Work Function Data Research Articles

Influence of Cu, Ag, Ir and Pt on the work function and near‐surface electron density of Re mass spectrometer filaments

AbstractThe influence of metal overlayers on the properties of polycrystalline Re mass spectrometer filaments has been examined using contact potential difference (CPD) measurements of the work function, x‐ray photoelectron spectroscopy (XPS) measurements of overlayer binding energy (BE) and ultraviolet photoelectron spectroscopy (UPS) measurements of electron energy distribution curves (EDCs). Work function data for CuRe, AgRe and IrRe are compared to earlier results for AuRe, PdRe and PtRe as part of the effort to understand the influence of surface additions to the work functions of ionization sources. The changes in work function owing to the addition of a metal overlayer do not always follow predictions of bulk electron transfer, such as the Miedema electronegativity. However, the observed changes in work function are consistent with a simple initial‐state understanding of the XPS core‐level data involving charge flow toward and away from the surface. The XPS binding energy shifts are consistent with shifts observed in the valence bands for Cu and Ag on Re.

Surface and bulk electrical properties of the hematite phase Fe2O3

Electrical properties of Fe2O3 were studied by using several electrical methods such as electrical conductivity, thermopower (Seebeck effect) and work function. The studies were performed at elevated temperatures (1053–1153 K) and under controlled oxygen activity (102∓105 Pa). Samples of different thickness varying between 103 nm and 1 mm were taken for the measurements of both electrical conductivity and thermopower. It has been found that the exponent of the po2 dependence resulting from the work function measurements (1/nφ) is about 1/2. Both thermopower and electrical conductivity data are well consistent with work function data for the thin film (1000 nm) of Fe2O3. The charge transport in Fe2O3 has been interpreted in terms of small polaron mechanism. Analysis of measured electrical parameters, regarding the thickness of studied specimens, indicates that the near-surface layer of Fe2O3 exhibits much higher deviation from stoichiometry than the bulk phase and resulting strong interaction between charge carriers. This effect has been interpreted in terms of segregation of intrinsic lattice defects to the surface, and presumably also to grain boundaries, of Fe2O3.

Interaction of oxygen with Si(111)7 × 7 and Si(100)2 × 1 surfaces modified with small amounts of Cs

The effect of small amounts of Cs (∼ 0.2 of the saturation Cs coverage at 300 K) on the oxidation of Si(111)7 × 7 and Si(100)1 × 2 at reaction temperatures of 300–700 K has been studied by means of Auger and electron energy loss (EEL) spectroscopies, thermal desorption and work function (WF) measurements. It has been found that at these low Cs coverages only the first stage of the oxidation process — the initial oxygen uptake — is affected. The result is: (i) an increase and temperature independence of the initial sticking coefficient, S 0; (ii) coverage independence of S upto certain (reaction temperature determined) oxygen coverages. EEL and WF data have shown that at 300 K oxygen occupies preferentially Cs promoted sites. The liberation of oxygen from this Cs influenced state in order to form usual Si-O bonding configurations takes place at temperatures exceeding 400 K as evidenced by the rise of sharp SiO related ∼ 5 and 7 eV loss peaks and reduction of the work function. Thermal desorption data showed that the oxidation of Si leads to an appearance of a higher temperature desorption state of Cs. Because of this oxygen induced stabilization of the Cs layer the removal of Cs requires annealing the mixed layer at 950 K. At oxygen coverages less than 1 ML this annealing procedure led also to complete SiO desorption.

Thermionic work function of chemically vapor deposited rhenium on tungsten

The vacuum work function of chemically vapor deposited (CVD) rhenium was investigated. The thermionic emission technique was used in order to obtain high temperature data. Data were collected as a function of temperature and time, and the total pressure was recorded. X-ray analysis showed that the rhenium was preferentially deposited with the 0001 planes approximately parallel to the emitting surface. The effective work function data were compared with the results of Campbell et al. [1]. It was concluded that the residual gas pressure was responsible for conflicting work function data.

High‐Temperature Kelvin Probe in Application to Ceramic Materals: Investigation of Nickel Oxide

A work function analysis system (Kelvin probe) which operates at high temperatures and under controlled gas atmospheres has been constructed and its application in studies of surface properties of ceramic materials is discussed. The technique allows in situ characterization of solid–gas interfaces and monitoring of surface processes, such as structural rear rangements, phase transitions, equilibration processes, and defect segregations. Experimental data for NiO single crystals illustrate that high‐temperature work function (WF) data may be used for a description of the defect structure model in the boundary layer and its changes as a result of segregation of extrinsic defects. WF measurements were applied to monitor the surface reequilibration during redox processes at 780°C in gas atmospheres of controlled oxygen partial pressure (1.74 × 102 to 4.3 × 104 Pa). The power of the WF dependence on the oxygen partial pressure determined for the oxidation and reduction experiments changes from 1/6 within 10 h to 1/1.5 after 220 h of annealing. During this treatment the bulk‐related oxygen exponent remains unchanged and close to 1/6. The experimental results are discussed in terms of the NiO defect structure in relation to the crystalline bulk and the near‐surface layer. The results indicate that equilibration of the near‐surface oxide layer involves not only the propagation of the nonstoichiometry from the surface into the bulk but also the segregation of impurities. The second process leads to strong changes in composition and resulting interactions between the defects in the outer layer which no longer can be described by the ideal defect model. The effect of ceramic interfaces (surfaces and grain boundaries) on equilibration kinetics in the gas–solid systems has been considered.

The effect of density of states, work function and exchange integral of polycrystalline and single crystal surfaces on the hydrogen evolution reaction

The influences of work function and density of states on the kinetics of the hydrogen evolution reaction on polycrystalline surfaces are reviewed. Work function data for single crystal surfaces are compared to the exchange current for the hydrogen evolution reaction through consideration of surface density. Correlation between the density of states and electrochemical kinetic parameters is reviewed and compared to calculated data.

Surface relaxation of Y2O3-stabilized ZrO2

Surface phenomena accompanying oxidation and reduction process of yttria-stabilized zirconia (Zr0.82Y0.18O2−δ) were studied using work function (WF) measurements. Relaxation experiments were performed at 780°C in the oxygen partial pressure range from 1.74×102 to 4.3×104 Pa. It is observed that the WF value after both the education and oxidation runs change continuously within about 150 h and remains constant at longer times. The WF of the individual oxidation and reduction runs depends on the partial pressure with a power which changes from 14 within initial 10 h to about 12 at 120 h. The oxygen partial pressure dependence of quantities related to the bulk defect structure remain constant at the same time. Oxidation and reduction experiments are interpreted in terms of the thermodynamic equilibrium between the gas phase and oxygen vacancies which are assumed to be predominant lattice defects both in the bulk and in the near-surface layer of zirconia as long as the power of the oxygen partial pressure dependence (determined from WF) is 14. The long term drift of both absolute WF values and the oxygen exponent is caused by surface segregation of impurities involving predominantly Al and Ca. It is concluded that the equilibration kinetics for the near-surface layer, determined by segregation, is much slower than the bulk kinetics controlled by diffusion of oxygen vacancies. Based on WF and SIMS data, the sandwich-type model for the near-surface layer of yttria-doped ZrO2 is developed. The sandwich involves the outermost layer enriched predominantly by Al and Ca, the sublayer enriched in Y and the crystalline bulk.

Work function and Auger characteristics of BaO on Rh-Nb substrate mixtures

The interaction of a BaO adsorbate with a Rh-Nb substrate mixture has been investigated using both interatomic Auger analyses as well as work function studies. The results show that the Ba interacts predominantly with Rh and the O with the Nb in a manner similar to that reported previously for Ir and W substrates, respectively. The intensity of either of these interactions, however, was not as strong as it was compared to the Ir-W case. The manner in which the Auger data varied for different binary Nb-Rh mixtures suggested that the preferential alignment of the BaO molecules noted in the Ir-W case does not occur for the Rh-Nb case and that this is probably due to the weaker interactions between adsorbate and substrate. The work function data also did not show the pronounced minimum near 50% that was observed with the aligned BaO on Ir-W. The correlation between the Ba + and O - ionicity and the work function lowering, however, still seems to be valid, regardless of any BaO alignment effect.

Adsorption of CO on Ru(10 [formula omitted]0) doped with different amounts of K and Cs

The interaction of CO with K and Cs doped Ru(10 1 0) has been studied by means of Auger electron spectroscopy (AES), thermal desorption (TD), electron energy loss spectroscopy (EELS), LEED and work function (WF) measurements. It was found that both alkali metals (AM) induce similar changes in the CO adsorption kinetics and energetics, with preferential filling of the promoted sites. The initial heat of CO adsorption, Δ H CO 0, associated with promoted CO ∗, increases linearly from 145 up to 224 kJ/mol with increasing alkali coverage, θ AM, the maximum Δ H CO 0 value being reached when coincident CO and AM desorption is detected. LEED and WF data for coadsorbate systems indicate formation of patches with different WF and compositions on the surface. EELS results show that the presence of AM causes an energy shift of the CO 2 π b → 2 π a transitions from 6.3 to ∼7.0 eV, removal of the AM ns valence transitions and plasmon excitations and appearance of new loss structures. Comparing the present data with the results on CO adsorption on Na doped Ru(10 1 0) the type and contribution of the possible direct COAM interactions are discussed.

Nickel monoxide-oxygen interaction between 20 and 1000 °c and its impact on the nickel monoxide reduction mechanism

The oxygen adsorption isobar and work function data for nickel monoxide are considered from the point of view of predominant surface oxygen species. The mechanism of oxygen interaction with the NiO surface in the range 20–1000° C is discussed, involving chemisorption equilibria and subsequent changes in oxide non-stoichiometry as a result of oxygen incorporation. It is concluded that five temperature ranges may be distinguished which correspond to different mechanisms of the NiO-oxygen interaction. The effect of NiO surface pre-treatment in oxygen on the mechanism of nickel nucleation is considered. Hypotheses are put forward to explain the specific changes in the number of metallic nuclei and in their volume according to whether oxygen pre-treatment of reduction is performed below or above the Neel temperature of NiO.

The coadsorption of oxygen and potassium on Ru(001): Evidence for the formation of K–O compounds

In order to study an adsorption system where bond formation between coadsorbates competes with the adsorbate–metal substrate bond, we have investigated the interaction between oxygen and a potassium monolayer on Ru(001). At low exposures of oxygen (0.4 L), vibrational, photoemission, Auger, and workfunction data indicate the formation of a KO2 species. EELS spectra are characterized by an intense K–O stretch at 240 cm−1 and the absence of Ru–O stretching modes. Auger spectra indicate a K:O stoichiometry of 1:2 and photoemission spectra contain features indicative of O–O bond formation. Subsequent exposure of oxygen results in the adsorption of atomic oxygen with a characteristic Ru–O stretching vibration at 615 cm−1. Annealing of the K–O layer results in the decomposition of KO2 to Kad+Oad, both being bonded more strongly to the metal substrate than the individually adsorbed species and thus indicating through-metal interactions between atomic oxygen and potassium.

Interaction of oxygen with an alkali modified Ru(001) surface

The interaction of oxygen with potassium predosed Ru(001) has been studied by means of thermal desorption (TD), Auger electron spectroscopy (AES), electron energy loss spectroscopy (EELS) and work function (WF) measurements. These results were compared with those reported earlier on Na and Cs precovered Ru(001). It was found that the initial sticking coefficient, S 0, increased linearly with alkali coverages θ am, in the sequence Na-K-Cs. The TD, WF and EELS data indicated predominance of the strong oxygen-ruthenium interactions over the alkali-oxygen interactions at low potassium and oxygen coverages (θ K and θ O, respectively). Above certain θ K and θ O (θ K > 0.12 and θ O > 0.6), the coincident O 2, K and K 2O TD peaks at 890 K, the EL feature at ~ 28 eV, and the appearance of a shoulder in the WF I–V curves were interpreted as the formation of K-O-Ru complexes gathered in patches on the Ru(001) surface. This behaviour was common for the three alkali metals when coadsorbed with oxygen.

Surface reactions on rare earth metals monitored by work function measurements

Surface reactions on clean, oxidized, and partially hydrided rare earth metal films with O 2, H 2, H 2O, CO, and CO 2 were monitored by dynamic work function measurements under ultrahigh vacuum conditions. The work function data are discussed in the light of electron spectroscopic evidence, where available. The main results of this study are: Initial oxidation of Yb seems to follow a different oxidation mechanism as compared to other rare earth metals. Reaction of H 2O with clean, oxidized, and hydrided rare earth metal surfaces yields surface hydroxyl species. H 2 uptake on divalent Yb metal is much slower than on trivalent Er metal, but occurs at a comparable rate on the oxidized systems. On oxidized surfaces, surface OH formation is substantiated for room-temperature reaction with H 2. The Δφ data indicate high reactivity of CO and CO 2 toward clean rare earth metal surfaces and are consistent with dissociative adsorption of both molecules.

Ordered oxygen overlayers on Cr(100) observed by leed and photoemission

Chemisorbed oxygen atoms on Cr(100) induce strong O(2p) derived surface resonances which are studied by angle resolved photoemission. Well ordered structures are observed after annealing (300°C). In the submonolayer range (θ O < 1) a study of the symmetry and dispersion of the O(2p) derived features shows the two-dimensional Bloch character associated with either a c(2 × 2)-O surface at low coverages (θ O≅0.25) or a (1 × 1)-O structure at high coverages (θ O≅0.9). When combined with LEED observations and work function data this study indicates that both structures coexist around θ O = 0.5 and chemisorbed oxygen is probably incorporated into the fourfold hollow sites. At θ O > 1, the onset of oxidation is clearly shown in the valence band and core level spectra and the data support the existence of a thin spinel-like oxide layer.

Surface oxidation of ytterbium

The initial stages of the oxidation of Yb have been investigated by electron energy loss spectrsocopy (ELS), Auger spectroscopy and work function measurements. The electron loss and Auger spectra of clean and oxygen covered Yb are analysed in some detail to obtain suitable diagnostic means for the oxidation reaction: the valence change of Yb atoms upon oxidation is reflected in pronounced shifts of plasmon energies and in the appearance of the 4d→4f giant resonance in ELS. These spectral changes were used to monitor the oxidation process. We find metal and oxide plasmons coexisting at moderate oxygen exposures, and differences in uptake curves derived from O KLL Auger signals and Yb3+ giant resonance signals; the formation of Yb3+ is somewhat delayed with respect to the oxygen uptake. Together with work function data these results provide complementary evidence for an island growth mechanism at the initial stages of the oxidation of Yb.

Surface oxidation of ytterbium

The initial stages of the oxidation of Yb have been investigated by electron energy loss spectrsocopy (ELS), Auger spectroscopy and work function measurements. The electron loss and Auger spectra of clean and oxygen covered Yb are analysed in some detail to obtain suitable diagnostic means for the oxidation reaction: the valence change of Yb atoms upon oxidation is reflected in pronounced shifts of plasmon energies and in the appearance of the 4d→4f giant resonance in ELS. These spectral changes were used to monitor the oxidation process. We find metal and oxide plasmons coexisting at moderate oxygen exposures, and differences in uptake curves derived from O KLL Auger signals and Yb 3+ giant resonance signals; the formation of Yb 3+ is somewhat delayed with respect to the oxygen uptake. Together with work function data these results provide complementary evidence for an island growth mechanism at the initial stages of the oxidation of Yb.

The work function as a monitor for thermal desorption spectroscopy: CO desorption from Ni sites on a Ni-rich (110) Cu-Ni surface

A new method for the evaluation of thermal desorption spectra is described. Using this approach, based on the usual assumptions of transition-state theory, the coverage dependence of the Gibbs free activation energy G D (θ) of desorption reactions can be determined. As experimental input it suffices to measure any physical property which is proportional to θ during time-programmed desorption as a function of time and surface temperature. If the isosteric heat and the activation energy of adsorption are known, the activation entropy of desorption and the Arrhenius preexponential factor can be estimated from G D. Work-function data on CO desorption from Ni-rich (111) Cu-Ni alloy surface are used to discuss this method.

Binding energies and chemical shifts of least bound core electron excitations in cubic ANB8−N semiconductors

AbstractCore electron binding energies EB with respect to the vacuum level and their chemical shifts are calculated for the least bound core levels of cations and anions of cubic ANB8−N semiconductors. Starting from the HF‐binding energy of the free atom absolute values of EB are obtained by adding core level shifts and relaxation energies. Core level shifts are calculated by means of an electro‐static model with ionic and bond charges according to Phillips' bond charge model. For the calculation of relaxation energies the linear dielectric theory of electronic polarization is applied. Valence and core electrons, and diagonal and non‐diagonal screening are taken into account. The theoretical results for chemical shifts of binding energies are compared with experimental values from XPS‐measurements corrected by work function data. Good agreement is obtained in all cases within the error limit of about one eV. Chemical and atomic trends of core level shifts, relaxation energies, and binding energies are discussed in terms of changes of atomic and solid state parameters. Chemical shifts and relaxation energies are predicted for various ternary ANB8−N compounds.

An ELS study of the adsorption of oxygen on W(110)

Oxygen adsorption onto W(110) at room temperature has been characterized by ELS with the help of Auger, LEED and work function data collected from the same system. Some form of reconstruction occurs up to about 20% coverage. Growth of p(2 × 1) structured islands then takes over up to half coverage after which adsorption occurs via a different type of bonding to the surface. The large gas exposures necessary to saturate the surface at room temperature do not cause bulk oxidation and modest heating of the high coverage surface does not cause reconstruction but merely a re-ordering in the oxygen overlayer. The surface complex formed at high coverage and high temperature leading to desorption of tungsten oxides is characterized in ELS by two sharp losses at 6 and 8 eV.

Adsorption of CO on Pd(100)

Adsorption of CO on a Pd(100) surface was studied in detail mainly by LEED, UPS, work function and thermal desorption measurements. Analysis of the ordered c(2√2×√2) R 45° structure occurring at Θ=0.5 revealed that in this phase each CO molecule is bridge bonded to 2 Pd atoms with Pd–C distances of 1.93±0.07 Å and a C–O bond length of 1.15±0.1 Å, the molecular axis being oriented normal to the surface. The mutual configuration of the adsorbed molecules is explained in terms of a short-range repulsive interaction model, which is supported by the observation that the isosteric heat of adsorption (Ead=38.5 kcal/mole) is constant up to a coverage of Θ?0.45. The photoelectron spectra exhibit two maxima at 7.9 (5σ+1π level) and 10.8 eV (4σ level) below the Fermi level which are in agreement with the observations with other Pd planes. This also holds for an electronic excitation associated with an energy of 13.5 eV as observed by electron energy loss spectroscopy. The variation of the sticking coefficient with coverage is described in terms of a second-order precursor state model with repulsive interactions. The pre-exponential factor for desorption (3×1016 sec−1) varies only little with coverage. The dipole moment of the adsorbate remains constant up to Θ?0.35 (0.17 D) where the overlayer starts to order, and exhibits an appreciable higher value near Θ=0.5. The maximum work function increase is Δφmax=0.93 V at Θ=0.5. The differential entropy of the adsorbed layer around 450 K was derived from the experimental adsorption isotherms. Up to Θ=0.35 the data fall well between the limits of two theoretical models for localised and delocalised adsorption of noninteracting particles. Deviations at higher coverages reflect again the onset of ordering. The energetic and work function data are in some variance to results reported earlier which is ascribed to the fact that these quantities may be sensitively influenced by spurious amounts of carbon impurities on the surface.