Electronic structure calculations provide predictions of energy thresholds for a wide range of surface chemical reactions, and they are the basis for significant advances in our understanding of surface reactivity. Despite the central role these calculations and their predictions play, testing the absolute accuracy of these calculations with direct experimental measurements has proven to be challenging.

The presentation will focus on using the results of state-resolved beam-surface scattering measurements to benchmark theoretical predictions. We find that when these measurements are performed on a cold surface, we observe sharp energetic thresholds for reaction. These observations result from our ability to control and vary precisely all energetic degrees of the system. These experimentally measured threshold energies can be compared directly with both electronic structure calculations, as well as with quantum dynamics predictions of chemical reactivity. We will compare experimental results for methane dissociation on Ni(111) with recent computational predictions from the Jackson group to illustrate this approach. We will also provide an update on our work extending this approach to other molecule-surface systems.

We have previously reported observations of large (>10^-16 cm²), low-energy (<500 eV), electron-promoted desorption cross-sections for benzene (C₆H₆) molecules adsorbed on the surface of amorphous solid water [1]. We will now report on the extension of this work to other molecular solids exhibiting varying degrees of hydrogen bonding within the molecular solid itself and between the solid surface and adsorbed benzene; specifically we have repeated our measurements employing substrates comprised of solid methanol (CH₃OH) and diethyl ether (CH₃CH₂OCH₂CH₃). Our report will detail our studies of the structure of adsorbed layers of C₆H₆ on the molecular solids and demonstrate the crucial role of hydrogen bonding in propagating electronic excitation to the solid-vacuum interface where C₆H₆ desorption can occur. Competitive electron-promoted chemistry in the form of H₂ formation will also be reported. Conclusions related to the impact of these observations on the early phase of icy interstellar grain chemistry will be discussed.

[1] Highly efficient electron-stimulated desorption of benzene from amorphous solid water ice, J. D. Thrower, M. P. Collings, F. J. M. Rutten, and M. R. S. McCoustra, Chem. Phys. Lett., 2011, 505, 106–111.

We have constructed a potential energy surface (PES) for H atoms interacting with fcc gold based on the form of the PES in Effective Medium Theory. The PES was adjusted to match energies calculated by DFT in many configurations, including many with the Au atoms displaced from their lattice positions. It describes both the interatomic forces and electron densities in full dimension with the accuracy of the ab initio energies used in its construction. Calculations describing the motion of H and Au atoms using this full dimensional adiabatic PES agree with results obtained previously using Ab Initio Molecular Dynamics, demonstrating the accuracy of the PES for configurations occurring in the scattering of H atoms from a surface at finite temperature.

The analytic expression for the total energy contains the embedded electron density leading to a self-consistent approach to simulating nonadiabatic trajectories. We find that nonadiabatic electron-hole pair excitation is the most important energy loss pathway for the H atom. The calculated energy distributions for scattered H atoms are in reasonable agreement with experimental results that are just becoming available. and determines the probability and mechanism for its adsorption. Analysis of trajectories calculated with and without nonadiabatic energy dissipation shows the adsorption or sticking probability as well as the mechanism of H atom adsorption is changed dramatically by nonadiabatic energy transfer

It is well known that irradiation of the solid with electrons or photons can cause its decomposition. This process, or, more adequately processes, is very fast (typically finalized within a time shorter than 10^-14 s) and is realized mainly by desorption of atoms or ions.Unfortunately, until now, existing models still do not give a proper value of the desorption yield and, simultaneously, a correct kinetic-energy distribution of emitted particles, as compared to the experimental observations.

In this talk a classical, quasi-quantum and quantum description of the positive ion desorption from ionic crystal surface, in which three potentials are involved, will be discussed and compared. It will be shown that the quantum description allows to explain some effects observed experimentally, such as a periodicity of small oscillations on the kinetic energy distribution (KED) curves (predicted by Wave-Packet Squeezing model) and emission through a temporarily existing potential barrier from the temporary bounded states located above the vacuum level. Moreover, analysis method of the ion KED oscillation and the fitting procedure which allows to determined a final effective desorption potential will be presented.

For two examples discussed, Li⁺ desorption from LiF and Na⁺ desorption from NaCl, desorption from two desorption sites can be distinguished – dominating ion desorption channel from adatom sites (more than 95%) and marginal one from the sites in the first surface layer. For the second desorption channel neighboring negative ions, due to surface relaxation lying in the first surface layer slightly above positive one, can act as a two-dimensional array of rosette-like apertures. In consequence, positive ions after passing through them may form diffraction pattern.

Finally, it appears that when desorption process is described using three potentials both the ions desorption efficiency and their kinetic-energy distribution are in agreement with the experimental results.

Molecular hydrogen is physisorbed on flat metal surfaces via van der Waals interaction. By taking advantage of the fact that molecular hydrogen exists in nuclear-spin isomers of ortho and para species [1], we have shown the interaction potential on Ag(111) is anisotropic with a slight perpendicular preference [2]. The Pd(210) surface has a step-like structure consisting of alternately aligned (100) and (110) terraces, and it has been shown that H₂ is rather strongly adsorbed on H-covered Pd(210) with a significant contribution of orbital hybridization [3]. On the other hand, it has been suggested that molecularly adsorbed species could be important for hydrogen absorption into the interior of Pd surfaces [4]. In the present study, we have investigated the adsorption of H₂ and D₂ on the Pd(210) with temperature-programmed desorption (TPD) combined with resonance-enhanced multi-photon ionization (REMPI).

When Pd(210) was exposed to H₂ at 115 K, TPD revealed a desorption peak at 180 K (α-peak) originating from the absorbed state as well as a peak at 280-320 K (b-peak) due to chemisorbed H. From the uptake rate of the α-peak, the absorption probability of H on Pd(210) was estimated to be 3×10^-3. When the surface was exposed to either H₂ or D₂ at 45 K, on the other hand, an additional TPD peak was observed at about 70 K (γ-peak), which was attributed to molecular adsorption. While a small difference between H₂ and D₂ was observed for the b-peak, the γ-peak temperature of D₂ was found to be higher than that of H₂ by 9 K, which corresponds to the difference in the adsorption energy of about 20 meV. Assuming that this difference is due to the zero-point energy difference in the adsorption potential, the adsorption potential was analyzed in terms of the Morse potential. By applying REMPI-TPD, furthermore, the TPD spectra of ortho-H₂ in the rotational state of J=1 and para-H₂ in J=0 were state-selectively measured. The desorption temperature of ortho-H₂ was found to be higher than that of para-H₂ by about 4 K, which corresponds to a difference in the adsorption energy of about 10 meV. We discuss that this large energy difference between the ortho and para species originates from the potential anisotropy on the basis of the first-order perturbation.

[1] K. Fukutani, T. Sugimoto, Prog. Surf. Sci. 88, 279 (2013).

[2] T. Sugimoto, K. Fukutani, Phys. Rev. Lett. 112, 146101 (2014).

[3] P. K. Schmidt et al., Phys. Rev. Lett. 24 87, 096103 (2001).

[4] S. Ohno et al., J. Chem. Phys. 140, 134705 (2014).