Model and theory for the determination of diffusion coefficients by Auger electron spectroscopy measurements and an application to oxygen diffusion along the [0001] and [101¯0] axes in single crystal zirconium

A simple hopping model of the diffusion of adsorbed species from a surface into the bulk of a material has been formulated and solved mathematically. The difference in the energy barriers for an atom moving between the atomic layers at the surface and in the bulk are explicitly considered. This model is also capable of describing the initial stages of diffusion, something that conventional solutions of the continuum diffusion equation cannot handle. Auger electron spectroscopy has been used to measure the dissolution rate of oxygen from Zr(0001) and Zr(101¯0) surface into the bulk. Satisfactory results were obtained by applying our model to the diffusion data for these two zirconium surfaces for two different heating schedules: (i) rapid temperature ramp-and-hold and (ii) continuous linear heating with respect to time. The resulting Arrhenius expressions for diffusion are: D = (0.115 ± 0.031)exp[(−44.45 ± 4.82)kcal/RT]cm²/s along Zr[0001] and D = (1.07 ± 0.26)exp[(−46.18 ± 4.22)kcal/RT]cm²/s along Zr[101¯0].     

Enhanced reactivity and selectivity in oxidation of Cu(1 0 0) and α-Cu-Al(5 at.%)(1 0 0) surfaces studied by electron and ion spectroscopies

In this study we investigate the influence of alloying on the reactivity and bonding of oxygen on α-Cu-Al(5 at.%)(1 0 0) oriented single crystal surfaces by X-ray photoelectron spectroscopy (XPS), ultra-violet spectroscopy (UPS) and low energy ion scattering (LEIS) spectroscopy, at room temperature. It was found that alloying results in an enhanced reactivity of both Cu and Al sites in comparison with the pure metals. According to adsorption curves calculated from XPS, saturation of the alloy surface occurs for exposures of ∼15 L. At saturation the total amount of adsorbed oxygen is similar for the alloy and pure copper surfaces. It was determined that first mostly Al sites are oxidized, followed by simultaneous oxidation of Cu and Al sites. At saturation the amount of oxygen bonded to Cu sites is ∼1.7 larger then that bonded to Al sites. From a comparison of the XPS and LEIS data analysis as a function of oxygen exposure it was found that oxidation of α-Cu-Al(5 at.%)(1 0 0) alloy is a multi-stage process with fast and slow stages. These stages involve an interplay of chemisorption, sub-surface diffusion of oxygen and Al segregation. UPS measurements show an increase in the work function of the alloy surface with oxygen adsorption. This is a contrast to pure Cu surfaces where the work function decreases at the initial stages of oxidation followed by an increase with oxygen exposure. Annealing to 400 °C drives the oxidized alloy surface into its thermodynamic state resulting in the formation of an aluminum oxide layer. Possible mechanisms to explain the enhanced reactivity of the alloy surface compared to that of pure copper are suggested and discussed.     

Oxidation of the hydrogenated diamond (100) surface

The surface composition and structure of natural diamond (100) surfaces subsequently oxidized with activated oxygen at T_sub≤35°C were investigated with high-resolution electron energy loss spectroscopy (HREELS), Auger electron spectroscopy, electron loss spectroscopy (ELS) and low-energy electron diffraction (LEED). Complete surface oxidation (oxygen coverage θ=1 ML) required doses of hundreds of kilolangmuirs of O₂. HREELS vibrational spectra permitted identification of the specific surface oxygen species, and also provided information about the diamond surface states. Most surface sites lost their hydrogen at least once before becoming oxidized. The oxygen coverage θ increased quickly at first, and then more slowly as saturation was approached; different mechanisms or sites may have accounted for the decreased rate. The relative distribution of oxygen species varied with the oxidation conditions. Ether, carbonyl and hydroxyl groups appeared during the initial stages of oxidation, but the hydroxyl groups disappeared at higher coverages. Bridge-bonded ether groups dominated at saturation coverage, although smaller amounts of carbonyl and hydroxyl were still observed. The carbonyl and C---H stretch frequencies increased with oxygen dose due to formation of higher oxidation states and/or hydrogen bonding between adjacent groups. ELS revealed only a low concentration of C=C dimers on the oxidized surfaces, and no evidence of graphitization.

Surfaces generated by oxygen addition and then desorption were more reactive than surfaces generated by hydrogen desorption. Oxidized surfaces that were heated in vacuum and then rehydrogenated did not recover the sharp LEED patterns and HREELS spectra of the original plasma-smoothed surface. This effect was presumably due to surface roughening caused by oxygen desorption as CO and CO₂, and creation of reactive high-energy sites that quickly bonded to available background gases and prevented large areas of organized surface reconstruction.     

Substitutional adsorption of K on Fe(100)

The adsorption of potassium on Fe(100) was studied by time-of-flight forward scattering and recoiling spectroscopy (TOF-SARS), low energy electron diffraction (LEED) and Auger electron spectroscopy (AES). After heating to 650 K of the potassium saturated surface the formation of a p(3 × 3) potassium superstructure was observed by LEED. TOF-SARS experiments ruled out the adsorption of potassium in the on-top, bridge and four-fold hollow site. The only site which is in agreement with all experimental results is the substitutional site where K replaces an Fe atom of the topmost layer of the crystal. This is the first time a substitutional adsorption site has been found on a bcc surface. On an fcc surface such an adsorption site has been found recently for adsorption of sodium and potassium on Al(111).     

Epitaxial growth of bcc Fe films on Al(1 0 0) surfaces using Ti as an interface stabilizer

An approach is described to promote epitaxial growth of thin metal films on single-crystal metal substrates by stabilizing the interface with an extremely thin metallic interlayer. A single atomic layer of a metal is deposited at the interface, Ti on Al(1 0 0) in this case, prior to the growth of the metal film of interest to produce an epitaxial interface in a system that is otherwise characterized by interdiffusion and disorder. The stabilized interface reduces interdiffusion and serves as a template for ordered film growth. Using Rutherford backscattering and channeling techniques along with low-energy electron diffraction and low-energy He⁺ scattering, it is demonstrated that an atomically thin layer of Ti metal deposited at the Fe-Al interface, a system well known for considerable intermixing at room temperature, reduces interdiffusion and promotes the epitaxial growth of Fe films on the Al(1 0 0) surface. The decrease in ion scattering yield for Al atoms, Fe-Fe shadowing and long-range order of the surface suggest that the epitaxial growth of Fe is greatly improved when the Ti interlayer is introduced prior to Fe deposition. Off-normal ion channeling experiments provide clear evidence for the bcc structure of Fe on the Ti/Al(1 0 0) template with the measured average (1 0 0) interplanar distance of 1.44 Å in the Fe overlayer.     

Oxidation of heated diamond C(100):H surfaces

This paper extends a previous study (Pehrsson and Mercer, submitted to Surf. Sci.) on unheated, hydrogenated, natural diamond (100) surfaces oxidized with thermally activated oxygen (O^*₂). In this paper, the oxidation is performed at substrate temperatures from T_sub=24 to 670°C. The diamond surface composition and structure were then investigated with high resolution electron energy loss spectroscopy (HREELS), Auger electron spectroscopy (AES), electron loss spectroscopy (ELS) and low energy electron diffraction (LEED).

The oxygen coverage (θ) increased in two stages, as it did during oxidation at T<80°C. However, there are fundamental differences between the oxidation of nominally unheated and heated diamond surfaces. This difference is attributed to simultaneous adsorption and rapid desorption of oxygen species at higher temperatures; the desorption step is much slower without heating. The initial oxidation rates were similar regardless of the substrate temperatures, but the peak coverage (θ) was lower at higher temperatures. For example, θ plateaued at 0.4±0.1 ML at 600°C. The lower saturation coverage is again attributed to oxygen desorption during oxidation. Consistent results were obtained on fully oxidized surfaces, which when heated in vacuum to T_sub=600°C, lost 60% of their adsorbed oxygen. ELS revealed few C=C dimers on the oxidized surfaces, and more graphitization than on unheated surfaces. Oxidation at elevated temperatures also increased the carbonyl to ether ratio, reflecting etching-induced changes in the types of surface sites. The carbonyl and C–H stretch frequencies increased with oxygen dose due to formation of higher oxidation states and/or hydrogen bonding between adjacent groups. The oxygen types did not interconvert when the oxidized surfaces were heated in vacuum. Oxygen desorption generated a much more reactive surface than heating-induced dehydrogenation of the smooth, hydrogenated surface.     

The interaction of D2O with Zr(0001) at 80 K

The adsorption of D₂O on Zr(0001) at 80 K and its subsequent reactions at higher temperatures have been studied by thermal desorption spectroscopy (TDS), work-function measurements (Δф), nuclear reaction analysis (NRA), LEED, infrared reflection spectroscopy (FTIR-RAS), Auger electron spectroscopy (AES), and static secondary ion mass spectroscopy (SSIMS). D₂O adsorption on Zr(0001) at 80 K is accompanied by a Δф of −1.33 eV. The adsorbed D₂O can be characterized into three layers by TDS: a chemisorbed layer (up to 0.23 ML), a second adsorbed layer, and an ice layer. The chemisorbed D₂O dissociates into OD_ad and D_ad at 80 K (possibly also into O_ad) and no desorption products could be detected, implying that the reaction products dissolved into the zirconium at temperatures appropriate for each component. The ice layer and most of the second adsorbed layer desorb as molecular water during heating. The water adsorbed at 80 K did not form any long-range ordered structure, but a (2 × 2) LEED pattern that was formed by heating the sample to temperatures above 430 K is believed due to be an ordered oxygen superstructure.     

Oxygen induced segregation of aluminum to α-Cu-Al(1 0 0) alloy surfaces studied by low energy ion scattering and X-ray photoelectron spectroscopy

The effect of annealing temperature on the surface composition of α-Cu-Al(1 0 0) alloys for aluminum concentrations of 5, 12 and 17 at% was investigated using X-ray photoelectron spectroscopy (XPS) and low energy ion scattering (LEIS). Two initial states of the sample surfaces were examined: sputter-cleaned and oxidized. The effect of annealing temperature on segregation is different for sputter-cleaned and oxidized samples. Aluminum preferential sputtering and strong oxygen induced aluminum segregation were detected on all examined samples. Whilst for the sputter-cleaned surfaces a small thermal induced segregation was observed, the combination of annealing and oxygen exposure resulted in aluminum enrichment in the 100-300% range relative to the bulk concentration. The segregation rate is proportional to the aluminum concentration for sputter-cleaned surfaces and displays a maximum for the oxidized α-Cu-Al(12 at.%)(1 0 0) surface.     

Segregation of Pt at clean surfaces of (Pt, Ni)3Al

Experimental evidence for surface segregation of Pt at (1 1 1) surfaces of ternary (Pt, Ni)₃Al alloys is presented, based upon Auger electron spectroscopy, low energy ion scattering, and angle-resolved X-ray photoelectron spectroscopy. Density functional calculations in the dilute limit confirm that Pt segregation is energetically favored.     

Photoemission studies of the interactions of CdTe and Te with Si(100)

The interactions between CdTe, and in particular Te, and the (100) surface of Si have been probed using photoemission and low energy electron diffraction with a view to investigating the mechanisms responsible for (100) and (111) growth orientations for CdTe on Si(100). The interfacial reactions have been studied both on room temperature deposition followed by annealing and on depositions at typical epitaxial growth temperatures. In both cases the same precursor stage of an ordered submonolayer of Te on the Si(100) surface has been identified. Line shape analysis of the Si 2p core level has suggested a structural model in which Te adatoms make up an incomplete monolayer bound in bridge sites. This model is in excellent agreement both with the (1 × 1) LEED pattern and recent SEXAFS studies of this surface. The implications of the cubic symmetry of this surface in terms of the subsequent growth orientation of CdTe are discussed. Termination of the surface by Te was also seen to induce band bending suggestive of Fermi level pinning at around midgap, in contrast to the passivating behaviour of other group VI elements on this surface. The Si 2p core level line shape analysis on termination by Te has also provided evidence to support the “covalent dimer” interpretation of the clean dimerised Si(100) surface.     

Adsorption of oxygen on Au(111) by exposure to ozone

Atomic oxygen coverages of up to 1.2 ML may be cleanly adsorbed on the Au(111) surface by exposure to O₃ at 300 K. We have studied the adsorbed oxygen layer by AES, XPS, HREELS, LEED, work function measurements and TPD. A plot of the O(519 eV)/Au(239 eV) AES ratio versus coverage is nearly linear, but a small change in slope occurs at Θ_O=0.9 ML. LEED observations show no ordered superlattice for the oxygen overlayer for any coverage studied. One-dimensional ordering of the adlayer occurs at low coverages, and disordering of the substrate occurs at higher coverages. Adsorption of 1.0 ML of oxygen on Au(111) increases the work function by +0.80 eV, indicating electron transfer from the Au substrate into an oxygen adlayer. The O(1s) peak in XPS has a binding energy of 530.1 eV, showing only a small (0.3 eV) shift to a higher binding energy with increasing oxygen coverage. No shift was detected for the Au 4f_7/2 peak due to adsorption. All oxygen is removed by thermal desorption of O₂ to leave a clean Au(111) surface after heating to 600 K. TPD spectra initially show an O₂ desorption peak at 520 K at low Θ_O, and the peak shifts to higher temperatures for increasing oxygen coverages up to Θ_O=0.22 ML. Above this coverage, the peak shifts very slightly to higher temperatures, resulting in a peak at 550 K at Θ_O=1.2 ML. Analysis of the TPD data indicates that the desorption of O₂ from Au(111) can be described by first-order kinetics with an activation energy for O₂ desorption of 30 kcal mol⁻¹ near saturation coverage. We estimate a value for the Au–O bond dissociation energy D(Au–O) to be 56 kcal mol⁻¹.     

Herringbone and triangular patterns of dislocations in Ag, Au, and AgAu alloy films on Ru(0 0 0 1)

We have studied the dislocation structures that occur in films of Ag, Au, and Ag_0.5Au_0.5 alloy on a Ru(0 0 0 1) substrate. Monolayer (ML) films form herringbone phases while films two or more layers thick contain triangular patterns of dislocations. We use scanning tunneling microscopy (STM) and low-energy electron diffraction (LEED) to determine how the film composition affects the structure and periodicity of these ordered structures. One layer of Ag forms two different herringbone phases depending on the exact Ag coverage and temperature. Low-energy electron microscopy (LEEM) establishes that a reversible, first-order phase transition occurs between these two phases at a certain temperature. We critically compare our 1 ML Ag structures to conflicting results from an X-ray scattering study [H. Zajonz et al., Phys. Rev. B 67 (2003) 155417]. Unlike Ag, the herringbone phases of Au and AgAu alloy are independent of the exact film coverage. For two layer films in all three systems, none of the dislocations in the triangular networks thread into the second film layer. In all three systems, the in-plane atomic spacing of the second film layer is nearly the same as in the bulk. Film composition does, however, affect the details of the two layer structures. Ag and Au films form interconnected networks of dislocations, which we refer to as “trigons.” In 2 ML AgAu alloy, the dislocations form a different triangular network that shares features of both trigon and moiré structures. Yet another well-ordered structure, with square symmetry, forms at the boundaries of translational trigon domains in 2 ML Ag films but not in Au films.     

In situ XPS study of Pd(1 1 1) oxidation at elevated pressure, Part 2: Palladium oxidation in the 10 mbar range

The oxidation of the Pd(1 1 1) surface was studied by in situ XPS during heating and cooling in 0.4 mbar O₂. The in situ XPS data were complemented by ex situ TPD results. A number of oxygen species and oxidation states of palladium were observed in situ and ex situ. At 430 K, the Pd(1 1 1) surface was covered by a 2D oxide and by a supersaturated O_ads layer. The supersaturated O_ads layer transforms into the Pd₅O₄ phase upon heating and disappears completely at approximately 470 K. Simultaneously, small clusters of PdO, PdO seeds, are formed. Above 655 K, the bulk PdO phase appears and this phase decomposes completely at 815 K. Decomposition of the bulk oxide is followed by oxygen dissolution in the near-surface region and in the bulk. The oxygen species dissolved in the bulk is more favoured at high temperatures because oxygen cannot accumulate in the near-surface region and diffusion shifts the equilibrium towards the bulk species. The saturation of the bulk “reservoir” with oxygen leads to increasing the uptake of the near-surface region species. Surprisingly, the bulk PdO phase does not form during cooling in 0.4 mbar O₂, but the Pd₅O₄ phase appears below 745 K. This is proposed to be due to a kinetic limitation of PdO formation because at high temperature the rate of PdO seed formation is compatible with the rate of decomposition.     

Photoelectron spectroscopy study of the K-covered ZnO(1 0 1&#x0304; 0) surface; annealing-induced changes in the electronic structure and the chemical composition

The electronic structure and the chemical composition of the K-covered ZnO(1 0 1&#x0304; 0) surface at temperatures between 300 and 1200 K are investigated by X-ray photoelectron spectroscopy and ultraviolet photoelectron spectroscopy. Adsorption of K on ZnO(1 0 1&#x0304; 0) at room temperature results in the formation of a two-dimensional disordered K overlayer and induces 0.2 eV downward bending of the substrate’s bands going from the bulk to the surface. Upon annealing the K-covered surface, initial downward bending turns to upward bending with maximum bending of 0.5 eV at 700-800 K. The thermally induced migration of the bulk O atoms and the resultant increase in the number of the O atoms on the surface is responsible for upward bending on the annealed surface. The accumulated O atoms interact with the predeposited K atoms on the surface to form non-stoichiometric K-O complexes with the O/K atomic ratio being 1.6-1.8 in the temperature range between 600 and 1000 K.     

Theoretical and experimental study of N(D) formation in nitrogen collisions with a metal surface

The dynamics of nitrogen collisions with metals partially covered by alkali atoms is studied both experimentally and theoretically. Our attention focuses on the formation of N⁻(¹D) metastable ions and their interaction with the surface. We present the electron energy spectra induced by slow collisions of N⁺ ions with partially cesiated Pd(111) surfaces under grazing incidence. These spectra display, as a function of Cs coverage, a sharp feature which is due to the autodetachment of N⁻(2p⁴, ¹D) to the N(2p³, ⁴S) ground state. Our calculations, performed with the coupled angular mode (CAM) method on the basis of the resonant electron exchange between the nitrogen atom in states of the 2p³ configuration and the metal surface, consistently explain how negative ions formed close to the surface can survive against electron loss to the metal during the outgoing trajectory and can later decay as free ions. In order to understand the alkali coverage dependence of the N⁻(¹D)-N(⁴S) peak intensity, the local character of the nitrogen interaction with the surface partially covered by adsorbate atoms has been taken into account.     

The role of co-adsorbed metal atoms in the chemistry of methyl species on Pt(111) formed by the decomposition of dimethylmercury and dimethylzinc

The chemistry of methyl species resulting from the decomposition of dimethylmercury (DMM) and dimethylzinc (DMZ) on Pt(111) in the range 300–400 K has been investigated by temperature prograrnmed desorption (TPD) and Auger electron spectroscopy (AES). In each case at 300 K, dissociative adsorption of the precursor results in the formation of an adlayer of methylmetal (CH₃M) moieties. These species are thermally stable to around 350 K before decomposing to yield mainly gaseous products, methane and hydrogen, and surface bound metal atoms. For DMM, subsequent heating to 400 K or direct dissociative adsorption at 400 K results in the formation of ethylidyne species. Ethylidyne formation is not observed in the thermal chemistry of DMZ at temperatures below 400 K and only transiently in the chemistry at 400 K. Complementary TPD and AES data indicate that, for DMM, desorption of the mercury atoms produced by CH₃Hg decomposition is the limiting factor in allowing the prevailing C₁ species to couple to form ethylidyne. In contrast, AES evidence indicates that zinc atoms remain on the surface to temperatures in excess of 750 K and hence prevent C---C coupling by blocking surface sites.