Electron Kinetic Energies from Vibrationally ... - ACS Publications

Nov 23, 2011 - ... Kinetic Energies from Vibrationally Promoted Surface. Exoemission: Evidence for a Vibrational Autodetachment Mechanism. Jerry L. La...
0 downloads 0 Views 3MB Size
ARTICLE pubs.acs.org/JPCA

Electron Kinetic Energies from Vibrationally Promoted Surface Exoemission: Evidence for a Vibrational Autodetachment Mechanism Jerry L. LaRue,† Tim Sch€afer,†,^ Daniel Matsiev,†,# Luis Velarde,†,r N. Hendrik Nahler,‡ Daniel J. Auerbach,† and Alec M. Wodtke*,†,§,|| †

Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106-9510, United States Department of Chemistry, Durham University, Durham DH1 3LE, U.K. § Institut f€ur Physikalische Chemie, Georg-August-Universit€at G€ottingen, Tammannstrasse 6, 37077 G€ottingen, Germany Max-Planck-Institut F€ur Biophysikalische Chemie, Karl Friedrich-Bonhoeffer-Institut, Am Fassberg 11, 37077 G€ottingen, Germany

)



ABSTRACT: We report kinetic energy distributions of exoelectrons produced by collisions of highly vibrationally excited NO molecules with a low work function Cs dosed Au(111) surface. These measurements show that energy dissipation pathways involving nonadiabatic conversion of vibrational energy to electronic energy can result in electronic excitation of more than 3 eV, consistent with the available vibrational energy. We measured the dependence of the electron energy distributions on the translational and vibrational energy of the incident NO and find a clear positive correlation between final electron kinetic energy and initial vibrational excitation and a weak but observable inverse dependence of electron kinetic energy on initial translational energy. These observations are consistent with a vibrational autodetachment mechanism, where an electron is transferred to NO near its outer vibrational turning point and ejected near its inner vibrational turning point. Within the context of this model, we estimate the NO-to-surface distance for electron transfer.

1. INTRODUCTION A full understanding of surface chemistry requires an accurate theoretical description of elementary reaction steps including but not limited to adsorption and bond-dissociation. Each of these elementary steps can be strongly influenced by energy transfer between the solid and the adsorbate. For example, energy transfer from molecular translation to surface degrees of freedom is essential to adsorption,1,2 while energy transfer between molecular vibration and the solid is involved in the dynamics of bond breaking and making in surface reactions.35 Although great progress has been made in understanding simple gas-phase chemical reactions,6 accurately modeling surface reactions is still one of the most important challenges in physical chemistry. Vibrational energy transfer dynamics at surfaces exhibits properties that point out the limitations of present theoretical treatments, which are most commonly performed using the BornOppenheimer approximation.7 There is now copious evidence of the importance of electronic nonadiabaticity in molecular vibrational energy transfer at metal surfaces. That is, molecular vibration can excite the solid’s electrons (and holes) and vice versa.8 “Molecular dynamics with electronic friction” is now one of the most popular approaches to deal with this issue.9 Here, electronic nonadiabaticity is treated as a weak influence on the adiabatic motion of the adsorbate. Despite the weak coupling assumption, the influence of electronic friction can be large. Vibrational lifetimes of CO(v = 1) on copper were about 109 shorter than those obtained in electronically adiabatic treatments, in good agreement with experiment.10,11 r 2011 American Chemical Society

Dynamics of chemical bond breaking might exhibit similarities with vibrational energy transfer; however, the amplitude of the interatomic motion during dissociation is much larger. Hence, it is unclear if friction theories apply to bond dissociation. Moreover, it is clear that electronic friction theories do not describe vibrational energy transfer in the strong coupling case.12 A good experimental approach to test this is surface scattering of highly vibrationally excited molecules, which probe the nature of large amplitude vibrational coupling to solids.1319 One remarkable observation within this context is multiquantum vibrational relaxation, first observed in scattering of NO(v = 15) from Au(111).17 Here, only a few percent of the scattered molecules are retained in their initial vibrational state, while the most probable scattering channel is loss of ∼78 vibrational quanta. Multiquantum vibrational relaxation can arise within an electronic friction framework,20 but vibrational relaxation is in this case sequential, involving multiple single-quantum energy transfer steps, where each transfer process occurs with a different electronhole pair. Evidence of the efficient transfer of many vibrational quanta to a single electron came from the observation of vibrationally promoted electron emission from a low work function surface.1315,19 Here, NO(v) was optically prepared (4 < v < 18), and the vibrationally promoted electron emission efficiency was recorded as a function of the vibrational state. Received: June 22, 2011 Revised: October 18, 2011 Published: November 23, 2011 14306

dx.doi.org/10.1021/jp205868g | J. Phys. Chem. A 2011, 115, 14306–14314

The Journal of Physical Chemistry A A vibrational energy threshold for vibrationally promoted electron emission was observed slightly above the surface work function.21 An inverse velocity dependence was also reported, showing that it is vibration and not translation that promotes the electron emission.18 A vibrational autodetachment mechanism was proposed involving electron transfer from the surface to the NO molecule and then to the vacuum. Here, an electron hops from the surface to the molecule as the affinity level of the molecule crosses the Fermi level of the surface near the outer turning point of vibration. The molecule then ejects the electron near the inner turning point of vibration, transferring vibrational energy to the electron in the process. The vibrational autodetachment mechanism represents a mechanistic foundation for the independent electron surface hopping (IESH) model.22 Here, potential energy surfaces (PESs) for both the neutral ground state and the negative ion excited state are calculated using density functional theory (DFT).22 These PESs are then used within a NewnsAnderson Hamiltonian23 to simulate the nonadiabatic electron hole pair (EHP) excitation resulting from the molecule surface encounter. The IESH model gives semiquantitative agreement with multiquantum vibrational relaxation experiments for NO(v = 15) on Au(111) and interestingly predicts a remarkable NO orientation dependence.24 There are, so far, only a few reports of theories designed to describe the EHP excitation distribution from surface chemical events.25 In this paper we present experimental observations of the electron kinetic energy distribution produced by scattering vibrationally prepared molecules of NO(v = 11 ( 2, 16 and 22) at two incidence energies of translation from a low work function (cesiated Au) surface, which follows up on a report of preliminary results.26 For comparison we also obtained results for NO(v = 0). This paper provides the first data of its kind on how changing the vibrational and translational energy of a target molecule influences the electron energy distribution in exoemission. We report a strong positive correlation between incidence energy of vibration and outgoing electron kinetic energy. In contrast, this work shows a weak inverse correlation between incidence energy of translation and outgoing electron kinetic energy. We show that these trends are consistent with energy limits set by the vibrational autodetachment mechanism. Within the context of this mechanism we may use the measured electron kinetic energy distributions to estimate the distance (∼10 Å) at which the electron is transferred between the surface and the NO molecule.

2. EXPERIMENTAL SECTION The experiments are performed in the moleculesurface scattering apparatus, which has been described only briefly in a prior publication.26 The apparatus (Figure 1) consists of three differentially pumped chambers: a molecular beam source chamber (region I, 107 Torr base pressure), an NO preparation and detection chamber (region II, 108 Torr base pressure), and an ultrahigh vacuum (UHV) surface science chamber (region III, 1010 Torr base pressure). A supersonic molecular beam of NO molecules originates in region I, the molecules are vibrationally excited in region II, and region III houses the surface science equipment where the low work function surfaces are prepared and the scattering experiments are carried out. During operation of the molecular beam, the pressures in regions IIII are 1  104 Torr, 8  107 Torr, and 5  1010 Torr, respectively. A pulsed piezoelectric valve (PV, 1 mm diameter nozzle, 10 Hz repetition rate, 3 atm stagnation pressure) is used to produce a supersonic molecular beam of