Vibrational Inelasticity of Highly Vibrationally Excited NO on Ag(111

Letter pubs.acs.org/JPCL

Vibrational Inelasticity of Highly Vibrationally Excited NO on Ag(111) Bastian C. Krüger,† Sven Meyer,† Alexander Kandratsenka,†,‡ Alec M. Wodtke,†,‡,§ and Tim Schaf̈ er*,† †

Institute of Physical Chemistry, Georg-August University of Göttingen, Tammannstraße 6, 37077 Göttingen, Germany Department of Dynamics at Surfaces, Max Planck Institute for Biophysical Chemistry, Am Faßberg 11, 37077 Göttingen, Germany § International Center for Advanced Studies of Energy Conversion, University of Göttingen, 37077 Göttingen, Germany ‡

ABSTRACT: Multiquantum relaxation of highly vibrationally excited nitric oxide on noble metals has become one of the best studied examples of the Born−Oppenheimer approximation’s failure to describe molecular interactions at metal surfaces. When first reported, relaxation of highly vibrationally excited NO occurring in collisions with Au(111) surfaces exhibited the largest vibrational inelasticity seen in molecule−surface collisions, and no system has been found to date exhibiting a greater vibrational inelasticity. In this work, we compare the relaxation of NO(v = 11) in scattering events on Ag(111) to that on Au(111). The relaxation probability and the average vibrational energy loss are much higher when scattering from Ag(111). We discuss possible reasons for this remarkable phenomenon, which may be related to the dissociation of NO, possible on Ag(111) at lower energy compared with Au(111).

I

nteractions between molecules and metal surfaces are often electronically nonadiabatic, where the molecule’s atomic motion couples to the electron bath of the solid.1−7 One of the best studied examples is the scattering of NO from Au(111), where NO vibration strongly couples to the electrons of the metal.1,2 Studies of this system include molecular beam surface scattering-based investigations of vibrational excitation and deexcitation varying surface temperature, incidence energy of translation, initial NO vibrational excitation, and molecular orientation.8−15 A striking example of electronically nonadiabatic behavior is seen when scattering highly vibrationally excited NO(vi = 15) from Au(111). Here, multiquantum relaxation occurs on a subpicosecond time scale, producing characteristic final vibrational state distributions peaking near v = 7, having transferred, on average, ∼1 eV of vibrational energy to the electrons of the metal.10,15 In contrast, almost no vibrational relaxation occurs when scattering vibrationally excited NO from insulating surfaces like LiF, as the large band gap in insulators prohibits the coupling to electron hole pairs (EHPs).15,16 Observations of NO multiquantum vibrational relaxation on Au(111) have become the subject of first-principles theoretical study as this represents perhaps the best experimental benchmark for Born−Oppenheimer failure in the description of molecular interactions at metal surfaces. One important approach to modeling and understanding NO/Au(111) scattering dynamics employs independent electron surface hopping (IESH),17 a method which has successfully reproduced several experimental outcomes.13,18 It accurately reproduces vibrational excitation and de-excitation of NO(vi = 0−3) in collisions with Au(111) when incidence translational energies are high.13 At lower incidence energies, deviations between experiment and theory are found and attributed to errors in the electronically adiabatic potential energy surface (PES).14 © XXXX American Chemical Society

Furthermore, deviations from experiment also increase at higher values of vi. These deviations might also be attributed to errors in the PES because the PES used in those calculations do not include a pathway to NO dissociation, which becomes important only at high energy.19 It is important to extend our experimental observations of NO vibrational energy exchange to other metal surfaces to help gain a deeper understanding into these issues. This is one way to experimentally alter the electronically adiabatic PES and eventually to better understand how the PES influences electronically nonadiabatic behavior. Moreover, it offers the opportunity for possible new insights into the role of electronically nonadiabatic interactions in chemical reactions taking place at metal surfaces. Of course, other physical properties that could be important to the interactions are also changed; these include: surface work function, coupling strength between molecule and surface, surface corrugation, surface states, and density of states (DOS) at the Fermi level. We present experimental measurements that lead to relaxed vibrational state distributions for NO(vi = 11) scattered from Ag(111). Ag(111) was chosen because the properties just mentioned are all similar to those of Au(111). Furthermore, its similarity to Au(111) means theoretical treatments developed for NO/Au should also be possible for NO/Ag. More importantly, the NO/Ag electronically adiabatic PES is substantially different than that of NO/Au; the attractive physisorption well depth33 is deeper and the barrier to dissociation of NO is lower.21 Experimental evidence of NO Received: November 3, 2015 Accepted: January 13, 2016

441

DOI: 10.1021/acs.jpclett.5b02448 J. Phys. Chem. Lett. 2016, 7, 441−446

Letter

The Journal of Physical Chemistry Letters

Figure 1. REMPI spectra of NO(vi = 11) scattered from Au(111) at 300 K (upper panel) and from Ag(111) at 670 K (lower panel). The incidence translational energy is in both cases 0.51 eV. The triangles mark the band heads of the gamma-band. The numbers in the parentheses (v, v″) indicate the vibrational quantum number in the electronically excited A2Σ state (v′) and the electronic ground state X2Π state (v″). Vibrational bands starting at high vibrational states in the electronic ground state are missing when scattering NO(vi = 11) from Ag(111). The enlargement of a small section of the spectrum in the inset shows the agreement between experiment and simulation.

dissociation on Ag(111) has also been reported,22 which is not the case for Au(111). Despite the many similarities of physical properties comparing Ag to Au, the measurements reported here show that the relaxed vibrational distributions of NO(vi = 11) are strongly dependent on the choice of noble metal. The magnitude of vibrational relaxation on Ag(111) is much larger than for the case of Au(111). In the remainder of this paper we explain our methods and observations and discuss possible reasons for this behavior. The experiments have been performed on a molecular beam surface scattering apparatus described in detail in the Experimental Methods. Highly vibrationally excited NO(vi = 11) molecules are produced by laser methods using the pump−dump−sweep technique.23 We vary the incidence translational energy of the NO molecules by expanding different mixtures of NO in H2 in a pulsed supersonic expansion. Scattered NO molecules are detected using resonance-enhanced multiphoton ionization (REMPI) via the NO γ-bands. Figure 1 shows (1 + 1) REMPI spectra obtained when NO(vi = 11) is scattered from Au(111) (upper panel) and Ag(111) (lower panel). Because of the open-shell character of the NO molecule the spectrum comprises 12 rotational branches, leading to a large number of ro-vibrational lines. For the case of Au(111) the many vibrational states produced in the relaxation process lead to a heavily congested spectrum, which can nevertheless be fully resolved and unambiguously assigned. To help clarify the assignment we have marked vibrational band heads by black triangles with labels γ(v′, v″); here, v′ refers to the vibrational quantum number of the A-state

used in the REMPI and v″ refers to the relaxed vibrational state of the NO molecule in the electronic ground state produced by scattering NO(vi = 11) with the noble-metal surface. NO vibrational states between v″ = 11 and 2 are seen in these spectra. The observed REMPI spectrum obtained from NO(vi = 11) scattering from Ag(111) is much simpler. Even to the untrained eye, it is clear that the observed behavior is very different. This is due to the fact that far fewer vibrational states are produced with significant populations than in the case of Au(111). It is obvious from the REMPI spectra that high vibrational states are only populated when scattering NO from the Au(111) surface. We subject these REMPI spectra to quantitative analysis, as has been described in a previous paper.10 In brief, we calculate term energies using Brown’s Hamiltonian24 and spectroscopic constants from refs 25 and 26. We then fit the calculated spectrum to the experimentally measured spectrum by varying the population of each vibrational state. An example of this analysis is shown as an inset to Figure 1. The resulting final vibrational distributions for NO(vi = 11, Ji = 0.5) scattered from Au(111) and Ag(111) are depicted in Figure 2a−c. We focus our attention on two important features easily observable when comparing the panels in this Figure. First, the final vibrational distribution is drastically shifted toward lower vibrational energies in the case of NO(vi = 11) scattering from Ag(111) at 670 K (two lower panels in Figure 2) in comparison with Au(111) at 300 K (upper panel). For scattering from Au(111), NO(vi = 11) loses on average 5.1 vibrational quanta corresponding to a fraction of 0.4 of the initial vibrational excitation. For scattering from Ag(111), NO(vi = 11) loses an average of 8.1 vibrational quanta, a 442

DOI: 10.1021/acs.jpclett.5b02448 J. Phys. Chem. Lett. 2016, 7, 441−446

Letter

The Journal of Physical Chemistry Letters

Figure 2. Relaxed vibrational state distributions of NO(vi = 11) scattered from (a) Au(111) with Einc = 0.51 eV and Tsurf = 300 K and from Ag(111) at Tsurf = 670 K with (b) Einc = 0.14 eV and (c) Einc = 0.51 eV. The vibrational state distribution is drastically shifted toward lower vibrational states in the case of Ag(111). Experiments on Ag(111) have been performed on a heated surface to avoid surface contamination; however, we do not expect a significant influence of the temperature on the relaxation probability.

Figure 3. Time-of-flight (TOF) spectra of NO scattered from the Ag(111) surface in vibrational state v″ = 3. Incident NO molecules are prepared in the state vi = 11, Ji = 0.5 2.5 mm from the Ag(111) surface. REMPI is performed 19.5 mm from the surface. The Figure shows the REMPI signal as a function of the delay between the laser pulses used for preparing NO(vi = 11) and probing scattered NO(v″ = 3, J = 11.5). The solid lines are the result of a fitting procedure described in detail in ref 27. The observed distribution corresponds to the average final translational energy of 0.51 eV. Inset A: Angular distribution for the NO(vi = 11, Ji = 0.5) → NO(vs = 3, Js = 11.5) scattering channel. The slight asymmetry of the distribution is caused by the tilted orientation of the sample in the sample holder. The narrow angular distribution indicates a direct scattering process. Inset B: From the fit to TOF data the final translational energy distribution is derived for three vibrational states.

fraction of 0.7 of the initial vibrational excitation. We point out that 0.7 is a lower limit to the fraction of vibrational energy transferred as we have made the unlikely assumption that no population is found in vs = 0 or 1, states that are obscured by thermal background and thus impossible to quantify in the present experiment. Second, we observe no effect of the incidence translational energy on the final vibrational distribution when scattering from Ag(111) (cf. two lower panels in Figure 2). This is in contrast with the relaxation of highly vibrationally excited NO at a Au(111) surface, at which the relaxation probability increases with incidence translational energy.10 Additional important insight into the scattering mechanism can be gained from the experimentally derived translational energy of scattered NO molecules. Figure 3 shows data from a time-of-flight (TOF) experiment for NO molecules leaving the surface in vs = 3 and Js = 11.5, which corresponds to a loss of 8 vibrational quanta (1.7 eV). For this particular state, the average final translational energy is nearly the same as the incidence translational energy (0.51 eV), which is significantly larger than the crystal’s thermal energy. This result immediately rules out a trapping/desorption-mediated vibrational energy-transfer mechanism and is consistent with the narrow angular distribution shown in the inset of Figure 3. Beyond this, it might seem surprising that outgoing NO translational energy is so close to the incidence translational energy. In the past, a Baule model has accurately described the amount of energy transferred from NO translation to Au.28 For NO collisions with a Ag surface, the Baule model would predict 0.16 eV outgoing translational energy for Ei = 0.51 eV. Comparing this to the observed 0.51 eV, we conclude that ∼0.35 eV excess translational energy has appeared in the NO(vs = 3) molecule due to vibration-to-translation coupling. Figure 3 (inset B) shows that the outgoing translational energy of NO(vs = 2 and 4) is shifted to higher and lower energies, respectively, compared with NO(vs = 3). On the basis of the analysis of TOF experiments for relaxation to these three different final vibrational states at the Ag(111) surface, we find that 0.17 ±

0.1 ΔEvib appears as outgoing NO translation. By comparison, on Au(111) at incidence translational energy of 0.57 eV we found that the outgoing excess translational energy beyond the Baule expectation was 0.11 ± 0.03 ΔEvib. We note that the efficiency of V−T coupling increases slightly with incidence translational energy, similar to reports on NO(vi = 3)/ Au(111).29 Multiquantum vibrational relaxation results from an electrontransfer event producing a transient NO−.13 For this process, the energy difference between the molecule’s affinity level and the Fermi energy is important. Hence, the work function of the noble metal, 4.7 eV for Ag(111) and 5.3 eV for Au(111), is expected to exert a significant influence because vibrational relaxation can only take place if the electron transfer is energetically allowed. We can get a qualitative idea of the work function effect on the vibrational state distribution using the IESH model. We implemented the model for NO/Au(111), as it has been described elsewhere.17 For NO/Ag(111) scattering simulations, we used the same parametrization of the PES as in the case of gold;17 only the work function has been lowered to 4.7 eV and the mass of the metal atoms has been reduced to that of Ag. (Assuming the metal atoms have the mass of Au does not give meaningfully different results.) Note that the IESH model reproduces NO vibrational relaxation at high incidence translational energies poorly.19 Nevertheless, calculations like this help to estimate the influence of the work function on the relaxation process. We find that the IESH model predicts a higher relaxation probability as well as a higher average vibrational energy loss for collisions of NO with Ag(111) compared with collisions with a Au(111); however, this effect is much smaller than seen in Figure 2. 443

DOI: 10.1021/acs.jpclett.5b02448 J. Phys. Chem. Lett. 2016, 7, 441−446

Letter

The Journal of Physical Chemistry Letters

electronic states, cannot be ruled out, it appears to us that features of the interaction potential related to the dissociation of NO (neglected in current models of electronically nonadiabatic vibrational energy transfer) could explain our observations.

Another previously reported experiment is relevant to the influence of work function on electronically nonadiabatic vibrational relaxation; these are experimental studies of NO(vi ≫ 0) vibrational relaxation at a Cs-covered gold surface, the work function of which is 1.6 eV.30 In that work, electron excitation resulting from NO vibrational relaxation is enough to produce emitted electrons whose energy distribution can be measured.31,32 The electron energy distributions showed that 0.8 ± 0.1 of the incidence vibrational energy can be converted to electronic excitation for NO relaxation on Cs covered gold. For NO relaxation on Ag, we see a similar fraction of vibrational energy converted to electron excitation, despite the fact that Ag’s work function is 4.7 eV. This appears to rule out the argument that the lower work function of Ag with respect to Au can explain the large differences in vibrational energy loss seen in Figure 2. We postulate that differences between the two adiabatic PES’s are responsible for the large differences in vibrational energy loss seen in Figure 2. There is some evidence of temperature-programmed desorption that NO binds more strongly to Ag(111)33 than to Au(111).34 Stronger binding also implies closer approach of NO to the Ag surface, accessing configurations where electron transfer is more likely. Evidence of NO dissociation on Ag(111) has also been reported.22 NO dissociation on Au has not been seen. Theory suggests that the barrier to NO dissociation on Ag(111) (3.1 eV) is lower than that on Au(111) (3.6 eV).21 Because a stretched geometry resulting from trajectories that pass near the transition state for dissociation increases the anionic character of the NO molecule,18 it is reasonable to postulate that nonadiabatic effects could be enhanced for the NO molecule on Ag near the dissociation transition state. Furthermore, enhanced vibrational inelasticity due to the influence of a dissociation transition state has been clearly demonstrated for H2 collisions on Cu(111).35 A similar enhancement of vibrational inelasticity may be at work in the NO/Ag system. NO(vi = 11) studied in this work has a vibrational energy of 2.5 eV, just shy of the dissociation barrier on Ag. For NO/Au(111), we observe large and growing deviations from theory with increasing NO vibrational excitation, where the highest vibrational energy was 3.1 eV. See figure 1 in ref 19. The potential energy surface used in the IESH model for NO/Au(111) does not allow NO dissociation; future work is needed to correct this. It appears to us most likely that the vibrational relaxation probability is enhanced for NO molecules with vibrational energy close to the dissociation energy. Hence, the reduced energy of the dissociation barrier with respect the NO vibrational energy, as is the case when comparing NO(vi = 11) on Ag(111) to Au(111) in this work, is likely to enhance vibrational relaxation. While we favor this explanation, it is not the only reasonable possibility. Up to now, all theoretical approaches on NO multiquantum vibrational relaxation at metal surfaces consider only the ground neutral electronic state and the anionic state. Thus, the influence of other electronically excited states is unknown and potentially important. Future studies to improve our theoretical models are clearly needed. In conclusion, we have shown that vibrational relaxation of NO(vi = 11) is dramatically enhanced when scattering from Ag(111) in comparison with scattering from Au(111). The reduced work function of Ag cannot explain the significant enhancing effect on the vibrational relaxation. Although other influences, for instance, electronically excited states or surface

■

EXPERIMENTAL METHODS The experiments are performed in a differentially pumped molecular beam surface-scattering apparatus that has been described in detail elsewhere.8−10,19 We expand mixtures of 10 and 60% NO in H2 in a pulsed piezoelectric valve to produce a molecular beam of 0.51 and 0.14 eV, respectively. NO molecules are excited to v = 11 employing the pump− dump−sweep technique via the A2Σ+(v = 2) state 2 cm in front of the surface.23 For the pump and the dump step we use the 204.7 and 336.1 nm output of two home-built narrow bandwidth OPO systems, which are pumped by the second harmonic of the same Nd:YAG laser (Lab 170-10, Spectra Physics).36 Residual molecules in the A2Σ+(v = 2) state are removed in the sweep step by excitation to a predissociative state using the 450.9 nm output of a Nd:YAG laser (Lab 23010, Spectra Physics) pumped dye laser (Precision scan, Sirah). Scattered molecules are detected by (1 + 1) resonanceenhanced multiphoton ionization (REMPI) via A2Σ+(v′ = 0−7) using the 245−315 nm radiation of a commercial OPO system (Sunlite, Continuum). Ions are detected on a microchannel plate detector (MCP, tectra, Chevron configuration), and the signal is visualized and recorded by an oscilloscope (LeCroy, Waverunner LT344) interfaced to an electrical computer. The Ag(111) surface is attached to a home-built sample holder and can be temperature-controlled between 90 K and the melting point of the crystal. The Ag(111) is sputtered with Ar+ ions (LK Technologies, NGI 3000, 3 kV, 20 min) and subsequently annealed to 970 K for 20 min. Cleanliness is verified using Auger electron spectroscopy (ESA-150, Staib Instruments). We perform the molecular beam surface scattering experiments at a surface temperature of 670 K. At this temperature we maintain reproducible experimental conditions and do not observe any adsorbate build-up at the surface using Auger electron spectroscopy. Note that the NO vibrational relaxation experiments on Au(111) shown in Figure 1 have been performed at 300 K; however, we compare the NO vibrational relaxation on Au(111) and Ag(111) directly, as vibrational relaxation does not depend on thermally excited EHPs. Hence, in contrast with vibrational excitation experiments we do not expect any surface temperature dependence. The obtained REMPI spectra are analyzed in a two-step procedure: (i) the spectrum is cut into smaller parts and vibrational bands are analyzed to obtain the rotational state distributions for all populated vibrational states; (ii) the whole spectrum is then fitted by a model containing population factors for each vibrational band and rotational state distributions are fixed to the form determined in step (i).

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. 444

DOI: 10.1021/acs.jpclett.5b02448 J. Phys. Chem. Lett. 2016, 7, 441−446

Letter

The Journal of Physical Chemistry Letters

■

(17) Shenvi, N.; Roy, S.; Tully, J. C. Nonadiabatic Dynamics at Metal Surfaces: Independent-electron Surface Hopping. J. Chem. Phys. 2009, 130, 174107. (18) Shenvi, N.; Roy, S.; Tully, J. C. Dynamical Steering and Electronic Excitation in NO Scattering from a Gold Surface. Science 2009, 326, 829−832. (19) Krüger, B. C.; Bartels, N.; Bartels, C.; Kandratsenka, A.; Tully, J. C.; Wodtke, A. M.; Schäfer, T. NO Vibrational Energy Transfer on a Metal Surface: Still a Challenge to First-Principles Theory. J. Phys. Chem. C 2015, 119, 3268−3272. (20) Barker, J. A.; Kleyn, A. W.; Auerbach, D. J. Rotational Energy Distributions in Molecule Surface Scattering: Model Calculations for NO/Ag(111). Chem. Phys. Lett. 1983, 97, 9−13. (21) Falsig, H.; Shen, J.; Khan, T.; Guo, W.; Jones, G.; Dahl, S.; Bligaard, T. On the Structure Sensitivity of Direct NO Decomposition over Low-Index Transition Metal Facets. Top. Catal. 2014, 57, 80−88. (22) So, S. K.; Franchy, R.; Ho, W. The Adsorption and Reactions of NO on Ag(111) at 80 K. J. Chem. Phys. 1989, 91, 5701−5706. (23) Bartels, N.; Krüger, B. C.; Meyer, S.; Wodtke, A. M.; Schäfer, T. Suppression of Spontaneous Emission in the Optical Pumping of Molecules: Pump−Dump−Sweep−Probe. J. Phys. Chem. Lett. 2013, 4, 2367−2370. (24) Brown, J. M.; Colbourn, E. A.; Watson, J. K. G.; Wayne, F. D. An Effective Hamiltonian For Diatomic Molecules: Ab Initio Calculations of Parameters of HCl+. J. Mol. Spectrosc. 1979, 74, 294−318. (25) Amiot, C. The infrared emission spectrum of NO: Analysis of the Δv = 3 sequence up to v = 22. J. Mol. Spectrosc. 1982, 94, 150−172. (26) Danielak, J.; Domin, U.; Ke, R.; Rytel, M.; Zachwieja, M. Reinvestigation of the Emission γ Band System (A2Σ+−X2Π) of the NO Molecule. J. Mol. Spectrosc. 1997, 181, 394−402. (27) Rahinov, I.; Cooper, R.; Yuan, C.; Yang, X.; Auerbach, D. J.; Wodtke, A. M. Efficient Vibrational and Translational Excitations of a Solid Metal Surface: State-to-state Time-of-flight Measurements of HCl(v = 2,J = 1) Scattering from Au(111). J. Chem. Phys. 2008, 129, 214708. (28) Golibrzuch, K.; Shirhatti, P. R.; Rahinov, I.; Auerbach, D. J.; Wodtke, A. M.; Bartels, C. Incidence energy dependent state-to-state time-of-flight measurements of NO(v = 3) collisions with Au(111): the fate of incidence vibrational and translational energy. Phys. Chem. Chem. Phys. 2014, 16, 7602−7610. (29) Golibrzuch, K.; Shirhatti, P. R.; Altschäffel, J.; Rahinov, I.; Auerbach, D. J.; Wodtke, A. M.; Bartels, C. State-to-State Time-ofFlight Measurements of NO Scattering from Au(111): Direct Observation of Translation-to-Vibration Coupling in Electronically Nonadiabatic Energy Transfer. J. Phys. Chem. A 2013, 117, 8750− 8760. (30) LaRue, J. L.; White, J. D.; Nahler, N. H.; Liu, Z.; Sun, Y.; Pianetta, P. A.; Auerbach, D. J.; Wodtke, A. M. The work function of submonolayer cesium-covered gold: A photoelectron spectroscopy study. J. Chem. Phys. 2008, 129, 024709. (31) LaRue, J.; Schäfer, T.; Matsiev, D.; Velarde, L.; Nahler, N. H.; Auerbach, D. J.; Wodtke, A. M. Vibrationally promoted electron emission at a metal surface: electron kinetic energy distributions. Phys. Chem. Chem. Phys. 2011, 13, 97−99. (32) LaRue, J. L.; Schäfer, T.; Matsiev, D.; Velarde, L.; Nahler, N. H.; Auerbach, D. J.; Wodtke, A. M. Electron Kinetic Energies from Vibrationally Promoted Surface Exoemission: Evidence for a Vibrational Autodetachment Mechanism. J. Phys. Chem. A 2011, 115, 14306−14314. (33) Behm, R. J.; Brundle, C. R. Summary Abstract: Decomposition of NO on Ag(111) at low temperatures. J. Vac. Sci. Technol. A 1984, 2, 1040−1041. (34) Engelhart, D. P.; Wagner, R. J. V.; Meling, A.; Wodtke, A. M.; Schäfer, T. Temperature Programmed Desorption of Weakly Bound Adsorbates on Au(111). Surf. Sci. 2015, DOI: 10.1016/ j.susc.2015.06.010.

ACKNOWLEDGMENTS We acknowledge support from the Alexander von Humboldt Foundation and the Deutsche Forschungsgemeinschaft under CRC 1073.

■

REFERENCES

(1) Bartels, C.; Cooper, R.; Auerbach, D. J.; Wodtke, A. M. Energy Transfer at Metal Surfaces: The Need to Go Beyond the Electronic Friction Picture. Chem. Sci. 2011, 2, 1647−1655. (2) Rahinov, I.; Cooper, R.; Matsiev, D.; Bartels, C.; Auerbach, D. J.; Wodtke, A. M. Quantifying the Breakdown of the Born-Oppenheimer Approximation in Surface Chemistry. Phys. Chem. Chem. Phys. 2011, 13, 12680−12692. (3) Hotzel, A.; Ishioka, K.; Knoesel, E.; Wolf, M.; Ertl, G. Can We Control Lifetimes of Electronic States at Surfaces by Adsorbate Resonances? Chem. Phys. Lett. 1998, 285, 271−277. (4) Harada, Y.; Masuda, S.; Ozaki, H. Electron Spectroscopy Using Metastable Atoms as Probes for Solid Surfaces. Chem. Rev. 1997, 97, 1897−1952. (5) Wodtke, A. M.; Matsiev, D.; Auerbach, D. J. Energy Transfer and Chemical Dynamics at Solid Surfaces: The Special Role of Charge Transfer. Prog. Surf. Sci. 2008, 83, 167−214. (6) Auerbach, D.; Wodtke, A. Vibrational Energy Transfer at Surfaces: The Importance of Non-Adiabatic Electronic Effects. In Dynamics of Gas-Surface Interactions, Díez Muiño, R., Busnengo, H. F., Eds.; Springer: Berlin, 2013; Vol. 50, pp 267−297. (7) Golibrzuch, K.; Bartels, N.; Auerbach, D. J.; Wodtke, A. M. The Dynamics of Molecular Interactions and Chemical Reactions at Metal Surfaces: Testing the Foundations of Theory. Annu. Rev. Phys. Chem. 2015, 66, 399−425. (8) Bartels, N.; Golibrzuch, K.; Bartels, C.; Chen, L.; Auerbach, D. J.; Wodtke, A. M.; Schäfer, T. Observation of Orientation-dependent Electron Transfer in Molecule−surface Collisions. Proc. Natl. Acad. Sci. U. S. A. 2013, 110, 17738−17743. (9) Bartels, N.; Golibrzuch, K.; Bartels, C.; Chen, L.; Auerbach, D. J.; Wodtke, A. M.; Schäfer, T. Dynamical Steering in an Electron Transfer Surface Reaction: Oriented NO(v = 3, 0.08 < Ei < 0.89 eV) Relaxation in Collisions with a Au(111) Surface. J. Chem. Phys. 2014, 140, 054710. (10) Bartels, N.; Krüger, B. C.; Auerbach, D. J.; Wodtke, A. M.; Schäfer, T. Controlling an Electron-Transfer Reaction at a Metal Surface by Manipulating Reactant Motion and Orientation. Angew. Chem. Int. Ed. 2014, 53, 13690−13694. (11) Cooper, R.; Li, Z.; Golibrzuch, K.; Bartels, C.; Rahinov, I.; Auerbach, D. J.; Wodtke, A. M. On the Determination of Absolute Vibrational Excitation Probabilities in Molecule-Surface Scattering: Case Study of NO on Au(111). J. Chem. Phys. 2012, 137, 064705. (12) Cooper, R.; Rahinov, I.; Li, Z.; Matsiev, D.; Auerbach, D. J.; Wodtke, A. M. Vibrational Overtone Excitation in Electron Mediated Energy Transfer at Metal Surfaces. Chem. Sci. 2010, 1, 55−61. (13) Cooper, R.; Bartels, C.; Kandratsenka, A.; Rahinov, I.; Shenvi, N.; Golibrzuch, K.; Li, Z.; Auerbach, D. J.; Tully, J. C.; Wodtke, A. M. Multiquantum Vibrational Excitation of NO Scattered from Au(111): Quantitative Comparison of Benchmark Data to Ab Initio Theories of Nonadiabatic Molecule−Surface Interactions. Angew. Chem. 2012, 124, 5038−5042. (14) Golibrzuch, K.; Shirhatti, P. R.; Rahinov, I.; Kandratsenka, A.; Auerbach, D. J.; Wodtke, A. M.; Bartels, C. The importance of accurate adiabatic interaction potentials for the correct description of electronically nonadiabatic vibrational energy transfer: A combined experimental and theoretical study of NO(v = 3) collisions with a Au(111) surface. J. Chem. Phys. 2014, 140, 044701. (15) Huang, Y.; Rettner, C. T.; Auerbach, D. J.; Wodtke, A. M. Vibrational Promotion of Electron Transfer. Science 2000, 290, 111− 114. (16) Wodtke, A. M.; Huang, Y.; Auerbach, D. J. Interaction of NO(v = 12) with LiF(001): Evidence for Anomalously Large Vibrational Relaxation Rates. J. Chem. Phys. 2003, 118, 8033−8041. 445

DOI: 10.1021/acs.jpclett.5b02448 J. Phys. Chem. Lett. 2016, 7, 441−446

Letter

The Journal of Physical Chemistry Letters (35) Rettner, C. T.; Auerbach, D. J.; Michelsen, H. A. Observation of Direct Vibrational Excitation in Collisions of H2 and D2 with a Cu(111) surface. Phys. Rev. Lett. 1992, 68, 2547−2550. (36) Velarde, L.; Engelhart, D. P.; Matsiev, D.; LaRue, J.; Auerbach, D. J.; Wodtke, A. M. Generation of Tunable Narrow Bandwidth Nanosecond Pulses in the Deep Ultraviolet for Efficient Optical Pumping and High Resolution Spectroscopy. Rev. Sci. Instrum. 2010, 81, 063106.

446

DOI: 10.1021/acs.jpclett.5b02448 J. Phys. Chem. Lett. 2016, 7, 441−446