Trapping-mediated Chemisorption Research Articles

Trapping-Mediated Chemisorption of Ethylene onSi(001)−c(4×2)

Adsorption of ethylene molecules on Si(001)-c(4 x 2) was studied using scanning tunneling microscopy at low temperatures. Ethylene molecules trapped at the surface at 50 K were imaged only after decay to chemisorption, each bonding to a Si dimer. Atomic-scale observations of temperature-dependent kinetics show that the decay exhibited Arrhenius behavior with the reaction barrier of 128 meV in clear evidence of the trapping-mediated chemisorption, however, with an anomalously small preexponential factor of 300 Hz. Such a small prefactor is attributed to the entropic bottleneck at the transition state caused by the free-molecule-like trap state.

Trapping-mediated chemisorption of disilane on Si(100)-2×1

Disilane adsorption probabilities have been measured on Si(100)-2×1 over a wide range of incident kinetic energies, incident angles, and surface temperatures using supersonic molecular beam techniques. The trapping-mediated chemisorption mechanism is shown to be the dominant adsorption pathway under the conditions investigated. The first step in such a mechanism, namely trapping into the physical adsorption well, has been studied directly via measurements at a surface temperature of 77 K. As expected, the trapping probability drops with increasing kinetic energy, but nearly 50% of incident molecules trap at 1 eV incident energy, indicating that trapping is quite efficient over a wide range of translational energies. Chemisorption probability values measured at higher surface temperatures are fit to a simple trapping-mediated chemisorption model that can be used to predict adsorption probabilities over a wide range of conditions. Measurements of the chemisorption probability at 500 K are independent of incident angle at kinetic energies of 0.75 eV and below. However, trapping probabilities measured at 77 K are shown to decrease with increasing angle of incidence at kinetic energies of 0.6 eV and above. This unusual effect is discussed in terms of molecular scattering during parallel momentum accommodation. In order to investigate the effect of surface hydrogen formed as a result of disilane decomposition, adsorption probabilities were measured as a function of monohydride coverage as well. On a monohydride-saturated surface the trapping probability is found to be lower than on a bare surface, most likely due to a decreased disilane physical adsorption binding energy compared to the bare surface. Also, the trapping probability varies linearly with hydrogen coverage between bare-surface and monohydride-saturated values. On the other hand, the hydrogen coverage dependence of the chemisorption probability is found to follow a simple second-order kinetic scheme based on chemisorption occurring at two vacant surface sites.

Interaction of cyclobutane with the Ru(001) surface: Low-temperature molecular adsorption and dissociative chemisorption at elevated surface temperature

We have studied the interaction of cyclobutane with the hexagonally close-packed Ru(001) surface. High-resolution electron energy loss spectroscopy (HREELS) has been used to identify the vibrational modes of both c-C4H8 and c-C4D8 adsorbed at 90 K as a function of cyclobutane exposure. We have observed a vibrational mode not observed in the gas phase at 2600 cm−1 (2140 cm−1) which is attributed to the strong interaction of the cyclobutane C–H (C–D) bonds with the ruthenium surface. Two different adsorption geometries for cyclobutane on Ru(001) have been proposed based on the dipolar activity of this softened C–H mode. We have also measured the trapping-mediated dissociative chemisorption of both c-C4H8 and c-C4D8 at surface temperatures between 190 and 1200 K. The measured activation energies with respect to the bottom of the physically adsorbed well for c-C4H8 and c-C4D8 are 10 090±180 and 10 180±190 cal/mol, respectively. The trapping-mediated chemisorption of cyclobutane is believed to occur via C–C bond cleavage, as judged by the absence of a kinetic isotope effect. The measured ratios of the preexponential factors for desorption relative to reaction of 21±2 and 47±4 for c-C4H8 and c-C4D8 respectively, are in the expected range considering the greater entropy gain associated with the transition state for desorption relative to the transition state for C–C bond cleavage.

Dissociative chemisorption of methane on Ir(111): Evidence for direct and trapping-mediated mechanisms

Molecular beam and bulb gas techniques were employed to study dissociative chemisorption of methane on Ir(111). The initial dissociative chemisorption probability (S0) was measured as a function of incident kinetic energy (Ei), surface temperature, and angle of incidence (θi). As the incident kinetic energy increases, the value of S0 first decreases and then increases with Ei indicating that a trapping-mediated chemisorption mechanism dominates methane dissociation at low kinetic energy, and a direct mechanism dominates at higher kinetic energies. The values of the reaction probability determined from molecular beam experiments of methane on Ir(111) are modeled as a function of Ei, θi, and surface temperature. These fits are then integrated over a Maxwell–Boltzmann energy distribution to calculate the initial chemisorption probability of thermalized methane as a function of gas and surface temperature. The calculations are in excellent agreement with results obtained from bulb experiments conducted with room-temperature methane gas over Ir(111) and indicate that a trapping-mediated pathway governs dissociation at low gas temperatures. At the high gas temperatures characteristic of catalytic conditions, however, these calculations indicate that a direct mechanism dominates methane dissociation over Ir(111). These dynamical results are qualitatively similar to the results of a previous study of methane dissociation on Ir(110), although the reactivity of thermalized methane is approximately an order of magnitude higher on the (110) surface of iridium.

Direct and trapping-mediated dissociative chemisorption of methane on Ir(111)

The initial probabilities of dissociative chemisorption 13CH 4 and CD 4 have been measured on the close-packed Ir(111) surface at both low and high total pressures. At pressures below ∼ 10 −3 Torr, trapping-mediated dissociative chemisorption is the dominant reaction mechanism since the gas temperature is equal to the wall temperature of 300 K, with activation energies for C H (C D) bond cleavage of 12.6 kcal/mol for 13CH 4 and 13.7 kcal/mol for CD 4 with respect to the gas phase energy zero of the methane infinitely far from the surface and at rest. For both methane isotopomers the ratio of the pre-exponential factor for desorption relative to that of reaction is ∼ 180 due to the larger phase space available to the molecule for desorption relative to reaction. The measured differences in C H versus C D bond activation for methane are attributed to zero-point energy differences between each isotopomer and point to classical over-the-barrier reaction dynamics as the reaction pathway for trapping-mediated chemisorption. The incident kinetic energy of the impinging methane was increased by diluting the methane in argon at a total pressure of 1 Torr which raised the gas temperature of the methane to the surface temperature. In this case, methane chemisorbs dissociatively via both trapping-mediated and direct C H bond activation. Direct activation of 13CH 4 is characterized by an activation energy of 17.4 kcal/mol and a pre-exponential factor of ∼ 0.61, while direct activation of CD 4 occurs with an activation energy of 18.1 kcal/mol and the same pre-exponential factor. These activation energies are averages of a distribution of activation barriers which occur because of the nature of the multidimensional potential energy surface describing the interaction of the methane with the surface.

The dissociative chemisorption of cyclopropane on Ir(110)

We have employed molecular beam techniques to investigate the dissociative chemisorption of cyclopropane on Ir(110) as a function of beam translational energy, Ei, from 1.5 to 48 kcal/mol, and surface temperature, Ts, from 85 to 1200 K. For Ts=85 K, c-C3H6 is molecularly adsorbed on Ir(110) with a trapping probability, ξ, of 0.97 at Ei=1.5 kcal/mol and ξ=0.90 at Ei=5 kcal/mol. For Ei≤5 kcal/mol, c-C3H6 is dissociatively adsorbed through a mechanism of trapping-mediated chemisorption, with initial probabilities of chemisorption, Pa, decreasing with increasing surface temperature from the intrinsic trapping probability at Ts=150 K, to Pa<0.05 above Ts=1000 K. The activation energy for trapping-mediated chemisorption of c-C3H6, referenced to the bottom of the physically adsorbed well and attributed to C–C bond cleavage, is 3.6±0.2 kcal/mol. For Ei≥10 kcal/mol, direct dissociative chemisorption increasingly contributes to the overall measured initial probability of chemisorption of cyclopropane. The initial probability of direct dissociative chemisorption of c-C3H6 increases approximately linearly from Pa=0.1 at Ei=10 kcal/mol, to Pa=0.5 at Ei=45 kcal/mol. No isotope effect is observed for the direct dissociative chemisorption of c-C3D6 for beam translational energies of 17 to 48 kcal/mol, indicating that C–C bond cleavage is the initial reaction coordinate for direct chemisorption of cyclopropane on Ir(110).

Surface morphology of homoepitaxial silicon thin films grown using energetic supersonic jets of disilane

The effect of different chemical mechanisms for adsorption, which depend on the incidence energy of disilane, on the film morphology is investigated by comparing deposition by high (∼2 eV) kinetic energy disilane jets, direct chemisorption; low (∼0.09 eV) kinetic energy disilane jets and ultra high vacuum chemical vapor deposition, trapping-mediated chemisorption. For substrate temperatures of 500–550 °C the mechanism for adsorption does not influence the film morphology as observed for films up to 3300 Å thick, which are comparable in smoothness to the starting substrate, as determined by atomic force microscopy.

Kinetics and dynamics of nitrogen adsorption on Ru(001): evidence for direct molecular chemisorption

The experimentally determined value of the molecular chemisorption probability, S 0 has an interesting dependence on incident kinetic energy, E i, and incident angle, θ i, for the N 2/Ru(001) system. An increase in the value of S 0 was observed at θ i = 60° as E i was increased from 0.2 to 0.4 eV providing evidence for a direct, activated, mechanism for molecular chemisorption. A simple model of the direct mechanism for high kinetic energy adsorption combined with a hard-cube model to describe trapping-mediated chemisorption at low kinetic energies can account qualitatively for the angular dependence at all incident energies.

Effect of preadsorbates on the dissociation dynamics of ethane on Ir(110)

We have examined the role of the preadsorbates oxygen and potassium in modifying the dynamics and kinetics of ethane activation on Ir(110). Although the effect of these adsorbates is commonly interpreted within the context of promotion and poisoning, we have followed a quantitative approach in determining the influence of O on the trapping-mediated chemisorption and that of O and K on the direct dissociative chemisorption of this alkane. Using a supersonic molecular beam, the direct dissociative chemisorption of perhydrido- and perdeuteroethane on Ir(110) was investigated using translational beam energies of 12–30 kcal/mole, parametric in oxygen and potassium precoverage from the clean surface to the adsorbate-saturated surface. We have observed that the probability of direct dissociative chemisorption decreases in the presence of dissociatively adsorbed oxygen as well as potassium. The magnitude of the poisoning is considered within a one-dimensional tunneling model and the extent of poisoning of both oxygen and potassium is quantified with a model that yields a range parameter of 8.5 Å for O and 13 Å for K. For trapping-mediated chemisorption using beam energies of 1–5 kcal/mole, preadsorption of 0.06 monolayers O was found to lower the activation energy of perhydrido- and perdeuteroethane activation by 400 and 300 cal/mole, respectively, with no effect on the preexponential factor, while a surface saturated with oxygen lowered these activation energies by 800 and 700 cal/mole, also with no effect on the preexponential factor.

Effect of internal energy on the dissociative chemisorption of oxygen on Ir(110)

We have investigated the effect of internal energy on the initial probability (in the limit of zero surface coverage) of dissociative chemisorption of O 2 on Ir(110) using supersonic molecular beam methods. Measurements have been carried out at six beam energies ranging from 3.3 to 9.2 kcal mol and at surface temperatures of 400 and 1000 K. Under these conditions the initial probabilities of chemisorption lie between approximately 0.70 and 0.85. Chemisorption is dominated by a trapping-mediated mechanism via a physically adsorbed state at 3.3 kcal mol and below. With increasing translational energy, contributions to adsorption via direct access to a molecularly chemisorbed state become dominant. At each of the six translational energies investigated, beams were prepared with gas temperatures of 290 K (internally “cold”) and 850 K (internally “hot”). We have observed an increase in the probability of chemisorption due to internal energy for this highly reactive system. For a surface temperature of 400 K, there is an increase of about 0.06 in the initial probability of chemisorption of O 2 for the internally hot beam at the lowest translational energies. For a surface temperature of 1000 K, increased initial probabilities of chemisorption for the internally hot beam are also observed, although the enhancement (0.03) is not so large as at the lower surface temperature for the same translational energies. For each surface temperature the enhancement in the initial probability of chemisorption is greatest at the lowest beam energies. We conclude that the enhanced chemisorption is due primarily to increased reactivity, due to internal energy, of the physically adsorbed precursor in the trapping-mediated chemisorption of oxygen.

Dynamics of the interaction of ethane with Ir(110)-(1×2)

Experimentally determined values of the initial adsorption probability of ethane on Ir(110)-(1×2) are presented which probe the dynamics of the interaction. The data were obtained from supersonic molecular beam measurements with an incident kinetic energy Ei ranging between 1.2 and 24 kcal/mol, surface temperatures TS between 77 and 550 K, and incident angle θi between 0° and 45°. Experimentally determined values of the initial trapping probability ζ0 of ethane into a physically adsorbed state at TS=77 K as a function of Ei and θi and experimentally determined values of the initial probability of dissociative chemisorption S0 as a function of Ei,θi, and TS are presented. The value of ζ0 is found to decrease with increasing Ei consistent with the fact that an increasingly larger fraction of the incident kinetic energy must be dissipated in order for the molecule to physically adsorb. The initial trapping probability has a relatively weak dependence on θi such that the value of ζ0 is found empirically to scale as Ei cos0.5 θi. Two distinct mechanisms of dissociative chemisorption on the bare surface are revealed. At low Ei a temperature-dependent trapping-mediated chemisorption mechanism dominates, while at relatively high Ei a temperature-independent direct mechanism dominates. For Ei less than 13.4 kcal/mol, the value of S0 decreases rapidly with increasing TS, consistent with a trapping-mediated mechanism. For a surface temperature of 154 K, S0 decreases with increasing Ei for 1.2≤Ei≤13.4 kcal/mol, in a manner similar to that for the molecular trapping probability. The data in the low Ei regime also support quantitatively a kinetic model consistent with a trapping-mediated chemisorption mechanism. The difference in the activation energies for desorption and chemisorption from the physically adsorbed, trapped state Ed−Ec is 2.2±0.2 kcal/mol. In the trapping-mediated chemisorption regime, the value of S0 is found to be rather insensitive to incident angle, scaling with Ei cos0.5 θi just as for trapping of molecular ethane into a physically adsorbed state. For a normal energy Ei cos2 θi greater than 8 kcal/mol, chemisorption via a direct mechanism becomes significant and increases with increasing Ei. Values of S0 in the direct chemisorption regime scale with normal energy and are independent of TS over the range from 350 to 1350 K.

Trapping-mediated dissociative chemisorption of ethane on Ir(110)-(1×2)

Evidence is presented to support a trapping-mediated dissociative chemisorption mechanism for ethane interacting with an Ir(110)-(1×2) surface. The data were obtained from supersonic molecular-beam measurements with an incident kinetic energy Ei ranging between 1.2 and 24.1 kcal/mol, a surface temperature Ts between 154 and 500 K, and an incident angle θi between 0° and 45°. For Ei less than approximately 13 kcal/mol, the probability of dissociative chemisorption S0 decreases rapidly with increasing Ts. For a surface temperature of 154 K, S0 decreases with increasing Ei for 1.2≤Ei ≤13.4 kcal/mol, consistent with a trapping-mediated chemisorption mechanism. Indeed, the data also support quantitatively a kinetic model consistent with a trapping-mediated chemisorption mechanism. The difference in the activation energies for desorption and chemisorption from the physically adsorbed, trapped state Ed −Ec is 2.2±0.2 kcal/mol. In the trapping-mediated regime, S0 is found to be rather insensitive to incident angle, scaling with Ei cos0.5 θi .