Surface Rate Processes Research Articles

Kinetic Monte Carlo simulations of surface growth during plasma deposition of silicon thin films

Based on an atomically detailed surface growth model, we have performed kinetic Monte Carlo (KMC) simulations to determine the surface chemical composition of plasma deposited hydrogenated amorphous silicon (a-Si:H) thin films as a function of substrate temperature. Our surface growth kinetic model consists of a combination of various surface rate processes, including silyl (SiH(3)) radical chemisorption onto surface dangling bonds or insertion into Si-Si surface bonds, SiH(3) physisorption, SiH(3) surface diffusion, abstraction of surface H by SiH(3) radicals, surface hydride dissociation reactions, as well as desorption of SiH(3), SiH(4), and Si(2)H(6) species into the gas phase. Transition rates for the adsorption, surface reaction and diffusion, and desorption processes accounted for in the KMC simulations are based on first-principles density-functional-theory computations of the corresponding optimal pathways on the H-terminated Si(001)-(2x1) surface. Results are reported for two types of KMC simulations. The first employs a fully ab initio database of activation energy barriers for the surface rate processes involved and is appropriate for modeling the early stages of growth. The second uses approximate rates for all the relevant processes to account properly for the effects on the activation energetics of interactions between species adsorbed at neighboring surface sites and is appropriate to model later stages of growth toward a steady state of the surface composition. The KMC predictions for the temperature dependence of the surface concentration of SiH(x(s)) (x = 1,2,3) species, the surface hydrogen content, and the surface dangling-bond coverage are compared to experimental measurements on a-Si:H films deposited under operating conditions for which the SiH(3) radical is the dominant deposition precursor. The predictions of both KMC simulation types are consistent with the reported experimental data, which are based on in situ attenuated total reflection Fourier transformed infrared spectroscopy.

Heterogeneous catalysis on the atomic scale

Information about the elementary processes underlying heterogeneous catalysis may be obtained by investigating well-defined single crystal surfaces. The success of this "surface science" approach for "'real" catalysis can be demonstrated, for example, with ammonia synthesis. The progress of catalytic reactions can be observed on an atomic scale by applying scanning tunneling microscopy and other surface physical techniques, as is shown with different examples in this paper: CO oxidation on a Pt(111) surface proceeds preferentially along the boundaries between adsorbed O and CO patches. Ru is practically inactive for the same reaction under lower pressure conditions but is transformed into RuO2 under atmospheric pressure conditions, where part of the surface Ru atoms function as coordinatively unsaturated sites (cus). In contrast, in the hydrogen oxidation reaction on Pt(111), an autocatalytic reaction step comes into prominence, and is responsible for the formation of propagating concentration patterns on the surface as a characteristic of nonlinear dynamics. Additionally, the limits of the concept of thermal equilibrium in surface rate processes are explored by applying ultrafast (femtosecond) laser techniques.

Heterogeneous catalysis on the atomic scale

Information about the elementary processes underlying heterogeneous catalysis may be obtained by investigating well-defined single crystal surfaces. The success of this “surface science” approach for “real” catalysis can be demonstrated, for example, with ammonia synthesis. The progress of catalytic reactions can be observed on an atomic scale by applying scanning tunneling microscopy and other surface physical techniques, as is shown with different examples in this paper: CO oxidation on a Pt(111) surface proceeds preferentially along the boundaries between adsorbed O and CO patches. Ru is practically inactive for the same reaction under lower pressure conditions but is transformed into RuO2 under atmospheric pressure conditions, where part of the surface Ru atoms function as coordinatively unsaturated sites (cus). In contrast, in the hydrogen oxidation reaction on Pt(111), an autocatalytic reaction step comes into prominence, and is responsible for the formation of propagating concentration patterns on the surface as a characteristic of nonlinear dynamics. Additionally, the limits of the concept of thermal equilibrium in surface rate processes are explored by applying ultrafast (femtosecond) laser techniques. © 2000 John Wiley & Sons, Inc. and The Japan Chemical Journal Forum Chem Rec 1:33–45, 2001

Surface Reconstruction and Rate Processes in Adsorbed Overlayers

Spontaneous or adsorbate-induced restructuring of surfaces can significantly influence the kinetics of elementary rate processes in adsorbed overlayers. Theoretically, this effect can be studied by employing statistical models of surface reconstruction. The review presented describes the results of simulations carried out in this field of surface science. The main attention is paid to such processes as desorption and surface diffusion. The systems under considerations are as follows: H/Si(001), H/W(001), CO/Pt(110), H/Cu(110), and H/Pt(001).

Evidence of the potential influence of planktonic community structure on the interannual variability of particulate organic carbon flux

The functioning of the biological pump during spring blooms was assessed with biogeochemical data from JGOFS (Joint Global Ocean Flux Study) process studies in the NE Atlantic during 1989 and 1990. A comparison of the integrated primary productivity signal (estimated by the 14C technique and from changes in ambient surface-water [tCO 2] and [NO 3 −]s during the spring blooms of the two years revealed close similarities. These observations suggest that the magnitude of the biological drawdown of CO 2 via photosynthetic activity during these periods was comparable. However, despite similarities in the magnitudes of these and other surface rate processes, sediment trap particulate organic carbon (POC) fluxes at 3100 m representing the spring bloom settlement events were 1.8 times greater in 1989 than in 1990. Taking into account the spatial and temporal resolutions of pelagic and deep trap datasets, these observations suggest that the coupling between organic carbon production in surface waters and its transfer to the deep ocean was stronger in the period studied in 1989 than in 1990. That is, the biological pump was more efficient in 1989. The size of the dominant phytoplankton species was observed to be the principal difference between the two spring bloom data sets. The potential influence of the observed algal size differences on the vertical POC flux was quantified from size-fractionated productivity data in conjuction with a food web-vertical flux model. The derived POC fluxes from the surface layer were two times greater in spring 1989 than in the bloom period in 1990, and a comparison of these flux estimates with those from other methods is favourable. Extrapolation of these derived shallow POC fluxes to 3100 m with existing empirical algorithms yields deep POC fluxes that are consistent with those collected by sediment traps at this depth. Differences in algal size between the two spring blooms can thus account for the observed interannual differences in deep-water POC fluxes without the need to invoke interannual differences in unmeasured mid-water processes. This work provides a clear demonstration that although observations of oceanic productivity may yield the input to the biological pump, they cannot, on annual timescales, reliably provide information on the efficiency of the pump in transferring carbon to the deep ocean.

Modeling the Kinetics of Heterogeneous Catalysis

A variety of microscopic processes occurring at the gas-solid interface (for example, adsorption, desorption, diffusion, reaction, surface reconstruction, and adsorbed layer ordering) play essential roles in heterogeneous catalysis. Much of the work devoted to studying these phenomena involves determining their kinetics. In this paper the authors provide a review of how the kinetics of these processes can be modeled. The methods which have been used to treat the kinetics of surface rate processes can be broadly divided into two categories, analytical and numerical. In the analytical approach, rate equations describing the time dependence of the fractional surface coverages, for instance, are written down. Generally, various approximations need to be made before these equations can be solved. In the numerical approach, which depends largely upon Monte Carlo techniques, the properties of various lattice gas models are studied using computer simulations. The authors focus on the numerical approach in this paper. 65 refs.

Kinetics of step-site filling for CO/Ni(9,1,1): A pulsed molecular beam-surface infrared study

Pulsed molecular beam-surface infrared measurements of the kinetics of CO populating step sites on Ni(9,1,1) are reported and interpreted in terms of elementary surface rate processes. An analytic model is developed to describe the distribution of CO between step and terrace sites in the equilibrium limit, and refine our previous determination of the binding energy difference between these sites to ΔEs-t=0.6±0.2 kcal/mol. Time-resolved surface infrared measurements indicate that the equilibrium step coverage is reached within 100 ms of the chemisorption event. This rapid migration across the (100) terraces to step sites implies a barrier to surface hopping of <5.5 kcal/mol. On a longer time scale of minutes, the CO population at step sites increases further as the equilibrium point is shifted by the dissociative adsorption of residual hydrogen. These slower step filling rates are described with a kinetic model, in which hydrogen adsorption is the rate-limiting step.

Kinetic modeling of surface rate processes

A hierarchical review is given of the modeling of surface rate processes. The following levels of sophistication of kinetic modeling are discussed (in the following order): (1) Langmuirian adsorption and desorption with random adsorbate distributions; (2) precursor-mediated adsorption and desorption in the absence of lateral interactions employing both kinetic and statistical approaches; (3) Langmuirian and precursor-mediated adsorption and desorption, with lateral interactions, treated using mean-field approximations; and (4) Monte Carlo simulations. Although this discussion encompasses the past sixty years of research in the field, emphasis is placed on the considerable advances in our understanding that have been made in the past thirty years.

The coverage dependence of the pre-exponential factor for desorption

The pre-exponential factor for desorption υ 0 is seldom measured as other than low surface coverage θ and is typically assumed to be independent of θ. Variations in υ 0 with θ are critically important in describing surface equilibrium and rate processes and in testing theories of desorption. It has often been overlooked that the functional depedence of υ 0 on θ may be readily determined from thermodynamic data used to determine isosteric heats of adsorption if the sticking coefficient S(θ) is known. We present υ 0(θ) values for approximately 45 adsorption species calculated from or reported in the literature for approximately adsorption systems. It is found that variations in υ 0 by factors of 10 6 are often observed. In approximately half of these examples υ 0 decreases by a factor of 10 3 or more and in the rest of the systems υ 0 is constant or decreases by less than 10 3; in very few cases does υ 0 increase significantly with coverage. Current theoretical models for υ 0 are briefly reviewed, and it is shown that none predicts such large coverage variations. However, analysis of low coverage models suggest that adsorbate-induced changes in the surface phonon spectrum may play a critical role.

Time-Resolved Electron Energy Loss Spectroscopy

Two recent instrumental improvements in high-resolution electron energy loss spectroscopy make possible the recording of complete surface vibrational spectra on the millisecond time scale. This is the first spectroscopic probe capable of directly measuring fundamental surface rate processes in real time with a resolution less than or equal to 1 millisecond. Such measurements are the key to understanding surface kinetics at the molecular level. This article summarizes experiments on the adsorption and decomposition of formic acid on Cu(100) to investigate the temperature and coverage dependence of the formate intermediate. Other results are cited that provide a detailed description of the decomposition of methanol on Ni(110). Also reported are direct measurements of the residence time of carbon monoxide on Cu(100) and the associated desorption kinetics.

Quantum statistical theory of laser/surface-catalyzed migration with application to surface rate processes

A microscopic Hamiltonian describing the dynamical processes of laser-induced migration of adatom is presented. The thermal-phonon-assisted and laser-stimulated migration rates are calculated via steepest descent method. Migration-limited rate processes are investigated by a diffusion equation with a diffusion constant governed by a master equation. Finally, applications to heterogeneous catalysis and microelectronics are discussed.

Dynamical model of selective vs. nonselective laser-stimulated surface processes. 2. Analytical methods and applications to surface rate processes

Dynamical model of selective vs. nonselective laser-stimulated surface processes. 2. Analytical methods and applications to surface rate processes

Surface rate processes and vapor pressures of magnetic solids

We present a theory of surface rate processes in magnetic solids that takes into account the coupling between magnetic and positional degrees of freedom of the atoms. This mechanism leads to an enhancement of the sublimation current over Arrhenius behavior as T c is approached from below, with a discontinuous derivative at T c . We also calculate the equilibrium vapor pressure, a quantity which exhibits marked departures from the behavior encountered in non-magnetic solids.

Concentration forcing of catalytic surface rate processes: Part I. Isothermal carbon monoxide oxidation over supported platinum

AbstractPeriodic feed switching between carbon monoxide/argon and oxygen/argon mixtures to a gradientless reactor has been found to significantly increase the average reaction rate as compared to the steady state rate, when the time averaged feed concentrations in both cases are stoichiometric. The rate enhancement achieved during periodic reactor operation is attributed to the attainment of more desirable surface concentrations on the platinum catalyst which dramatically increase the surface rate processes leading to carbon dioxide production.

Compensation Effect in Heterogeneous Catalytic Reactions including Hydrocarbon Formation on Clays

A SERIES of comparable reactions is said to exhibit compensation behaviour when the Arrhenius plots show systematic variations of the activation energy (E) and the pre-exponential factor (A) according to the expression Many examples of such behaviour have been reported for diverse reaction systems, but hitherto no explanation of the effect has been generally accepted. Indeed, it has been suggested1–3 that obedience of data from a sequence of reactions to equation (1) may arise through the method of analysis of the kinetic measurements or it may be an artefact. But before compensation behaviour is dismissed as a phenomenon of little theoretical significance, we suggest that in at least one field, heterogeneous catalysis, obedience to equation (1) may be considered to result from a useful mechanistic model. Many examples of compensation behaviour among reactions catalysed by solids have been reported and Bond4 has discussed the theoretical implications. The present preliminary report describes a further sequence of surface rate processes which exhibit the compensation effect and proposes a model to explain these observations.

Surface Rate Processes and Sensitivity of High Explosives

The nature and rate of linear surface regression of TNT, RDX, Tetryl, and PETN have been studied utilizing a hot-plate pyrolysis technique over surface temperatures ranging as high as 500°C. The data, when compared to similar data for benzoic acid, suggest that the primary surface rate process appears to be an endothermic melt flow having an apparent activation energy suggestive of heat transfer and viscous flow as a rate controlling step. These data, when used as a measure of surface heat dissipation, correlate very well with the measured values of impact sensitivity. Also, they offer an explanation for the fact that attempts to correlate impact and/or friction sensitivity with melting point alone could lead to misleading results when PETN and RDX are considered.