence band photoemission study of Pb adsorption on Rh(1 0 0) and Rh(1 1 0)

We have studied the adsorption of Pb on the Rh(1 0 0) and (1 1 0) surfaces by photoemission and low energy electron diffraction (LEED), and tested the chemical properties by adsorption of CO. Pb forms two distinct c(2 × 2) phases on Rh(1 0 0), according to the temperature of the substrate. The phase formed below about 570-620 K, denoted α-c(2 × 2), reduces the coverage of adsorbed CO but does not affect the valence band spectrum of the molecule. The phase formed above this temperature, denoted β-c(2 × 2), also reduces the coverage of adsorbed CO but the valence band spectrum of the adsorbed CO is strongly affected. The two phases are also characterised by a slightly different binding energy of the Pb 5d_5/2 level, 17.54 eV for the α phase and 17.70 for the β phase. The Pb/Rh(1 1 0) surface shows two ordered Pb induced phases, c(2 × 2) and p(3 × 1). CO adsorbs on the first with reduced heat of adsorption and with a valence band spectrum that is strongly altered with respect to CO adsorbed on clean Rh(1 1 0), but does not adsorb on the p(3 × 1) structure at 300 K. We compare the present results with previous results from related systems.     

Theoretical study of hydrogen adsorption on the GaN(0 0 0 1) surface

Ab initio density functional theory, using the B3LYP hybrid functional with all-electron basis sets, has been applied to the adsorption of H on the (0 0 0 1) surface of wurtzite GaN. For bulk GaN, good agreement is obtained with photoemission and X-ray emission data for the valence band and for the Ga 3d and N 2s shallow core levels. A band gap of E_g = 4.14 eV is computed vs the experimental value (at 0 K) of 3.50 eV. A simple model, consisting of a (2 × 2) structure with 3/4-monolayer (ML) of adsorbed H, is found to yield a density of states in poor agreement with photoemission data for H adsorbed on surfaces prepared by ion bombardment and annealing. A new model, consisting of co-adsorbed Ga (1/4 ML) and H (1/2 ML), is proposed to account for these data.     

The dehydrogenation of CH4 on Rh(1 1 1), Rh(1 1 0) and Rh(1 0 0) surfaces: A density functional theory study

CH₄ dehydrogenation on Rh(1 1 1), Rh(1 1 0) and Rh(1 0 0) surfaces has been investigated by using density functional theory (DFT) slab calculations. On the basis of energy analysis, the preferred adsorption sites of CH_x (x = 0-4) and H species on Rh(1 1 1), Rh(1 1 0) and Rh(1 0 0) surfaces are located, respectively. Then, the stable co-adsorption configurations of CH_x (x = 0-3) and H are obtained. Further, the kinetic results of CH₄ dehydrogenation show that on Rh(1 1 1) and Rh(1 0 0) surfaces, CH is the most abundant species for CH₄ dissociation; on Rh(1 1 0) surface, CH₂ is the most abundant species, our results suggest that Rh catalyst can resist the carbon deposition in the CH₄ dehydrogenation. Finally, results of thermodynamic and kinetic show that CH₄ dehydrogenation on Rh(1 0 0) surface is the most preferable reaction pathway in comparison with that on Rh(1 1 1) and Rh(1 1 0) surfaces.     

The CuGaSe2(0 0 1) surface: A (4 × 1) reconstruction

We report on the formation of a stable (4 × 1) reconstruction of the chalcopyrite CuGaSe₂(0 0 1) surface. Using Ar⁺ ion-bombardment and annealing of epitaxial CuGaSe₂ films grown on GaAs(0 0 1) substrates it was possible to obtain flat, well-ordered surfaces showing a clear (4 × 1) reconstruction. The cleanliness and structure were analyzed in situ by AES and LEED. AES data suggest a slight Se-enrichment and Cu-depletion upon surface preparation. Our results demonstrate that (0 0 1) surfaces of the Cu-III-VI₂(0 0 1) material can show stable, unfacetted surfaces.     

Ruthenium adsorption and diffusion on the GaN(0 0 0 1) surface

We report first principles calculations to analyze the ruthenium adsorption and diffusion on GaN(0 0 0 1) surface in a 2×2geometry. The calculations were performed using the generalized gradient approximation (GGA) with ultrasoft pseudopotential within the density functional theory (DFT). The surface is modeled using the repeated slabs approach. To study the most favorable ruthenium adsorption model we considered T₁, T₄ and H₃ special sites. We find that the most energetically favorable structure corresponds to the Ru- T₄ model or the ruthenium adatom located at the T₄ site, while the ruthenium adsorption on top of a gallium atom (T₁ position) is totally unfavorable. The ruthenium diffusion on surface shows an energy barrier of 0.612 eV. The resultant reconstruction of the ruthenium adsorption on GaN(0 0 0 1)- 2×2 surface presents a lateral relaxation of some hundredth of Å in the most stable site. The comparison of the density of states and band structure of the GaN(0 0 0 1) surface without ruthenium adatom and with ruthenium adatom is analyzed in detail.     

Ground-state properties of arsenic deposited on the Ge(0 0 1)(2 × 1) surface

We present results of ab initio calculations of structural, electronic and vibrational properties of the Ge(0 0 1) surface covered with a monolayer of arsenic. The fully occupied π_u bonding and π_g antibonding electronic states due to the As-As dimer formation are quite close in energy and their ordering is same as that found on the Si(0 0 1) surface. Using our calculated atomic and electronic structures, surface lattice dynamics was studied by employing a linear response approach based on density functional perturbation theory. A comparison of the phonon spectrum of the Ge(0 0 1)/As(2 × 1) surface with that of the clean Ge(0 0 1)(2 × 1) surface indicates the presence of several new characteristic phonon modes due to adsorption of As atoms.     

Growth mode and electronic structure of Au nano-clusters on NiO(0 0 1) and TiO2(1 1 0)

The growth mode and electronic structure of Au nano-clusters grown on NiO and TiO₂ were analyzed by reflection high-energy electron diffraction, a field-emission type scanning electron microscope, medium energy ion scattering and photoelectron spectroscopy. Au was deposited on clean NiO(0 0 1)-1 × 1 and TiO₂(1 1 0)-1 × 1 surfaces at room temperature with a Knudsen cell at a rate of 0.25-0.35 ML/min (1 ML = 1.39 × 10¹⁵ atoms/cm²:Au(1 1 1)). Initially two-dimensional (2D) islands with thickness of one Au-atom layer grow epitaxially on NiO(0 0 1) and then neighboring 2D-islands link each other to form three-dimensional (3D)-islands with the c-axis oriented to the [1 1 1] direction. The critical size to form 3D-islands is estimated to be about 5 nm². The shape of the 3D-islands is well approximated by a partial sphere with a diameter d and height h ranging from 2.0 to 11.8 nm and from 0.95 to 4.2 nm, respectively for Au coverage from 0.13 to 4.6 ML. The valence band spectra show that the Au/NiO and Au/TiO₂ surfaces have metallic characters for Au coverage above 0.9 ML. We observed Au 4f spectra and found no binding energy shift for Au/NiO but significant higher binding energy shifts for Au/TiO₂ due to an electron charge transfer from Au to TiO₂. The work function of Au/NiO(0 0 1) gradually increases with increase in Au coverage from 4.4 eV (NiO(0 0 1)) to 5.36 eV (Au(1 1 1)). In contrast, a small Au deposition(0.15 to 1.5 ML) on TiO₂(1 1 0) leads to reduction of the work function, which is correlated with an electron charge transfer from Au to TiO₂ substrate.     

Oxidation of a cesium-covered Ni (1 1 0) surface studied by metastable-induced electron spectroscopy

An initial stage of oxidation of a cesium-covered Ni (1 1 0) surface has been studied by metastable-induced electron spectroscopy (MIES) and low-energy electron diffraction (LEED). The MIES brought spectra with Cs 6s induced peak (P_6s), Cs 5p (P_5p), O 2p induced peak (P_ox) and a structure related to the substrate Ni 3d states (P_3d). The work function change Δφ showed an oscillatory behavior in the progress of surface oxidation. The process is divided into three stages: (i) at low O₂ exposures, Δφ > 0 with unchanging P_5p and P_6s; (ii) at moderate exposures, Δφ < 0 with a drastic decrease in the P_6s intensity; (iii) at higher exposures, Δφ > 0 with shifts of peaks P_5p and P_ox to higher energies, together with an appearance of peak P_3d. A three-step model of initial oxidation of alkali-covered Ni (1 1 0) surfaces is presented.     

Metal organic chemical vapor deposition of ultrathin ZrO2 films on Si(1 0 0) and Si(1 1 1) studied by electron spectroscopy

The growth of ultrathin ZrO₂ films on Si(1 0 0)-(2 × 1) and Si(1 1 1)-(7 × 7) has been studied with core level photoelectron spectroscopy and X-ray absorption spectroscopy. The films were deposited sequentially by chemical vapor deposition in ultra-high vacuum using zirconium tetra-tert-butoxide as precursor. Deposition of a > 50 Å thick film leads in both cases to tetragonal ZrO₂ (t-ZrO₂), whereas significant differences are found for thinner films. On Si(1 1 1)-(7 × 7) the local structure of t-ZrO₂ is not observed until a film thickness of 51 Å is reached. On Si(1 0 0)-(2 × 1) the local geometric structure of t-ZrO₂ is formed already at a film thickness of 11 Å. The higher tendency for the formation of t-ZrO₂ on Si(1 0 0) is discussed in terms of Zr-O valence electron matching to the number of dangling bonds per surface Si atom. The Zr-O hybridization within the ZrO₂ unit depends furthermore on the chemical composition of the surrounding. The precursor t-butoxy ligands undergo efficient C-O scission on Si(1 0 0), leaving carbonaceous fragments embedded in the interfacial layer. In contrast, after small deposits on Si(1 1 1) stable t-butoxy groups are found. These are consumed upon further deposition. Stable methyl and, possibly, also hydroxyl groups are found on both surfaces within a wide film thickness range.     

Density functional study of hypophosphite adsorption on Ni (1 1 1) and Cu (1 1 1) surfaces

Surface structures and electronic properties of hypophosphite, H₂PO₂⁻, molecularly adsorbed on Ni(1 1 1) and Cu(1 1 1) surfaces are investigated in this work by density functional theory at B3LYP/6-31++g(d, p) level. We employ a four-metal-atom cluster as the simplified model for the surface and have fully optimized the geometry and orientation of H₂PO₂⁻ on the metal cluster. Six stable orientations have been discovered on both Ni (1 1 1) and Cu (1 1 1) surfaces. The most stable orientation of H₂PO₂⁻ was found to have its two oxygen atoms interact the surface with two PO bonds pointing downward. Results of the Mulliken population analysis showed that the back donation from 3d orbitals of the transition metal substrate to the unfilled 3d orbital of the phosphorus atom in H₂PO₂⁻ and 4s orbital's acceptance of electron donation from one lone pair of the oxygen atom in H₂PO₂⁻ play very important roles in the H₂PO₂⁻ adsorption on the transition metals. The averaged electron configuration of Ni in Ni₄ cluster is 4s^0.634p^0.023d^9.35 and that of Cu in Cu₄ cluster is 4s^1.004p^0.033d^9.97. Because of this subtle difference of electron configuration, the adsorption energy is larger on the Ni surface than on the Cu surface. The amount of charge transfers due to above two donations is larger from H₂PO₂⁻ to the Ni surface than to the Cu surface, leading to a more positively charged P atom in Ni_nH₂PO₂⁻ than in Cu_nH₂PO₂⁻. These results indicate that the phosphorus atom in Ni_nH₂PO₂⁻ complex is easier to be attacked by a nucleophile such as OH⁻ and subsequent oxidation of H₂PO₂⁻ can take place more favorably on Ni substrate than on Cu substrate.     

Adsorption of water on TiN (1 0 0), (1 1 0) and (1 1 1) surfaces: A first-principles study

We use first-principles density functional theory-based calculations in the analysis of the interaction of H₂O with (1 0 0), (1 1 0) and (1 1 1) surfaces of TiN, and develop understanding in terms of surface energies, polarity of the surface and chemistry of the cation, through comparison with H₂O adsorption on ZrN. While water molecule physisorbs preferentially at Ti site of (1 0 0) and (1 1 1) surfaces, it adsorbs dissociatively on (1 1 0) surface of TiN with binding stronger than almost 1.32 eV/molecule. Our analysis reveals the following general trends: (a) surfaces with higher energies typically lead to stronger adsorption, (b) dissociative adsorption of H₂O necessarily occurs on a charge neutral high energy surface and (c) lower symmetry of the (1 1 0) plane results in many configurations of comparable stability, as opposed to the higher symmetry (1 0 0) and (1 1 1) surfaces, which also consistently explain the results of H₂O adsorption on MgO available in literature. Finally, weaker adsorption of H₂O on TiN than on ZrN can be rationalized in terms of greater chemical stability of Ti arising from its ability to be in mixed valence.     

Core-level shifts of the c(8 × 2)-reconstructed InAs(1 0 0) and InSb(1 0 0) surfaces

We have studied In-stabilized c(8 × 2)-reconstructed InAs(1 0 0) and InSb(1 0 0) semiconductor surfaces, which play a key role in growing improved III–V interfaces for electronics devices, by core-level photoelectron spectroscopy and first-principles calculations. The calculated surface core-level shifts (SCLSs) for the ζ and ζa models, which have been previously established to describe the atomic structures of the III–V(1 0 0)c(8 × 2) surfaces, yield hitherto not reported interpretation for the As 3d, In 4d, and Sb 4d core-level spectra of the III–V(1 0 0)c(8 × 2) surfaces, concerning the number and origins of SCLSs. The fitting analysis of the measured spectra with the calculated ζ and ζa SCLS values shows that the InSb spectra are reproduced by the ζ SCLSs better than by the ζa SCLSs. Interestingly, the ζa fits agree better with the InAs spectra than the ζ fits do, indicating that the ζa model describes the InAs surface better than the InSb surface. These results are in agreement with previous X-ray diffraction data. Furthermore, an introduction of the complete-screening model, which includes both the initial and final state effects, does not improve the fitting of the InSb spectra, proposing the suitability of the initial-state model for the SCLSs of the III–V(1 0 0)c(8 × 2) surfaces. The found SCLSs are discussed with the ab initio on-site charges.