Supersonic Molecular Beam Techniques Research Articles

Nucleation of pentacene thin films on silicon dioxide modified with hexamethyldisilazane

Nucleation and growth of pentacene on silicon dioxide surfaces modified with hexamethyldisilazane, HN[Si(CH3)3]2, has been examined using supersonic molecular beam techniques and atomic force microscopy. Similar to growth on clean SiO2 surfaces, the rate of deposition at a fixed incident flux decreases with increasing kinetic energy of incident pentacene, indicative of trapping mediated adsorption. Unlike clean, unmodified SiO2 surfaces, however, growth on the modified surface exhibits characteristics of heterogeneous nucleation, where the maximum island density is independent of deposition rate. Further, islands in the submonolayer regime are two molecules high, unlike the one molecule high islands observed on clean SiO2.

Initial Adsorption Dynamics of O2 on Si(111)-7*7 Surface at Room Temperature

Initial adsorption dynamics of O2 on Si(111)-7×7surface at room temperature was investigated by a supersonic molecular beam technique in conjunction with real-time X-ray photoelectron spectroscopy using synchrotron radiation. The initial sticking probability and the saturation coverage as a function of the translational kinetic energy of impinging O2 were measured in the wide range from 0.03 eV, namely thermal gas generated by back-filling condition, to 2.3eV. We proposed a trapping-mediated process as a dominant adsorption mechanism at low kinetic energy region, while at high kinetic energies approximately over 0.06eV the direct adsorption overlapping the trapping-mediated one whose contribution became small with increasing incident energies were expected to be a dominant adsorption mechanism. A gradual increase of the saturation coverage with increasing kinetic energy was clearly observed in the incident energy of 0.4eV to 1.7eV where the direct adsorption took over. The peak area intensity over 1.7eV regions is approximately 1.8 times larger than that for 0.03eV. These results indicate that the incident energy essentially induces the activated adsorption corresponding to the enhancement of the further O2 dissociative adsorption even at room temperature.

Analysis of the Molecular Beam Inelastic Scattering at Solid Surfaces Based on the Theoretical Model of Pure and Classical Binary Collision

The influence on the gas-surface inelastic collision processes at the well defined surfaces of molecular structural anisotropy, corrugation of the interaction potential and the center-of-mass position of the impinging molecules has been investigated utilizing a supersonic molecular beam technique in conjunction with the recently developed model calculation based on the simple and pure classical binary-collision. The model calculations are found to reproduce the obtained experimental results in details.

Surface Products and Coverage Dependence of Dissociative Ethane Adsorption on Pt{110}-(1 × 2)

The dissociation of ethane on Pt{110}-(1 x 2) has been studied using supersonic molecular beam and temperature-programmed reaction techniques. The study unequivocally shows that the stable dissociation product of ethane on Pt{110}-(1 x 2) at all coverages is CCH2 at 350-400 K and CCH at 440 K. Temperature-programmed-reaction (TPR) experiments indicate that the CCH2 species decomposes to CCH with a reaction-limited peak temperature of 430 K. Above 450 K, the CCH species becomes unstable and decomposes with a peak temperature of 540 K. By 600 K, ethane dehydrogenates completely to form a surface carbon layer. The sticking probability is initially 0.02 at 370 K and 0.03 at 600 K and follows a linear (1-2theta) dependence for coverages of up to theta = 0.4 ML, where theta is defined as the number of C2Hx units per (1 x 2) unit cell. However, a much weaker coverage dependence at 800 K suggests that the carbon agglomerates into high-density islands.

Mass and lattice effects in trapping: Ar, Kr, and Xe on Pt(111), Pd(111), and Ni(111)

Abstract The trapping probabilities of argon, krypton, and xenon on Pd(1 1 1) and Ni(1 1 1) have been investigated using supersonic molecular beam techniques. The trapping probability of argon exhibits normal incident energy in a similar fashion on both Pd(1 1 1) and Pt(1 1 1) because the mass of argon is significantly less than the surface mass of either Pd or Pt. In contrast, dynamic corrugation in the gas–surface potential is observed for krypton trapping on Pt(1 1 1) and Pd(1 1 1), resulting in a decreased angular dependence of the trapping probability compared to argon. For xenon trapping on Pd significant lattice deformation during the gas–surface collision appears to give rise to total energy scaling. The trapping probability of xenon on Pd(1 1 1) remains high at unusually high incident kinetic energies due to the overall enhanced energy transfer from the incident atom to the lattice. Trapping probabilities of Ar, Kr, and Xe are significantly lower on Ni(1 1 1) than on either Pt(1 1 1) or Pd(1 1 1) despite the lower surface mass of the Ni atoms. This result is attributed to the lower binding energy of the rare gases on Ni(1 1 1) and the higher Debye temperature of Ni. The energy scaling of Ar trapping on Ni(1 1 1) is determined by static corrugation, but the energy scaling for Kr and Xe on Ni(1 1 1) may involve the effects of dynamic corrugation. In the latter cases, the greater stiffness of the nickel lattice decreases the dynamic corrugation relative to Pt(1 1 1) and Pd(1 1 1).

Inelastic scattering process of H2O on Pt(1 1 1) studied by supersonic molecular beam techniques

Abstract In order to understand the inelastic scattering process, the incident kinetic energy dependence of the angular intensity distribution is measured of H2O scattered from Pt(1 1 1) at 500 K. The comparisons of H2O scattering with Ne scattering from Pt(1 1 1) and with the prediction of the simple classical binary collision model reveal that not only phonon creation in the surface but also a large amount of H2O rotational mode excitation and/or chattering scattering takes place during collision, especially for the high incident kinetic energy conditions.

The nucleation and growth of silicon thin films on silicate glasses of variable composition using supersonic gas source molecular beam deposition

Supersonic molecular beam techniques have been used to study the nucleation and growth of Si thin films on glass surfaces of variable composition using Si2H6 as the precursor to film growth. We have examined, in particular, the early stages of growth using scanning electron microscopy. Making use of molecular beam techniques to control accurately the precursor exposure we have examined trends in the evolution of the Si island density as a function of the composition of the glass, x, in (2⋅SiO2)1−x(Al2O3⋅CaO)x. The silica composition (1−x) for these samples was varied between 0.25 and 0.75, and comparisons were also made to the nucleation of Si on SiO2 thin films made by thermal oxidation and Corning 1737 display glass. We have found that the incubation time τinc varies only weakly with substrate composition, increasing by only a factor of 3 over the range 1−x=0.25–1.0. Examination of a later stage of nucleation and growth, the time for coalescence, τcoal, indicated a stronger dependence on composition, and this metric varied by a factor of 8 over the same range of composition. These results indicate that the intrinsic reactivity of the surface scales with the silica content of the surface. The maximum island density shows a much stronger, superlinear dependence on silica content, increasing by a factor of 15 as 1−x increased from 0.25 to 1.0. For the silica rich compositions, i.e., SiO2 and 1737, Nmax is essentially independent of substrate temperature and the results can be interpreted by a model for nucleation that is purely heterogeneous, and where surface diffusion plays a minimal role. In contrast, on the most silica dilute glass surface (1−x=0.25), Nmax exhibits an Arrhenius temperature dependence with an apparent activation energy of 1.1 eV. Coupled with the observation of a broader island size distribution on this surface, we conclude that surface diffusion plays a role in nucleation and growth on this silica dilute surface, possibly via Ostwald ripening.

Effects of atomic hydrogen on the selective area growth of Si and Si1−xGex thin films on Si and SiO2 surfaces: Inhibition, nucleation, and growth

Supersonic molecular beam techniques have been used to study the nucleation of Si and Si1−xGex thin films on Si and SiO2 surfaces, where Si2H6 and GeH4 have been used as sources. A particular emphasis of this study has been an examination of the effects of a coincident flux of atomic hydrogen. The time associated with formation of stable islands of Si or Si1−xGex on SiO2 surfaces—the incubation time—has been found to depend strongly on the kinetic energy of the incident molecular precursors (Si2H6 and GeH4) and the substrate temperature. After coalescence, thin film morphology has been found to depend primarily on substrate temperature, with smoother films being grown at substrate temperatures below 600 °C. Introduction of a coincident flux of atomic hydrogen has a large effect on the nucleation and growth process. First, the incubation time in the presence of atomic hydrogen has been found to increase, especially at substrate temperatures below 630 °C, suggesting that hydrogen atoms adsorbed on Si-like sites on SiO2 can effectively block nucleation of Si. Unfortunately, in terms of promoting selective area growth, coincident atomic hydrogen also decreases the rate of epitaxial growth rate, essentially offsetting any increase in the incubation time for growth on SiO2. Concerning Si1−xGex growth, the introduction of GeH4 produces substantial changes in both thin film morphology and the rate nucleation of poly-Si1−xGex on SiO2. Briefly, the addition of Ge increases the incubation time, while it lessens the effect of coincident hydrogen on the incubation time. Finally, a comparison of the maximum island density, the time to reach this density, and the steady-state polycrystalline growth rate strongly suggests that all thin films [Si, Si1−xGex, both with and without H(g)] nucleate at special sites on the SiO2 surface, and grow primarily via direct deposition of adatoms on pre-existing islands.

The adsorption dynamics of small alkanes on ( [formula omitted]) surfaces of platinum group metals

The adsorption dynamics of methane, ethane, propane, and n-butane on Ni(1 1 1) as well as methane and n-butane on Pd(1 1 1) have been investigated using super-sonic molecular beam techniques, and stochastic trajectory simulations. For each alkane the initial adsorption probability was measured as a function of incident energy and incident angle, and compared with corresponding values for other alkanes on Pd(1 1 1) and Pt(1 1 1). In general, at a fixed incident energy and angle, the trapping probability is highest on Pd(1 1 1), followed by Pt(1 1 1) and Ni(1 1 1), a result which deviates from the prediction of simple mass matching arguments [Surf. Sci. 187 (1987) 67]. The lower binding energy of the alkanes and the higher lattice force constant for Ni(1 1 1) compared with Pt(1 1 1) and Pd(1 1 1) give rise to the lower trapping probability. Three-dimensional stochastic trajectory simulations for alkane trapping on the three metals clearly indicate that incoming molecules lose considerable energy to Pd lattice vibrations, resulting in a high trapping probability. On the other hand, the stiffer Ni lattice prevents the excitation of surface phonons, consistent with experimental results. A simple scaling principle based only on surface masses and Debye temperatures was applied to scale the trapping probabilities of methane, ethane, and propane on Pd(1 1 1) and Ni(1 1 1) from that on Pt(1 1 1), and quantitative agreement was obtained within about 30%.

Oxidation of Aqueous HF-treated Si(001) Surface Induced by Translational Kinetic Energy of O2 at Room Temperature

The oxidation induced by the translational kinetic energy of O2 on the Si(001) surface treated with HF solution were investigated by combining synchrotron radiation photoemission spectroscopy with the supersonic molecular beam techniques. The oxidation at room temperature did not progress up to 3600 L of O2 exposure with incident energy of 0.04 eV, whereas the oxidation states of up to Si4+ species were formed in the case of 3.0 eV. The oxide-layer thickness was estimated to be 0.26 nm at the final oxidation stages. We concluded that the Si atoms at the top layers were oxidized by the incident energy of 3.0 eV.

Density functional study of gas–surface reactivity: GeH 4 dissociative adsorption onto semiconductor surfaces

Ultrasoft pseudopotential total energy calculation based on density functional theory (DFT) with generalized gradient approximation (GGA) have been used to investigate (1) the possible energy pathway for direct dissociative adsorption of GeH 4 onto both Si(001) and Ge(001) surfaces to evaluate their gas–surface reactivity in comparison with measured gas–surface reactivity using supersonic molecular beam techniques and (2) the coupling of internal distortion of GeH 4 with SiSi dimer distortion on Si(001) and GeGe dimer distortion on Ge(001) surfaces, respectively, at their transition states to understand how experimental variables such as incident kinetic energy of GeH 4 and substrate temperature govern their gas–surface reactivity. Firstly, our calculated transition state structures strongly indicate the gas–surface reactivity for direct dissociative adsorption of GeH 4 onto both Si(001)-(2×2) and Ge(001)-(2×2) surfaces increases with both incident kinetic energy and substrate temperature. But the increase of gas–surface reactivity with increasing substrate temperature, i.e. inducing thermal vibration, is not as significant as that with increasing incident kinetic energy of GeH 4, i.e. inducing bond-breaking distortion. Secondly, our evaluated gas–surface reactivity for direct dissociative adsorption of GeH 4 onto Si(001) is approximately a factor of 6 better than that onto Ge(001). This calculated result is in reasonable agreement with its counterpart (approximately a factor of between 10 and 2 depending on the incident kinetic energy) derived from measured gas–surface reactivity using supersonic molecular beam techniques. Finally, we attribute this higher evaluated gas–surface reactivity for the direct dissociative adsorption of GeH 4 onto Si(001) surface in comparison with that onto Ge(001) surface to (1) the less energy conversion from incident kinetic energy of GeH 4 to the energy of internal distortion of GeH 4 at the transition state and (2) the more populated vibrational levels of SiSi dimer on Si(001) surface than that of GeGe dimer on Ge(001) surface provided by the same substrate temperature to reach their transition states.

Density Functional Study of the Effect of SiH4/GeH4 and Si(001)/Ge(001) on Gas‐Surface Reactivity during Initial Dissociative Adsorption

AbstractUltrasoft pseudopotential total energy calculation based on density functional theory (DFT) with generalized gradient approximation (GGA) has been used to investigate 1) the energetic profile for the initial dissociative adsorption of XH4 (X = Si and Ge) onto Si(001) and Ge(001) surfaces to evaluate their gas‐surface reactivity in comparison with relevant measured gas‐surface reactivity using supersonic molecular beam techniques, and 2) the effect of different gaseous molecular precursors, i.e. XH4 (X = Si and Ge), and different surfaces, i.e. Si(001) and Ge(001), on their gas‐surface reactivity during initial dissociative adsorption. Our evaluated gas‐surface reactivity for GeH4 is approximately a factor of 18.45 better than that for SiH4 on Si(001)‐(2×2) surface. This calculated result is about three to four times higher than observed gas‐surface reactivity (as much as a factor of 5 depending on the incident kinetic energy) derived from measured gas‐surface reactivity using supersonic molecular beam techniques. We believe that the better evaluated gas‐surface reactivity for GeH4 than SiH4 is due to 1) the forming of a stronger bond of Si‐H between H within GeH4 and buckled‐down Si atom on the Si(001)‐(2×2) surface and 2) the smaller distortion of Ge‐H bond within GeH4 at the transition state. Additionally, our evaluated gas‐surface reactivity for SiH4 on Si(001)‐(2×2) surface is approximately a factor of 21.69 better than SiH4 on Ge(001)‐(2×2) surface. This calculated result is about two times higher than observed gas‐surface reactivity. We attributed this better evaluated gas‐surface reactivity for SiH4 on Si(001)‐(2×2) surface to 1) the smaller distortion of Si‐H bond within SiH4 and 2) the nature of weaker bond of Ge‐H between H within SiH4 and buckled‐down Ge atom on Ge(001)‐(2×2) surface in comparison with that of stronger bond of Si‐H between H within SiH4 and buckled‐down Si atom on Si(001)‐(2×2) surface even though there is the slightly shorter bond length of Ge‐H between H within SiH4 and buckled‐down Ge atom on Ge(001)‐(2×2) surface at the transition state.

The Adsorption Dynamics of Molecular Methane, Propane, and Neopentane on Pd(111): Theory and Experiment

The molecular adsorption dynamics of methane, propane, and neopentane on Pd(111) was studied using supersonic molecular beam techniques and stochastic trajectory simulations. The sticking probability of the alkanes was measured as a function of incident energy, incident angle, and alkane coverage. In each case, the trapping probability decreased with increasing incident translational energy, which is expected for nonactivated molecular trapping. Nonnormal energy scaling was observed for the trapping probability of all of these alkanes, indicating corrugation of the gas−surface interaction potentials. The modified Kisliuk model was employed to describe the self-coverage dependence of the alkane on Pd(111). The trapping probability of each of these alkanes on Pd(111) is higher than that on Pt(111), consistent with the lower mass of palladium. Using the Morse potential for the methyl (methylene)−platinum two body potential obtained from the stochastic trajectory analyses of alkanes trapping on Pt(111), the trapping probabilities of alkanes on Pd(111) were predicted nearly exactly. Energy transfer calculations for alkane trapping on Pd(111) clearly indicate that excitation of the lattice vibrations and molecular cartwheeling rotations are the principal mechanisms for determining trapping.

IN-SITU CORE-LEVEL PHOTOELECTRON SPECTROSCOPY OF ADSORBATES ON SURFACES INVOLVING A MOLECULAR BEAM — GENERAL SETUP AND FIRST EXPERIMENTS

In this contribution we introduce a new apparatus combining high-resolution photoelectron spectroscopy and supersonic molecular beam techniques designed to follow simple surface reactions like adsorption, desorption or oxidation in situ in a time- and temperature-resolved manner. Using high brightness synchrotron radiation high resolution core-level spectra can be obtained in less than 3 s. The molecular beam allows for local pressures of up to 1 · 10-5 mbar and translational energies of the molecules of up to 2 eV in the case of CO. Preliminary examples of CO adsorption on Pt(111) are shown, where kinetic parameters can be derived from the coverage dependence of the different adsorption sites.

酸素分子の並進運動エネルギーで誘起されるＳｉ（００１）表面酸化状態の放射光光電子分光法によるリアルタイム観察

Employing Si-2p and O-1s photoemission spectroscopy using monochromated synchrotron radiation and the supersonic molecular beam technique, we have performed “real time” in situ observations of oxidation states on the Si (001) -2×1 surface at room temperature. High-resolution Si-2p photoemission spectra, which unambiguously resolve components for oxidation states [Si1+, Si2+, Si3+ and Si4+], were successfully measured requiring only 43 sec per spectrum. We found that the Si4+ species gradually increased to reach the oxide thickness of 0.57 nm just after the saturation of Si1+, Si2+ and Si3+ species with a translational energy of 2.9 eV.

Ar scattering on Ru (0001): a comparison to the washboard modelPresented at the Stereodynamics 2000 Conference on Dynamics and Stereodynamics of Chemical Reactions, El Escorial, Madrid, December 1–5, 2000.

The scattering of Ar atoms on a Ru (0001) surface has been studied by applying supersonic molecular beam techniques. Variation of the incidence energy (Ei) results in a rich variety of angular and energy distributions. At a low energy, clear zero-order diffraction has been identified. At higher energies (≥1 eV), argon scattering of Ru (0001), however, seems to be dominated by a totally different mechanism, which appears to be almost quantitatively reproduced by the washboard model, developed by Tully (J. C. Tully, J. Chem. Phys., 1990, 92, 680, ). This is surprising since his comparison of the washboard model to a reference of classical trajectory calculations for the Ar/Pt (111) system showed essential deviations. Our attempt to reproduce the experimental Ar/Ru (0001) data by similar simulations was unsuccessful, most likely due to shortcomings in the description of the solid state.

Real time observation of oxygen chemisorption states on [formula omitted] during supersonic oxygen molecular beam irradiation

Employing Si 2p and O 1s photoemission spectroscopy using monochromated synchrotron radiation and the supersonic molecular beam technique, we have performed real time in situ observations of oxidation states on Si(0 0 1)-2×1 at room temperature. High-resolution Si 2p photoemission spectra, which unambiguously resolve oxide components [Si 1+, Si 2+, Si 3+ and Si 4+], were successfully measured requiring only 43 s per spectrum. We found that the Si 4+ species gradually increases to reach the oxide thickness of 0.57 nm just after the saturation of Si 1+, Si 2+ and Si 3+ species with a translational energy of 2.9 eV.

Si 2p and O 1s photoemission from oxidized Si(0 0 1) surfaces depending on translational kinetic energy of incident O 2 molecules

The influence of translational kinetic energy of incident O 2 molecules for the passive oxidation of the partially oxidized Si(0 0 1) surface has been studied by photoemission spectroscopy. The incident energy of O 2 molecules was controlled up to 3 eV by a supersonic molecular beam technique. Two incident energy thresholds (1.0 and 2.6 eV) were found out in accordance with the first-principle calculations. Si 2p and O 1s photoemission spectra measured at representative incident energies showed the incident energy induced oxidation at the backbonds of the dimer and the second layer (subsurface) Si atoms. Moreover, the difference of oxygen chemical bonds was found out to be as the low and the high binding energy components in the O 1s photoemission spectra. They were assigned to bridge sites oxygen and dangling bond sites oxygen, respectively.

The adsorption dynamics of molecular carbon dioxide on Pt( [formula omitted]) and Pd( [formula omitted

The dynamics of the nondissociative adsorption of carbon dioxide on Pt(1 1 1) and Pd(1 1 1) were investigated by supersonic molecular beam techniques at a surface temperature of 80 and 60 K, well below the desorption temperatures of 95 and 88 K, respectively. The initial trapping probability of CO 2 on both Pt(1 1 1) and Pd(1 1 1) scale with normal and near-normal energy scaling functions, indicating smooth gas–surface potentials. On the other hand, trapping on the carbon dioxide-saturated Pt(1 1 1) exhibits a more corrugated gas–surface potential, on which trapping scales as E Tcos 0.7 θ. The probability of carbon dioxide trapping increases with self-coverage and is successfully described by the modified Kisliuk model, indicating that trapping into the extrinsic precursor state is more efficient than trapping on the clean surface. The coverage of carbon dioxide into Pt(1 1 1) saturates at about 0.275 ML at a surface temperature of 80 K as a result of beam exposure at normal incidence at 8.5 kJ/mol. The stochastic, three-dimensional classical dynamics for the simulation of CO 2 on Pt(1 1 1) indicates that the excitation of surface phonons on the first bounce determines trapping on Pt(1 1 1). The same set of Morse potentials were employed to predict the CO 2 trapping on Pd(1 1 1). In general, the prediction agrees reasonably with the experimental results. The difference between the calculated trapping probabilities from simulations and the experimental values is within 20%.

The prediction of trapping probabilities for ethane by molecular dynamics simulations: scaling from Pt [formula omitted] to Pd [formula omitted

The dynamics of ethane trapping on Pd(1 1 1) were investigated by supersonic molecular beam techniques and three-dimensional stochastic trajectory simulations. The initial trapping probability was measured over the range of incident energy, E T, from 10 to 34 kJ/mol and incident angles, θ, from 0° to 45° at a surface temperature of 95 K. The trapping probability scales with E Tcos 0.9 θ, indicating a corrugated gas–surface potential. Stochastic trajectory simulations utilizing a Morse potential whose parameters were obtained experimentally from ethane trapping on Pt(1 1 1) [J. Chem. Phys. 104 (1996) 3134] predict the corresponding experimental values of the initial trapping probability of ethane trapping on Pd(1 1 1) to within 30%. Calculations of energy transfer for ethane after the first bounce on Pd(1 1 1) clearly indicate that vibrational excitation of the lattice phonons account primarily for the increase in trapping probabilities of ethane on Pd(1 1 1) compared with Pt(1 1 1). The enhancement in phonon excitation results from increasing the mass ratio of ethane to the metal atoms.