Search

The reduction in the substrate temperature achieved by low-pressure diamond deposition from halogenated precursors is believed to involve surface processes including halogen adsorption, desorption, and reaction with adsorbed hydrogen. In this work we studied the adsorption of fluorine and chlorine on a diamond (100) single-crystal surface in UHV using thermal desorption spectroscopy (TDS), Auger electron spectroscopy (AES) and low-energy electron diffraction (LEED). Elemental halogens were dosed by solid-state electrolysis of either lanthanumtrifluoride or silver chloride, respectively, at elevated temperatures. The single crystal was regularly cleaned by pretreatment in a hydrogen plasma and showed clear reflexes of a (2×1) reconstructed surface in LEED. In AES, upon successive halogen dosing, a linear increase in the intensity of the halogen Auger signal and a simultaneous decrease in the intensity of the 272 eV carbon signal could be seen initially. No LEED overstructures could be observed upon adsorption on C(100). TDS shows that fluorine desorbs at temperatures above 830°C, whereas chlorine desorbs completely in the temperature range between 500 and 700°C. Halogen uptake of the diamond surface can be increased either by thermally activating the surface or by dosing of atomic hydrogen prior to halogen adsorption.

The coadsorption of H 2 O and alkalis (Li, Na and K) on a stepped Ni(s)(111) surface with nominal (775) orientation was studied by UPS (UV-light induced photoelectron spectroscopy), LEED, TDS (thermal desorption spectroscopy) and work function change measurements (Δ Φ ). On clean Ni(775) five H 2 O desorption states denoted as A ( T m =160 K), B (175 K), C (225 K), D (260 K) and E (340 K) were detected. The A and B states are also observed at similar desorption temperatures on flat Ni(111) and are therefore attributed to H 2 O adsorption on the (111) terraces of Ni(775) ( B state) or to the desorption of ice multilayers ( A state). Consequently, the C , D and E states are ascribed to the adsorption and a partial dissociation at step sites. The C state is associated with a “five peak” UPS spectrum interpreted as due to the overlap of the 1b 1 , 3a 1 and 1b 2 orbitals of molecular water and 2 σ and 1 π emission from OH groups. A preferred adsorption of the alkalis at step sites for low coverages is deduced from the Δ Φ changes. For Li adsorption the saturation of the step sites is accompanied by a significant change in the slope of the Δ Φ versus coverage curve. The increase of the effective dipole moment (less negative for low coverages, adsorption at step sites) is caused by screening of the positively charged alkalis in a step-down adsorption geometry. For the coadsorption of H 2 O and alkalis (at low alkali precoverages) only H 2 O desorption states C , D and E from step sites are influenced. The C state shifts to lower peak temperature, whereas the D and E states (recombination from dissociation products) decrease in intensity with increasing alkali coverage. Alkali coadsorption on Ni(775) is accompanied by the dissociation of water, depending on the nature and the precoverage of the alkali. The tendency to induce dissociation is highest for Li and decreases with increasing ionic radius from Na to K.

The desorption kinetics of H 2 O and D 2 O were studied on a flat Ru(0001) and a stepped Ru(1018) surface by TDS, Δo measurements (120 to 300 K) and LEED. In accordance with other authors we observed three typical desorption peaks for H 2 O on Ru(0001) (C, 155 K; A 2 , 190 K; and A 1 , 212 K), and the respective peaks for the isotope effect for D 2 O on Ru(0001), characterized by a missing A 1 peak. Isothermal Δo measurements were performed in order to determine the different desorption kinetics for H 2 O and D 2 O. Neither for D 2 O nor for H 2 O the desorption of the A 2 state can be described by first order kinetics. Especially for the A 2 peak of H 2 O, we observed an initial range of first order kinetics followed by a strong deviation, which could be described by Avrami kinetics or lateral interactions. Also, Δo measurements during desorption give evidence for a complicated adsorption kinetic process. In order to modify the water bilayer structure, a stepped Ru(1018) surface was used for comparison. These measurements are in progress and should give further information about the isotope effect.

CO 2 normally does not adsorb on atomically flat surfaces with high work functions (e.g. Cu(111)) even at lower temperatures. However, CO 2 adsorption can be achieved by activating such surfaces either by introducing steps or by alkali preadsorption. In the present paper we report about thermal desorption spectroscopic (TDS), work function change (ΔΦ) and photoelectron spectroscopic (UPS, XPS) investigations of CO 2 + K on the Cu(111) surface in the temperature range from 78 to 900 K. Our results demonstrate that the behaviour of the system can roughly be divided into two ranges depending on the amount of preadsorbed K : (1) at small to medium precoverages (0 < Θ K < 0.2, ionic K) mainly adsorbed molecular CO 2 is present, associated with a three peak UP spectrum and a desorption in TDS around 105 K. (2) At higher coverages (0.2 < Θ K < 0.38 ; metal-like K) two high temperature desorption states between 650 and 750 K are found in TDS. The UP spectrum at 78 K is more complex and arises from a mixture of molecularly bound CO 2 and a surface carbonate. The existence of two different carbon oxide species is manifested also in the O Is spectra from XPS. For Θ K < 0.2 ΔΦ increases with CO 2 adsorption. For higher coverages a decrease up to a minimum followed by an increase up to a saturation value is observed. There the coadsorptional behaviour of K and CO 2 is believed to result from a direct CO 2 -K interaction forming a surface carbonate species.

It is well established that for submonolayer coverages the strong interaction between CO + K is not only influenced by the adsorptives but also by the substrate itself. We report about investigations of CO + K on the Cu(111) surface, where the CO Cu (111) interaction is rather weak. The analytical techniques employed include thermal desorption spectroscopy, LEED and work function change measurements. We find the interaction between K + CO on Cu(111) to be very complex. Three different regimes of CO K interaction can be distinguished. (1) At very low Θ K (0.0–0.02) TDS measurements show a shift of the CO desorption maximum (for low CO coverages) towards lower temperature. ΔΦ indicates a decrease and finally a vanishing of the effective negative CO dipole moment. (2) At Θ K = 0.02−0.155 a new CO desorption maximum appears, shifting continuously from 165 to 270 K with increasing K coverage. ΔΦ indicates a positive effective dipole moment (increasing 2π∗ backdonation). Between Θ K = 0.13 and 0.18 around 300 K a very sharp and highly asymmetric CO desorption peak can be detected with no coincident K desorption. (3) At Θ K > 0.2, where the transition from “ionic-” to “metal-like” K occurs, the strong interaction between CO K is evidenced by a thermal stabilization of the adlayer to temperatures > 700 K. Under these conditions several additional desorption peaks are detected between 400 and 800 K. Desorption of CO 2 was detected under these Θ K conditions, probably due to a partial dissociation and recombination reaction.

The adsorption of atomic H on Cu(110) was studied with LEED, thermal desorption spectroscopy (TDS) and work function change measurements (Δφ) in the temperature range of 90–400 K. For the production of atomic H, molecular hydrogen was dosed through a resistivily heated Ta tube onto the Cu(110) surface. At a source temperature of 1325 K the H2 dissociation rate was > 80%. LEED as well as TDS experiments were performed to calibrate the coverage θH. From the same experiments the sticking coefficient S0 at 90 K was deduced to be 0.05 ± 0.04. The kinetics of the reconstruction reaction (1 × 3)H → (1 × 2)H at constant θH and temperatures between 100–200 K and the de-reconstruction (1 × 2)H → (1 × 1) above 250 K with the simultaneous desorption of H2 were determined from isothermal Δφ measurements. For the (1 × 2)H → (1 × 1) de-reconstruction at constant T between 257 and 283 K the experimental data can be fitted assuming a constant Avrami exponent of n = 1.27 and an activation energy of 1.02 ± 0.07 eV. The reconstruction (1 × 3)H → (1 × 2)H at low temperatures (115–140 K) is found to be much more complex.

The coadsorption of Na alkali atoms and H 2 O molecules on a stepped Ni(s)(111) surface with nominal (775) orientation was studied with thermal desorption spectroscopy (TDS), work function change measurements (ΔΦ), low electron energy diffraction (LEED) and Auger electron spectroscopy (AES). On the clean Ni(775) surface five H 2 O desorption states denoted as A (T=155-160 K), B (T=175 K), C (T=225 K), D (T=260 K) and E (T=335 K) can be found. State A is due to the adsorption of ice multilayers, state B to H 2 O bilayer clusters adsorbed at the terraces while C, D and E are step induced states

To expand previous investigations about the adsorption of H2O on a stepped Ni(s)(111) surface with nominal (221) orientation, the adsorption of H2O on a stepped Ni(s)(111) with nominal (11 11 9) orientation was studied in the present paper. The surfaces exhibit the same (111)-terrace and (11 1 )-step orientation but they differ in the terrace width, which is about three times greater for Ni(11 11 9) than for Ni(221). The adsorption was studied with thermal desorption spectroscopy (TDS), low energy electron diffraction (LEED) as well as work function change measurements (Δφ). The LEED data from the clean Ni(s)(111) surface are consistent with the atomic arrangement expected for the “ideal” non-reconstructed (11 11 9) surface. In TDS experiments five desorption states denoted as A (TM = 155–160 K), B (TM = 174–177 K), C (TM = 225 K), D and E (TM = 260 K) were found. A and B exist at similar temperatures on flat Ni(111) as well as Ni(221) and are assigned to adsorption of H2O clusters at terrace sites (B) or adsorption of ice multilayers (A). The step-induced states C, D and E, which were found at higher temperatures, are related to the adsorption of H2O monomers at steps (C) and the recombination of dissociation products (D and E). As expected, the amount of H2O molecules adsorbed at step sites as well as the contribution by step molecules to the work function change is about three times smaller than for Ni(221), corresponding to the lower step density of Ni(11 11 9). At higher coverages terrace sites are occupied by H2O molecules developing a modified bilayer structure. Between 120 and 150 K a (√3 × √3)R30° structure, which has also been found on other flat surfaces with hexagonal symmetry, could be observed. Upon heating to 160 K the LEED pattern changed into a (2 × 2) structure. This behaviour is believed to be due to a temperature- and coverage-induced reordering of state B.

The adsorption of ethylene oxide (EtO) in the presence of “metallic” K (θ(K) in the range of a first continuous overlayer) on Ni(111) is associated with a ring breaking reaction of EtO and the formation of an EtO + K reaction layer. The products are thermally stabilized to temperatures > 450 K. UPS spectra for EtO and “metallic” K demonstrate even at 100 K the formation of new surface species. Annealing to 450 K leads in the He I UPS spectrum to peaks positioned at 4.5, 8.8, 9.5 and 11.0 eV below EF. They are attributed to aldehyde-like species. This interpretation is supported by HREELS. For EtO and “metallic” K the ring breaking reaction is detected below 200 K with a vanishing of the 850 cm−1 loss feature typical for molecular EtO (C2O ring deformation mode). At 450–500 K annealing temperature new intense losses at 1140 and 1410 cm−1 and the splitting of the CH stretching mode into two losses (2720 and 3060 cm−1) are attributed to the formation of aldehyde-like radicals. For comparison, UPS and HREELS measurements have been performed for acetaldehyde and K on Ni(111) which support this interpretation.

CO adsorption on a high indexed Ni surface with nominal (551) orientation was measured with LEED and thermal desorption spectroscopy (TDS). LEED from the clean Ni(551) surface exhibits a (1 × 1) surface layer expected for the ideal atomic arrangement of a fcc (551) surface. TDS spectra show a sequential filling of four desorption states labelled A (TM = 450 K), B (420 K), C (320 K) and D (230 K). The desorption states A and B are related to the desorption from (111) and (110) microfacets of the Ni(551) surface, respectively. Desorption states C and especially D, which are missing on “flat” Ni(111) or Ni(110), are attributed to CO adsorbed at step sites. With LEED in a definite regime of CO coverage half and one-fourth order streaks in direction perpendicular to the (111) steps of Ni(551) are detected. Near the saturation of states C a weak c(4 × 2) overlayer develops.

Ethylene oxide (Et-O) adsorption on clean and potassium precovered Ag(110) was studied using LEED, angular resolved photoemission spectroscopy (ARUPS), high resolution electron energy loss spectroscopy (HREELS) and thermal desorption spectroscopy (TDS). The binding of Et-O towards clean Ag(110) is weak, complete molecular desorption occurs below 200 K. LEED and spectroscopic data suggest a binding of Et-O to Ag(110) via the oxygen atom in a short bridge adsorption site. Low K precoverages, (θ(K) T ≈ 55 K to higher temperature. However, in the low coverage range of θ(K), Et-O adsorption is still molecular. For higher K precoverages θ(K) ⩾ 0.7, the interaction character between Et-O and K changes drastically. An Et-O + K reaction layer is formed with thermal stabilization to temperatures ⩾ 450 K. This reaction layer is suggested to be due to an Et-O ring breaking reaction at ⩽, 200 K in the presence of metallic K and a polymerization into a larger compound.

Ethylene oxide (Et-O) adsorption on clean and K-modified Ni(111) surfaces was studied with thermal desorption spectroscopy (TDS) and work function change measurements (Δo) in the temperature range of 100–1000 K. On clean Ni(111), in agreement with previous measurements, evidence is found for a molecular adsorption of Et-O. The binding energy towards the Ni(111) surfaces is weak. With TDS desorption maxima are detected at 120 K (Et-O ice layer), 150 K (Et-O from a second adsorption layer) and finally at 195–215 K from fractional Et-O monolayer coverages. On clean Ni(111), Et-O lowers the work function with a maximum change of Δo = −1.6 eV. The negative sign of Δo is consistent with a perpendicular orientation of the Et-O molecular axis with the negative end — the O atom — directed towards the substrate. On K-dosed Ni(111) the Et-O adsorption behavior is significantly changed for low K coverages (θ(K) < 0.15) and drastically altered for higher coverages, θ(K) ⩾ 0.36. Up to θ(K) = 0.15 the Et-O adsorption is still molecular. Due to the lateral interaction between Et-O and K the effective Et-O dipole moment as judged from Δo measurements and the repulsive interaction between neighboring Et-O (TDS peak narrowing) decreases to zero. For higher K coverages, gq(K) ⩾ 0.36, Et-O is thermally stabilized to temperatures above 400 K. This behavior is believed to be due to a ring breaking of Et-O and the formation of a K-stabilized radical. Above 540 K, evolution of H2 and other molecular fragments gives evidence for further decomposition processes.

The adsorption of bromine on the (110) surface of silver has been studied by ultraviolet ( hυ = 21.2 and 40.8 eV) photoelectron spectroscopy in the temperature range of 100–300 K. Four different adsorption and reaction states could be detected. For fractional monolayer coverages Br 2 adsorbs dissociatively on the Ag(110) surface. The chemisorption of bromide leads to new emission features at about 3 and 5.2 eV below E F , which are assigned as occupied antibonding structures (3 eV) and as bonding Br4p x, y orbitals (5.2 eV). At 100 K, further bromide adsorption leads to the formation of an AgBr layer with molecular adsorbed bromine on top of this corrosion layer. The He I spectrum is dominated by structures at 3.5, 5.8 and 7.5 eV which are due to emission from the π g , π u and σ g molecular orbitals of Br 2 . The buildup of the AgBr layer is clearly demonstrated by desorbing the molecular bromine at about 150 K. The resulting spectrum of the AgBr layer shows peaks at 2.5 and 3.4 eV with p- and mixed-in d-character and peaks at 4.1, 5.2 and 6.1 eV which are primarily d-like. Heating of the AgBr layer up to 300 K results in a transformation from a 2D layer into a 3D agglomeration of larger AgBr clusters on top of a Br/Ag(110) chemisorption layer.

Photoelectron spectroscopy (UPS), thermal desorption spectroscopy (TDS), isotope exchange experiments, work function change (δφ) and LEED were used to study the adsorption and dissociation behavior of H2O on a clean and oxygen precovered stepped Ni(s)[12(111) × (111)] surface. On the clean Ni(111) terraces fractional monolayers of H2O are adsorbed weakly in a single adsorption state with a desorption peak temperature of 180 K, just above that of the ice multilayer desorption peak (Tm = 155 K). In the angular resolved UPS spectra three H2O induced emission maxima at 6.2, 8.5 and 12.3 eV below EF were found for θ ≈ 0.5. Angular and polarization dependent UPS measurements show that the C2v symmetry of the H2O gas-phase molecule is not conserved for H2O(ad) on Ni(s)(111). Although the Δφ suggest a bonding of H2O to Ni via the negative end of the H2O dipole, the O atom, no hints for a preferred orientation of the H2O molecular axes were found in the UPS, neither for the existence of water dimers nor for a long range ordered H2O bilayer. These results give evidence that the molecular H2O axis is more or less inclined with respect to the surface normal with an azimuthally random distribution. H2O adsorption at step sites of the Ni(s)(111) surface leads in TDS to a desorption maximum at Tm = 225 K; the binding energy of H2O to Ni is enhanced by about 30% compared to H2O adsorbed on the terraces. Oxygen precoverage causes a significant increase of the H2O desorption energy from the Ni(111) terraces by about 50%, suggesting a strong interaction between H2O and O(ad). Work function measurements for H2O+O demonstrate an increase of the effective H2O dipole moment which suggests a reorientation of the H2O dipole in the presence of O(ad), from inclined to a more perpendicular position. Although TDS and Δφ suggest a significant lateral interaction between H2O+O(ad), no changes in the molecular binding energies in UPS and no “isotope exchange” between 18O(ad) and H216O(ad) could be observed. Also, dissociation of H2O could neither be detected on the oxygen precovered Ni(s)(111) nor on the clean terraces.

The interaction of NH3 with clean Ru(001) surfaces has been studied using LEED (low energy electron diffraction), ESDIAD (electron stimulated desorption ion angular distribution), TDS (thermal desorption spectroscopy), and work function changes (Δφ). Four different binding states (denoted as α1, α2, β and γ) were detected with TDS. At low coverages, NH3 desorbs from the α1 state with a TDS peak maximum at ~ 310 K. The broadening of the TDS peaks and their shift to lower temperature with increasing NH3 coverage are related to repulsive lateral interactions between neighboring NH3 molecules. At higher NH3 coverages (θNH3≳ 0.15), a second desorption peak (α2) develops at T = 180 K, accompanied by a (2 × 2) LEED structure. With further increase of NH3 exposure a sharp desorption peak (β state) is found at T = 140 K, and is interpreted as due to NH3 species desorbing from a second adsorption layer. Finally a desorption peak due to multilayer adsorption (γ state) is found at 115 K. At low NH3 coverages (α1 state), a “halo”-like H+ ESDIAD pattern gives evidence of randomly oriented or freely rotating NH3, monomers, bounded via the N atoms to the surface with the H atoms pointing away from the surface. This orientation of NH3 is supported by work function measurements showing a linear decrease of Δφ in the α1 state. Structural information concerning the adsorption geometry of NH3 in the β state has been obtained from LEED and ESDIAD. During the formation of the second NH3 layer (β) a (2√3 × 2√3)R30° LEED pattern is observed and is accompanied by an ESDIAD pattern with a hexagonal outline. A structural model of the β-state bonding, in which second layer NH3 molecules are bonded via threefold hydrogen bonds to the first layer NH3, is proposed.

H 2 O adsorption on clean Ni(110) surfaces at T ≦ 150 K leads at coverages below θ ≊ 0.5 to the formation of chemisorbed water dimers, bound to the Ni substrate via both oxygen atoms. The linear hydrogen bond axis is oriented parallel to the [001] surface directions. With increasing H 2 O coverage (θ ≧ 0.5) , the accumulation of further hydrogen bonded water molecules induces some modification of the dimer configuration, producing at θ ≊ 1 a two-dimensional hydrogen bonded network with a slightly distorted ice lattice structure and long range order.

The adsorption and reaction of H 2 O on clean and oxygen precovered Ni(110) surfaces was studied by XPS from 100 to 520 K. At low temperature ( T 2 O on the clean surface with nearly constant sticking coefficient was observed. The O 1s binding energy shifted with coverage from 533.5 to 534.4 eV. H 2 O adsorption on an oxygen precovered Ni(110) surface in the temperature range from 150 to 300 K leads to an O 1s double peak with maxima at 531.0 and 532.6 eV for T =150 K (530.8 and 532.8 eV at 300 K), proposed to be due to hydrogen bonded O ads … HOH species on the surface. For T >350 K, only one sharp peak at 530.0 eV binding energy was detected, due to a dissociation of H 2 O into O ads and H 2 . The s-shaped O 1s intensity-exposure curves are discussed on the basis of an autocatalytic process with a temperature dependent precursor state.

The structure and kinetics of NH3 interacting with clean and sodium or oxygen-dosed Ni(110) have been studied using ESDIAD (electron stimulated desorption ion angular distribution), LEED and thermal desorption spectroscopy. NH3 adsorbs in molecular form at 80 K. No azimuthal ordering is found for molecular NH3 on either clean or O-dosed Ni(110), but NH3 reacts with O above 200 K to form a surface species “inclined” along [001] azimuths, identified as OH(ad). Coadsorption of NH3 with Na causes a reorientation of the molecular axes of the NH3 bonded near Na.

The oxygen induced p(2×2) and c(2×2) LEED symmetries on Ni(100) were studied with UPS He I (s- and p-polarization), He II and XPS (O Is and XAES O(KLL) spectra), revealing only some minor differences between both states. In the UPS He II analyzed under a polar angle of 45° in the [100] azimuthal direction, the half width of the O 2p maximum from the c(2 × 2) overlayer increases to ΔE=2.5 eV, suggested to be due to a splitting of the 2p levels. During the transition from p(2×2) to c(2×2), with XPS, only a small increase of the O Is BE from 529.9 to 530.1 eV is observed, whereas the X-ray excited O(KL23L23) peak shifted from 514.3 to 515.0 eV. In contrast to the results from the p(2×2) and c(2×2) overlayers on Ni(100), where the O 2p intensity nearly vanishes with s-polarization, the O 2p emission from the (2×1) structure on Ni(110) is found to be less dependent from the polarization. The O 2p intensity profile, measured as a function of polar angle in the [100] and [110] azimuthal direction, is nearly identical in both cases.

ELS and simultaneous quartz microbalance investigations have been carried out on clean aluminum and its surface oxide layer. The loss spectrum of clean Al is interpreted by the collective features of the conduction electrons: volume and surface plasmons, the latter being extremely sensitive to a small oxygen uptake. In the very beginning the oxidation is characterized by a loss peak at 7.3 eV which is attributed to a single electron transition from the O(2p) level to an unfilled state near the Fermi level of the metal. The decreasing intensity of the 7.3 eV loss and the increasing of a second loss at 19.2 eV with further oxygen uptake are tentatively explained by the formation of Al2O3 and interband transitions of amorphous Al oxide. The formation of A12O3 is supported by the gravimetric measurements of oxygen mass gain.

The total energy distribution of electrons emitted from clean Cu(100) and oxygen covered surfaces is analysed. A primary electron energy of 400 eV enabled the investigation of characteristic losses (ELS), Cu MVV Auger transitions and true secondary electrons in a single spectroscopic run. Oxygen exposure up to 108 L at elevated temperature (~400 K) results in a Cu density of states (DOS) strongly affected by O(2p) electrons. The Auger lines of Cu, atomic-like for clean surfaces, reveal DOS effects after some 107 L oxygen exposure: all MVV transitions shift down by ~2 eV in spite of a fixed M23 level; the M23VV Auger line splitting is vanishing due to a broadened valence band maximum allowing the deexcitation of the final two-hole state of intraatomic transitions. Heating the oxygen covered crystal to 820 K is accompanied by the removal of much surface oxygen and an electronic state resembling an earlier oxidation state without DOS effects in the Cu Auger spectrum.

The beginning oxidation of evaporated Al films was measured under UHV conditions by means of AES, gravimetric oxygen uptake and work function changes. The high resolution of our quartz oscillator microbalance (Δm = 7 × 10-11g/cm2 with a time response 7.5 × 10-8g/cm2. At room temperature a decrease of work function with an equilibrium value of -0.75 eV (oxygen uptake 9.10-8g/cm2) with increasing oxidation was observed. This is interpreted as incorporation of oxygen and diffusion of Alδ+ ions to the surface. But at T = 80 K an increase of work function with a maximum value of 0.5 eV was found. This is attributed to adsorbed oxygen because the incorporation is much slower at this temperature. The AES of pure Al is characterized by the low energy 64 eV peak. The decrease of the 64 eV peak and the increase of two new peaks at 50 and 37.5 eV during the oxidation were measured as a function of oxygen uptake and attributed to the change of electron density of states in the conduction band of aluminum.

The interaction of oxygen with Ni(110) surfaces was investigated by means of electron loss spectroscopy (ELS) and Auger electron spectroscopy (AES). The ELS of clean Ni revealed the well known losses due to interband transitions and plasmon features. New losses due to oxygen adsorption were measured for exposure >10 L at Δ E = 5.8, 7.4, 13.9 and 23.9eV. The drastic change in the ELS at this stage of Ni oxidation coincides with the splitting of the M 23 loss peak, the changes in the MVV and L 3 VV Auger line shapes and a shift of the O(KLL) peak to lower energy.

The presence of oxygen on a Ni(110) surface promotes the adsorption and decomposition of H 2 O at 300 K. Angle resolved UPS (ultraviolet photoemission spectroscopy), ESDIAD (electron stimulated desorption ion angular distribution) and isotope experiments all indicate that OH(ad) is formed on the surface, presumably via a hydrogen abstraction reaction, H 2 O + O(ad) → 2 OH(ad). The molecular axes of the OH(ad) species are inclined with respect to the surface normal, and are oriented along [001] and [001] azimuthal directions.

At 300 K oxygen adsorbs dissociatively on Ag(111) with a sticking coefficient S0 − 10−5. The uptake was followed with XPS using the O 1s intensity calibrated via the O(2×1) symmetries on Ni(110) and Ag(110). The maximum coverage amounts to ϑ − 0.6 (2×109 L, −1 Torr). This atomic oxygen state is characterized in the UPS by two induced maxima from bonding and antibonding 2p orbitals 8.8 and 3.4 eV below EF. The O 1s BE is found to be 530.2 eV; in the EELS one loss is measured at 27 meV (220 cm−1). At 100 K, the interpretation of the photoelectron as well as of the EEL spectra is hindered due to H2O adsorption from the residual gas.

The interaction of water vapour with clean as well as with oxygen precovered Ni(110) surfaces was studied at 150 and 273 K, using UPS, ΔΦ, TDS, and ELS. The He(I) (He(II)) excited UPS indicate a molecular adsorption of H 2 O on Ni(110) at 150 K, showing three water-induced peaks at 6.5, 9.5 and 12.2 eV below E F (6.8, 9.4 and 12.7 eV below E F ). The dramatic decrease of the Ni d-band intensity at higher exposures, as well as the course of the work function change, demonstrates the formation of H 2 O multilayers (ice). The observed energy shift of all water-induced UPS peaks relative to the Fermi level (Δ E max = 1.5 eV at 200 L) with increasing coverage is related to extra-atomic relaxation effects. The activation energies of desorption were estimated as 14.9 and 17.3 kcal/mole. From the ELS measurements we conclude a great sensitivity of H 2 O for electron beam induced dissociation. At 273 K water adsorbs on Ni(110) only in the presence of oxygen, with two peaks at 5.7 and 9.3 eV below E F (He(II)), being interpreted as due to hydroxyl species (OH) δ- on the surface. A kinetic model for the H 2 O adsorption on oxygen precovered Ni(110) surfaces is proposed, and verified by a simple Monte Carlo calculation leading to the same dependence of the maximum amount of adsorbed H 2 O on the oxygen precoverage as revealed by work function measurements. On heating, some of the (OH) δ- recombines and desorbs as H 2 O at ≅ 320 K, leaving behind an oxygen covered Ni surface.

The chemisorption of CO on Cu, Ni and CuNi alloy surfaces was examined by SIMS, work function measurements and desorption spectroscopy. Using a dynamic SIMS technique the M+, M+2, MCO+ and M2CO+ emission at different temperatures (100–400 K) was measured as a function of CO exposure. In agreement with the work function and desorption experiments an increase of M+ and MCO+ emission due to the CO adsorption on Cu was found only at low temperatures (100–190 K). On the Ni surface an increase of Ni+, NiCO+ and Ni2CO+ was measured up to 400 K. The adsorption of CO on CuNi alloy surfaces — as derived from the work function measurements — can be described by the assumption of two different states of adsorbed carbon monoxide. They can be characterized by different binding energies and from sign and magnitude different work function changes. These states were interpreted as adsorption at Ni or Cu sites of the alloy surfaces, respectively. To a certain extent the SIMS results from the alloy surfaces are incompatible with the work function measurements and desorption spectroscopy and the SIMS studies on the pure metals. A Cu+ emission with comparable intensity to the Ni+ emission was found for alloys with bulk concentrations of 60 and 40 at% Cu at 300 K. The ratio Ni+Cu+ was nearly independent of CO pressure and temperature. The measured ratios of Cu+2(Cu+ + Ni+), Ni+2(Cu+ + Ni+) and CuNi+(Cu+ + Ni+) with values about 10−2 can be explained the basis of a statistical arrangement of Cu and Ni atoms in the alloy surface. The intensities of the MCO+ emissions are 102 times smaller than the corresponding values of the pure metals. No emission of M2CO+ was found on CuNi during CO adsorption.

The adsorption of H2O on both clean and modified Ni(110) surfaces has been studied using a variety of methods: electron stimulated desorption ion angular distribution (ESDIAD), thermal desorption spectroscopy (TDS), and low energy electron diffraction (LEED). Fractional monolayers, θ(H2O) 0.5–1 larger hydrogen bonded clusters with long range c(2 × 2) symmetry are formed. Upon heating to ⩾ 200 K, a fraction of the H2O dissociates, forming OH(ad). TDS of H2O from clean Ni(110) reveals four binding states having peak temperatures of 155, 210, 245 and 370 K. They are related to multilayer desorption (155 K), desorption from larger bilayer clusters (210 K), desorption from H2O dimer clusters which might be stabilized by OH (245 K), and recombination of OH to yield H2O(g) (360 K). Dissociation of H2O is promoted by surface oxygen. For the adsorption of H2O on oxygen-dosed Ni(110) at θ(O) > 0.08, a mixture of molecular and dissociative adsortion occurs immediately at 80 K, producing inclined OH. Isotropic exchange of H216O with 18O(ad) is observed even for binding states in which dissociation is believed not to to occur and is related to a proton exchange involving H2O(ad) hydrogen bonded to O(ad).