Constant Sticking Coefficient Research Articles

Adsorption states and mobility of trimethylacetic acid molecules on reduced TiO2(110) surface

Combined scanning tunneling microscopy (STM), X-rays photoelectron spectroscopy (XPS) and density functional theory (DFT) studies have probed the bonding configurations and mobility of trimethylacetic acid (TMAA) molecules on the TiO(2)(110) surface at RT. Upon TMAA dissociation through deprotonation, two distinctly different types of stable chemisorption configurations of the carboxylate group (TMA) have been identified according to their position and appearance in STM images. In configuration A, two carboxylate O atoms bond to two Ti(4+) cations, while in configuration B one O atom fills the bridging oxygen vacancy (V(O)) with the other O bounded at an adjacent regular Ti(4+) site. Calculated adsorption energies for the configurations A and B are comparable at 1.28 and 1.36 eV, respectively. DFT results also show that TMA may rotate at RT about its O atom that filled the V(O) (in configuration B), with a rotation barrier of approximately 0.65 eV. Both the observation of the constant initial sticking coefficient and preference for TMAA molecules to dissociate at selective sites indicate that TMAA adsorption is mediated by a mobile precursor state. Several possible molecular (physisorbed) states of TMAA have indeed been identified by DFT, all being highly mobile at RT. In contrast, the TMA diffusion in the chemisorbed (dissociative) state is a very slow with a calculated barrier of 1.09 eV for diffusion along the Ti row.

Deposition dynamics and chemical properties of size-selected Ir clusters on TiO 2

We report a study of Ir n /TiO 2 samples prepared by size and energy-selected deposition of Ir n + ( n=1, 2, 5, 10, 15) on rutile TiO 2(1 1 0) at room temperatures. The Ir clusters are found to be formally in the zero oxidation state, and there are no significant shifts in Ir 4f binding energy with cluster size. Over a wide range of impact energies, both Ir XPS intensity and peak position are constant, indicating constant sticking coefficient, and no impact-driven redox chemistry. Low energy ion scattering spectroscopy (ISS) suggests that the deposited Ir clusters remain largely intact, neither fragmenting nor agglomerating, and retaining 3-D structures for the larger sizes. For impact energies above 10 eV/atom, comparison of ISS and XPS data show that the Ir clusters are penetrating into the TiO 2 surface, with the extent of penetration increasing with both per atom energy and cluster size. Temperature programmed desorption (TPD) of CO is used to further characterize the deposited Ir n . This system shows pronounced substrate-mediated adsorption (SMA) in low CO exposures, with strong dependence on cluster size. ISS and sputtering experiments indicate that CO adsorbed via SMA is bound differently than CO adsorbed in high dose experiments. In experiments with sequential C 16O and C 18O doses, facile C 16O → C 18O exchange is observed for Ir 5 and larger clusters, but not for Ir 2. The peak CO desorption temperature is found to decrease with cluster size. The cycle of CO adsorption and heating comprising a TPD experiment have a dramatic effect on the sample morphology, leading to encapsulation of Ir by a thin TiO x layer.

Modeling of Cu Transport in Sputtering Using a Monte Carlo Simulation

For the Cu-sputtering process, conventional simulations have been carried out to examine individual phenomena such as the production of plasma, the emission of atoms from the target, the transport of emitted atoms through space and the sticking of atoms onto the wall surface. We have developed a simulator which enables the analysis of the Cu-sputtering process from the production of plasma to the transport and deposition of sputtered atoms. In the present simulation, the flux of Ar ions incident onto the target is determined by a plasma simulation using a particle-in-cell/Monte-Carlo (PIC/MCC) model. Sputtered atoms are assumed to be emitted following the cosine law with a Maxwellian velocity distribution. The test-particle Monte Carlo method is used in the simulation of the transport of the sputtered atoms. Sputtered atoms are assumed to stick to the walls with a constant sticking coefficient. With a sticking coefficient of 0.93, determined experimentally, the dependence of the film growth rate on two parameters, i.e., target–wafer distance and pressure, agreed well with the experimental results.

Bromine adsorption, reaction, and etching of Cu(100)

The interaction of Br 2 with Cu(100) was characterized using temperature-programmed desorption (TPD) and low-energy electron diffraction (LEED). Initial exposure to Br 2 resulted in the formation of a c(2 × 2) LEED pattern and CuBr desorption peaks at 870 and 1000 K. These desorption peaks saturated at a dose of approximately 5 L and were attributed to Br chemisorption. Continued exposure to Br 2 resulted in the growth of Cu 3Br 3 desorption peaks associated with the formation of bulk CuBr. The Cu 3Br 3 desorption peaks exhibited a strong coverage dependence. At low CuBr coverages, desorption peaks at 450 and 485 K were observed, while at intermediate coverages a third peak at 540 K was observed, and at high coverages a single broad peak at 500 K was observed. Similar results were obtained whether the CuBr layer was formed by exposure to Br 2 or by deposition of Cu 3Br 3 onto a c(2 × 2) layer. The two higher-temperature peaks were shown to be consistent with sublimation of α and β CuBr, while the lowest-temperature peak could not be associated with sublimation of any bulk phase of CuBr. The lowest-temperature peak was attributed to either a grain-size effect or to the interaction of CuBr with chemisorbed Br. At 325 K, growth of CuBr resulted in a hexagonal LEED pattern that disappeared upon annealing to 400 K. Again, the same results were obtained for CuBr formed by reaction with Br 2 and by vapor deposition of Cu 3Br 3, suggesting that the reaction produces near-equilibrium structures. The hexagonal LEED pattern was attributed to compressed epitaxial CuBr(111). The reaction of Br 2 with Cu(100) was characterized by a constant sticking coefficient, indicating that the reaction is adsorption-rate limited over the range studied. The sticking coefficient was found to depend on temperature in two distinct ways: (i) annealing the c(2 × 2) layer irreversibly increases the sticking coefficient, and (ii) the sticking coefficient reversibly decreases with increasing reaction temperature. The first effect was attributed to structural changes in the chemisorbed layer.

Monitoring adsorption and desorption on a metal surface by optical non-resonant reflectivity changes

A linear optical reflection technique is demonstrated as a simple and non-intrusive means for monitoring the adsorption and desorption of carbon monoxide on Cu(100). The non-resonant reflectivity change induced by CO adsorption on copper is of the order of −1%, allowing submonolayer sensitivity of 0.01 ML to be achieved. The induced reflectivity change exists over a broad frequency range, and is found to correlate linearly with the coverage in both the infrared ( λ=4.76 μm) and the visible ( λ=632.8 nm). CO adsorption is found to proceed with a constant sticking coefficient up to a coverage of 0.5 ML, after which Langmuir adsorption kinetics are followed. The changes in the adsorption kinetics correlate quantitatively with structural changes in the CO monolayer. This technique monitors the surface coverage directly, and is simple and inexpensive to implement. As an optical surface probe it allows time-resolved measurements and can be extended to high-pressure environments.

Interaction of Oxygen with Thin Cobalt Films

The interaction of oxygen with thin cobalt films supported on oxidized Si(100) substrates was studied using Auger electron spectroscopy (AES) and work function changes (Δφ) at 130 and 300 K. Oxygen uptake curves were derived from the ratio changes of the OKLL and CoLMM Auger signals at 513 and 778 eV, respectively. They showed a constant sticking coefficient (so) up to ≈15 langmuirs, especially at 130 K, possibly indicating adsorption through a molecular precursor that diffuses over the chemisorbed layer. At 300 K, so = 0.2, while at 130 K, 0.5 ≥ so ≥ 0.3, depending on the dosing pressure. Up to ≈10 langmuirs O2 at 300 K, the work function increased 0.20 eV. At that coverage, the Co MVV Auger transitions indicated an oxide formation. From this coverage onward, the work function decreased, saturating at ≈80 langmuirs with a value ≈−1.2 eV below the clean Co surface, pointing to oxygen diffusion into the bulk.

Interaction of oxygen and Ni on W(110)

The adsorption of Ni on O-covered W(110) and of oxygen on Ni-covered W(110) and subsequent behavior on heating have been studied via thermal desorption, XPS, Auger, work function, and LEED measurements. Adsorption of Ni on O/W(110) does not lead to NiO formation, but on heating to T > 400 K to dewetting and at even higher temperatures to segregation which seems complete near 1000 K. After segregation Ni forms (111)-oriented islands, while O is forced into a p(1 × 8) structure, which corresponds to p(1 × 1) with regularly spaced defect lines every eighth (non-primitive) unit cell, running along 〈100〉 directions. There is evidence for a dead zone between Ni and O islands, as also observed in other systems. Ni desorption temperature is reduced appreciably, relative to Ni on clean W(110). Adsorption of oxygen on Ni W (110) occurs on a saturated Ni monolayer (corresponding to Ni W = 1.29 , i.e. virtually to the density of a Ni(111) plane) with constant sticking coefficient s = 0.74 at 90 K up to O W = 0.9 . Maximum O uptake corresponds to O Ni ≈ 1 or O W ≈ 1.3 . For lower Ni coverage, Ni W = 0.94 ( Ni 0.73 W(110) relative to saturation Ni coverage), s is initially higher but decreases more rapidly than on Ni 1 W (110) . In both cases O adsorption leads to disorder of the composite surface. For Ni W (110) chemisorption occurs up to O W = 0.3 at 90 K, accompanied by a linear increase in work function. Above this coverage oxidation sets in, as indicated by Ni core level shifts; φ remains constant to nearly saturation coverage, where a steep increase, attributable to O 2 species occurs. On heating these desorb at 150–200 K. NiO decomposes between 400 and 600 K with O becoming chemisorbed on the W surface. Additional heating leads to desorption of WO 2 between 1000 and 1250 K, with remaining O coverage O/W = 0.7. Ni-O segregation occurs above 600 K. Oxygen adsorption also occurs on Ni 1O 0.6/W(110) to the extent of O W = 0.7. Similarities and differences between these results and those obtained for O-Cu, O-Pd and O-Ag interactions on W(110) are discussed.

Ba deposition on Si (1 0 0)2 × 1

The Ba deposition on Si(1 0 0)2 × 1 at room temperature has been investigated by LEED, AES, TDS, EELS and work function measurements. Ba grows, though disordered, layer by layer with a constant sticking coefficient. Barium on Si(1 0 0)2 × 1 gives rise to two adsorption states which exhibit relatively high binding energy for Θ ≤ 2 ML, a strong Ba-Si ionic interaction and relatively low binding energy for Θ ≤ 2 ML where the Ba overlayer has a metallic character. Heating of the Ba-Si interface at T < 700°C promotes Ba-Si interaction, probably with a tendency to form silicide compounds. Higher temperatures cause the appearance of the 2 × 4, 2 × 1 and 2 × 3 diffraction patterns.

Dehydrogenation of cyclohexene on ordered Sn/Pt(111) surface alloys

The adsorption and dehydrogenation of cyclohexene on Pt(111) and two Sn/Pt(111) surface alloys has been studied since cyclohexene is a likely intermediate in the dehydrogenation of cyclohexane to benzene, a prototypical reaction for catalytic reforming. Our investigations used TPD, AES, LEED, and sticking coefficient measurements. The two ordered, Pt-Sn surface alloys were prepared by annealing monolayer amounts of Sn vapor-deposited onto Pt(111). Depending upon the conditions used, the annealed surface exhibited a p(2 × 2) or (√3 × √3)R30° LEED pattern corresponding to the (111) face of Pt 3Sn or a surface alloy of composition Pt 2Sn, respectively. Sticking coefficient measurements showed that the initial sticking coefficient, S 0, was equal to unity on all three substrates and determined a saturation coverage of 0.14 ML for the cyclohexene monolayer on all three substrates at 180 K; Sn in these surface alloys has no measurable effect on the adsorption kinetics and the monolayer coverage compared to Pt(111) at 180 K. Precursor mediated adsorption kinetics on all three surfaces are indicated by the constant initial sticking coefficient up to cyclohexene coverages of at least 0.02 ML. However, the cyclohexene binding energy decreases with increasing Sn concentration in the surface layer, indicating that Sn has a substantial electronic effect on cyclohexene adsorption on the Pt(111) surface. The di-σ bonded cyclohexene on the Pt(111) surface is changed to hydrogen-bonded cyclohexene on the (√3 × √3)R30°Sn / Pt(111) surface. This change is attributed to the stronger influence of Sn on the surface capacity for forming di-σ bonds than hydrogen bonds. At small cyclohexene coverages, the alloying of Pt with Sn increases both the activity and selectivity of gas phase benzene production under UHV conditions. At higher cyclohexene coverages, the self-poisoning of the surface by cyclohexene or reaction intermediates becomes the determining factor in the reactions of cyclohexene and the addition of Sn to the surface only slightly increases the selectivity of gas phase benzene production. However, carbon deposition on the surface is strongly suppressed.

NH 3-adsorption on Ge using a cylindrically shaped crystal

The low-temperature adsorption of ammonia was studied on a cylindrically shaped Ge crystal in the orientation range (001)-(113)-(111)-(110) using the methods of low-energy electron diffraction (LEED and SPALEED) and photoelectron spectroscopy (UPS, ARUPS and XPS). Below 100 K, ammonia condenses. At 110 < T < 200 K adsorption is molecular on all orientations. At 110 < T < 130 K, adsorption proceeds via a mobile precursor with constant sticking coefficient up to ML-saturation with the ML coverage depending slightly on orientation. The ionisation energy decreases by 2.3 eV indicating nitrogen end-down adsorption and a charge transfer to the surface. Up to 0.5 ML on (001), corresponding to one molecule per dimer, the adsorbate is orientated perpendicularly. Above 0.5 ML, a transition to a tilted and more weakly bound configuration occurs. On (111) and all other orientations, only this weakly bound tilted form is observed. On (001), the surface order and reconstruction remains whereas it decreases elsewhere. On (111), the intensity of the Ge 3d surface core level component decreases upon adsorption because a part of it is shifted to the bulk position. On (001), the clean Ge3d surface component is only slightly shifted by adsorption and an additional component of equal intensity is formed. We assign them to the up-atom and the down-atom, respectively, of the dimers after adsorption. At 200 K, adsorption occurs preferentially on (001) and its vicinals confirming that the stronger bond is related with the surface dimers. Adsorbed ammonia is dissociated under the influence of electron and high-energy photon radiation.

Sticking coefficients of adsorbing proteins

The protein sticking coefficient, φ, the fraction of collisions that result in adsorption, is a function of the molecular interactions between the protein and the surface. A random walk and diffusionto-capture model was used to describe the kinetics of protein adsorption. The assumption of a constant sticking coefficient leads to a first-order model of the kinetics. A solution of the problem of adsorption from a semi-infinite medium with first-order kinetics at the boundary was obtained by numerical simulation on the computer. The results of the computer simulations match the time dependence observed experimentally. A correlation was developed to estimate φ from experimental data. φ has been found to be in the range 10 −5 −10 −8 for several protein adsorption kinetic studies reported in the literature.

Interaction of CO with Cu + cations: CO adsorption on Cu 2O(100)

The interaction of CO with coordinately unsaturated Cu + cations has been investigated on the Cu 2O(100) surface with TDS and photoemission. CO adsorbs molecularly with a constant sticking coefficient for exposures up to 0.3 L, suggesting a precursor state and nonactivated adsorption at 120 K. The activation energy for desorption at low coverages is 16.7 kcal mol , significantly higher than most reported values for metallic copper surfaces, and in agreement with previous suggestions of a higher heat of adsorption for CO at Cu + cations in the Cu/ZnO methanol synthesis catalyst. In contrast to CO adsorption on metallic copper surfaces, photoemission demonstrates a significant decrease in the surface dipole upon adsorption due to a net charge transfer to the surface and the resulting polarization of the adsorbed CO molecules. These electronic changes and the higher heat of adsorption are due to an increased σ-donor character for CO as evidenced by a direct interaction with Cu 4p-derived electronic states. No evidence is seen for a direct interaction with the Cu 3d bands. The enhanced σ bonding relative to CO on metallic copper surfaces can be rationalized in terms of the decreased Cu 4s/4p-derived density of states of Cu + cations.

Synchrotron-radiation-induced deposition of boron and boron carbide films from boranes and carboranes: Decaborane

Boron has been deposited successfully on Si(111) from the synchrotron-radiation-induced decomposition of decaborane (14), i.e., B10H14. The rate of deposition is limited by the adsorption rate of decaborane (14) on the surface. In addition there is some indication that there is an activation barrier to dissociative adsorption. The synchrotron-radiation- induced growth rate of boron thin films from decaborane (14) is linear with coverage for a large range of thickness, suggesting a constant sticking coefficient for decaborane adsorption at room temperature.

Adsorption, desorption and dissociation of N 2 on W(110)

N 2 adsorbs molecularly on W(110) at T s ⩽ 100 K with high and nearly constant sticking coefficient ( s ≈ 0.7). Saturation coverage is estimated as N 2/ W = 0.7, from the p(4 × 1) LEED pattern seen at saturation and by analogy with CO adsorption. Desorption at ∼ 130–150 K seems to occur from more than a single binding state, even for partial coverages. The tightly bound component (75% of total N 2 adsorbed, independent of coverage) desorbs with first-order kinetics as shown by isothermal measurements with E = 10.3 kcal, v = 10 15 s −1. Approximately 10% of total N 2 is converted to atomically bound N during desorption. The amount converted is independent of desorption temperature over the range 128–148 K, suggesting an activation energy equal to that of desorption of the tightly bound component, i.e. ∼ 10 kcal. This suggests that dissociation from preadsorbed N 2 occurs via a different channel than dissociation from impinging energetic N 2 molecules.

Adsorption and reaction of chlorine with the titanium(0001) surface

The authors have studied the adsorption of chlorine on the Ti(0001) surface using Auger electron spectroscopy (AES), low-energy electron diffraction (LEED), and thermal desorption spectroscopy (TDS). Chlorine overlayers were generated by using an electrochemical source of molecular chlorine at temperatures ranging from 30 to 600/sup 0/C. Adsorption is dissociative with an approximately constant sticking coefficient at low coverages. The adsorbed chlorine is sensitive to electron beam stimulated desorption (ESD) but does not thermally desorb below 750/sup 0/C. They did not observe any ordered LEED patterns unless a high-coverage surface was annealed to a temperature of 620/sup 0/C. The resulting complex LEED pattern suggests that the adsorbed chlorine forms a coincidence lattice on the titanium surface. The data is consistent with the formation of a pseudo-halide surface layer related to TiCl/sub 3/.

Halide formation and etching of Cu thin films with Cl2 and Br2

Etching mechanisms of Cu thin films under the exposure of an argon ion beam and halogen molecules (Cl2, Br2) were examined in a UHV system using a quartz crystal oscillator and x-ray photoelectron spectrometer. Chlorine exposure at room temperature produced a very stable chlorine overlayer on the Cu surface. The passivation of Cu by chemisorbed Cl reduced the etch rate of Cu film under argon ion bombardment for pressure up to 10−3 Torr of chlorine. Cuprous chloride was formed at 210 K, where multilayers of Cl2 are adsorbed on the Cu surface. Br2, however, diffused into the Cu substrate at 10−6 Torr at room temperature. A linear growth rate of the halide film was observed. The etch rates of copper were not enhanced by ion bombardment in the presence of either halogen. Large amounts of free halogens were incorporated in the halide films. These results suggest that the halogenation of a copper surface is dominated by the unactivated place exchange mechanism with constant sticking coefficient. The relative flux ratios of neutral and ion species as well as surface temperature were found to play important roles in halide formation and in the etching of copper.

Growth rates of vapor grown filaments

Abstract The growth rates and rate-determining steps for filamentary crystals of varioys substances are surveyed. Such simple crystalline shapes yield reliable growth rate data and clear morphological information. Numerous reported data are classified in the following categories: (1) Growth rates which can be represented by the Hertz-Knudsen (HK) equation and a constant sticking coefficient. (2) Growth rates corresponding to the HK equation at a low temperature but becoming lower at high temperatures. (3) Growth rates approaching the rates given by the HK equation with increasing temperature. (4) Growth rates having a peak value at a specified temperature. For each category, growth models and rate-determining steps are discussed.

Nitridation of Si(111) by nitrogen atoms. II

Annealed Si(111) crystals have been reacted at T < 1050°C with nitrogen atoms produced by electron dissociation of N 2 gas near 10 −4 Torr. The initial stage of the reactions at T ≳ 850°C is rate limited by the incident nitrogen flux, has a constant sticking coefficient or order 1, and produces a well ordered “8 × 8” LEED pattern for surfaces free of impurities. The “8 × 8” LEED pattern is shown to arise from a coincidence lattice with real space unit vectors of length 8 11 times the length of the unreconstructed Si(111) unit cell. The previously reported quadruplet pattern is produced for surfaces with a few percent carbon contamination. These LEED patterns become more intense as the Si(111) 7 × 7 spots fade away, and arise from structures of monolayer thickness growing in islands. Auger and electron energy loss (ELS) signals of these monolayer structures are distinct from each other. After the completion of a reacted monolayer, the reaction proceeds more slowly and produces poorly ordered LEED patterns which gradually replace those mentioned above. The kinetics of the reactions following the quadruplet monolayer are faster than those following the “8 × 8”. Thermal desorption measurements indicate that the growth becomes limited by diffusion through the reacted layers for films about 25–30 Å thick. Auger and ELS signals and LEED patterns of thick multilayers formed on either monolayer structure are similar to each other and to those of the quadruplet monolayer. The reactions carried out at lower temperatures generate Auger and ELS signals which are different from those of the T ≳ 850°C reactions, and suggest the presence of thermodynamically unstable configurations produced by slower surface diffusion.

Interaction of H 2O with a clean and oxygen precovered Ni(110) surface studied by XPS

The adsorption and reaction of H 2O on clean and oxygen precovered Ni(110) surfaces was studied by XPS from 100 to 520 K. At low temperature ( T<150 K), a multilayer adsorption of H 2O 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 2O 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 2O 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.

Adsorption of N 2,CO and O 2 on Ni(110) at 20 K

On Ni(110) the adsorption of N 2, CO and O 2 was investigated by UPS at low temperatures down to 20 K. For all three gases chemisorptive bonding was found for the first adsorbate layer. CO and N 2 chemisorb molecularly and O 2 dissociatively. For CO and O the chemisorption state is well known. For N 2 we found a mixture of chemisorbed and physisorbed species already in the submonolayer region. This is concluded from angle-resolved UPS. At higher exposures physisorbed molecules were found on top of the chemisorbed layer. The peak positions of the valence levels as seen in the UP spectra are shifted by the same amount as the work function is changed by the underlying adsorbate layer. For the chemisorption layer a constant sticking coefficient was found for CO and N 2 whereas it decreased with coverage for O 2.