Library

Notes: We have determined the angular distributions of H2, HD, and D2 desorbing from Cu(111) for surface temperatures in the range 370–800 K. These are found to be strongly peaked and symmetric about the surface normal in every case. Results for all three isotopes are found to be indistinguishable, being close to a cos 12θf distribution at 600 K, slightly narrower at 370 K, and slightly broader at 800 K. Results are discussed in terms of other previous desorption measurements and related to adsorption data via detailed balance.

Notes: We examine experimental data on the activated dissociative adsorption and associative recombination of hydrogen at copper surfaces with respect to the role played by molecular vibrational states. We use models describing the variation of the adsorption probability with the vibrational state, kinetic energy, and angle of the incident molecules, establishing the parameters of the models by a nonlinear least squares fit to adsorption data. Using the principle of detailed balance, we apply these models to the corresponding data on associative desorption thus comparing adsorption data with desorption angular, velocity, and internal state distributions. The most consistent picture resulting from this analysis is that the adsorption has significant contributions from both H2(v=0) and H2(v=1) and that these components have markedly different translational thresholds. Within the framework of this picture we are able to resolve the apparent contradiction between the strong angular dependence of the kinetic energy required for adsorption and the lack of angular dependence of the mean kinetic energy of desorption. We also partially resolve the apparent discrepancy in interpretation of the role of H2(v=1) in recent adsorption experiments.

Notes: We have determined the dependence of the dissociative adsorption probability in the zero coverage limit, S0, for H2 on Cu(111) as a function of translational energy, Ei, and incidence angle, θi, vibrational state, v, and rotational state, J. We have also obtained information on the effect of surface temperature, Ts, on this probability. These results have been obtained by combining the findings of two separate experiments. We have obtained the form of the dependence of S0 on Ei at Ts=925 K for a range of quantum states from desorption experiments via the principle of detailed balance. We have obtained absolute S0 values from direct molecular beam adsorption experiments, which reveal that S0 scales with the so-called "normal energy,'' En=Ei cos2 θi. The desorption experiments provide detailed information for J=0 to 10 of H2(v=0) and for J=0 to 7 of H2(v=1). The beam experiments additionally provide information on the adsorption of H2(v=2), averaged over J. All measurements are consistent with adsorption functions with an s-shaped form, which can be described by S0=A(1+erf(x))/2, where x=(En−E0)/W. Values of W are ∼0.16 and 0.13 eV for v=0 and v=1, respectively, at Ts=925 K, falling by about 0.05 eV for Ts=120 K, and with only a slight dependence on J. Values of A are insensitive to v and J, with a value of ∼0.25. S(En,v,J) curves are thus similar for different v and J, but shifted in En.In contrast, we find that the values of E0, which determine the mid-point of the curves, have a strong dependence on v and J. Specifically, E0 for H2(v=0) molecules is about 0.6 eV, falling to 0.3 and 0.1 eV for H2(v=1) and H2(v=2), respectively. Translational energy is thus about twice as effective as vibrational energy in promoting dissociation. E0 rises with increasing J at low J, before falling at high J, indicating that rotational motion hinders adsorption for low rotational states (J〈4), and enhances adsorption for high rotational states (J(approximately-greater-than)4). Results are compared with similar studies on the D2/Cu(111) system and with recent calculations. Finally, these results are used to predict the dependence of the rate of dissociation on temperature for a "bulb'' experiment with ambient hydrogen gas in contact with a Cu(111) surface. This simulation yields an activation energy of 0.47 eV for temperatures close to 800 K, compared to a literature value of 0.4 eV from experiment. Analysis of the temperature dependence reveals that the dominant reason for the increase in rate at high temperature is the increase in population of the high energy tail of the translational energy distribution. © 1995 American Institute of Physics.

Notes: We have determined the alignment of D2(v,J) desorbed from Cu(111). The measurements reveal a small preference for "helicoptering'' motion that increases with increasing J. At low J, the alignments are much smaller than predicted by recent calculations. We believe that the anisotropic potential may scramble the alignment as the molecules leave the surface. © 1996 American Institute of Physics.

Notes: We have investigated the dependence on the rotational and vibrational states of the translational energy of D2(v,J) formed in recombinative desorption from Cu(111). These results provide information about the effect of rotational energy relative to that of vibrational and translational energy on the dissociative chemisorption of D2 on Cu(111). The range of rovibrational states measured includes rotational states J=0–14 for vibrational state v=0, J=0–12 for v=1, and J=0–8 for v=2. D2 molecules were detected in a quantum-state-specific manner using three-photon resonance-enhanced multiphoton ionization (2+1 REMPI). Kinetic energies of desorbed molecules were obtained by measuring the flight time of D2+ ions in a field-free region. The mean kinetic energies determined from these measurements depend strongly on the rotational and vibrational states. Analyzing these results using the principle of detailed balance confirms previous observations that vibrational energy is effective, though not as effective as translational energy, in promoting adsorption. Rotational motion is found to hinder adsorption for low rotational states (J≤5) and enhance adsorption for high rotational states (J≥5). Even for high J states, however, rotational energy is less effective than either vibrational energy, which is 30%–70% more effective than rotational energy, or translational energy, which is 2.5–3 times more effective than rotational energy in promoting adsorption. The measured internal state distributions for the rovibrational states listed above are consistent with the observed dependence of the kinetic energy of the de- sorbed molecules with the rotational state. In addition, the analysis performed yields the dependence of the adsorption probability on kinetic energy separately for each rovibrational state. These functions have very similar sigmoidal shapes for all states examined. Changing the quantum state is primarily associated with a shift in the position, or threshold energy, for the curves. The level at which these functions saturate or level off at high energy is independent of rotational state but varies nonmonotonically with the vibrational state.