Reaction dynamics at a metal surface; halogenation of Cu(110).

Scanning Tunnelling Microscopy (STM) is opening up a new field of reaction dynamics, followed one-molecule-at-a-time, only recently applied to reaction at a metal surface. Here we combine experiment with theory in studying the motions involved in the successive breaking by electron-induced reaction of the two carbon-halogen bonds, C-Cl or C-I, in physisorbed p-dihalobenzene, to form chemisorbed halogen-atoms and organic residue on Cu(110) at 4.6 K. We characterize the geometry of the physisorbed initial state, p-dichlorobenzene (pDCB) and p-diiodobenzene (pDIB), at the copper surface, as well as the successive final states of both chemisorbed reaction products: electron #1 giving rise to the first halogen-atom and a chemisorbed halophenyl and electron #2 giving a second halogen-atom and a chemisorbed phenylene. The major findings reported are (a) the distance and angular distributions of the chemisorbed reaction products relative to the physisorbed reagent molecule, (b) an approximate ab initio calculation, coupled with classical molecular dynamics (MD), of the repulsion between the products on the excited potential-energy surfaces, pes*, following excitation by electrons #1 or #2, and subsequently MD on the ground-state pes with inclusion of inelastic surface-interaction as a means to understanding the above, (c) observation of the changing dynamics with the chemistry of the halogen-atom, and (d) characterization of the effects of secondary encounters among the reaction products in the constrained space of the more highly localized reaction of pDIB. Item (d) shows clear evidence of high reactivity in surface-aligned collisions with restricted impact parameter, termed Surface Aligned Reaction, SAR, characterized by STM.

Molecular dynamics of haloalkane corral formation and surface halogenation at Si(111)-7 x 7.

Long-chain organic molecules, 1-halododecane, RX (X = Cl,Br), adsorbed on Si(111)-7 x 7 were shown to form stable dimeric corrals; type I around corner holes and type II around corner adatoms S. Dobrin et al. [Surf. Sci. Lett. 600, L43 (2006)]. Here we examine the molecular dynamics of corral formation, in which mobile physisorbed adsorbates spontaneously convert to immobile. At high coverage the mechanism gives evidence of involving collisions between mobile vertical monomers, giving types I and II immobile horizontal dimers, vD +vD -->h2 (I, II). At low coverage mobile vertical monomers collide with immobile horizontal ones to form largely type-II corrals, vD + h-->h2 (II). Thermal reaction of corrals with X = Br brominates the surface by two distinct molecular pathways, thought to have more general applicability: "daughter-mediated" reaction of vertical v(A) with a low activation energy (here Ea approximately 5 kcal mol(-1)) and "parent-mediated" reaction of horizontal h or h2 with high activation energy (here Ea = 29 kcal mol(-1)).

Dynamics of harpooning studied by transition state spectroscopy. Part III. Li...FCH3.

The Van der Waals complex Li...FCH3 has been formed in a crossed molecular beam apparatus. The transition state (TS) for the reaction Li*(2p 2P) + FCH3-->LiF + CH3 was accessed at various configurations by laser-excitation of the Li...FCH3 complex by tunable visible radiation, lambda 1. Photoinduced depletion of the complex by excitation to this TS was found to occur across a broad range of lambda 1 from 570 to 850 nm. This 'action spectrum' consisted of two broad regions located to either side of the atomic transition line of Li (2p 2P<--2s 2S). The first region, between 700 and 850 nm, was dominated by sharp maxima in the depletion intensity. A broad peak with weakly-resolved structure characterized the second region, between 570 and 680 nm. These findings were interpreted by means of high-level ab initio calculations of the potential-energy surfaces in the TS region. The peaks in the photodepletion spectrum were assigned to specific electronic transitions, their shapes and intensities being explained in terms of calculated transition-dipole moments and rovibrational wavefunctions.

Photoinduced charge-transfer reaction at surfaces. Part I. (HCl)m..Nan/LiF(001) + hv (640 nm)-->(HCl)m - 1 Cl-Nan+/LiF(001) + H(g).

A sub-monolayer of atomic sodium, Nan, was deposited on LiF(001) at 50 K and characterized by temperature-programmed desorption, X-ray photoelectron spectroscopy, and titration with HCl. The Nan was dosed with HCl to form (HCl)m..Nan/LiF(001), which was then irradiated by 640 nm laser-radiation to induce a charge-transfer (CT) reaction. Reaction-product atomic H(g) was observed leaving the surface, by two-color Rydberg-atom time-of-flight (TOF) spectroscopy. These H-atoms gave evidence of arising from the photoinduced harpooning reaction between the sodium clusters, Nan, on the substrate, and (HCl)m adsorbed on the Nan. The translational energy distribution, its vibrational structure, and the angular distribution of H(g) gave information regarding the harpooning event. Translationally and vibrationally excited HCl(g) was shown, by resonance-enhanced multiphoton ionization (REMPI), to be formed as an alternate product; by way of (HCl)m..Nan/LiF(001) + 602 nm-->(HCl)m - 1 Nan/LiF(001) + HCl(g)(v > or = 0).

Some concepts in reaction dynamics.

The objective in this work has been one which I have shared with the two other 1986 Nobel lecturers in chemistry, D. R. Herschbach and Y. T. Lee, as well as with a wide group of colleagues and co-workers who have been responsible for bringing this field to its current state. That state is summarized in the title; we now have some concepts relevant to the motions of atoms and molecules in simple reactions, and some examples of the application of these concepts. We are, however, richer in vocabulary than in literature. The great epics of reaction dynamics remain to be written. I shall confine myself to some simple stories.

Nonequilibrium processes.

Nonequilibrium phenomena have been studied for over half a century, particularly as a means to understanding the mechanism of energy transfer. Application of the insights and techniques of molecular physics to chemistry has resulted in a view of chemistry as constituting an aspect of the study of strong collisions, and chemical reaction as a special type of energy transfer. Increasing use has been made in experimental work of nonequilibrium environments for the study of chemical processes. The nature and purpose of such experiments are reviewed here, very briefly, and an attempt is made to point to areas that appear ripe for development over the coming decade.

Energy Distribution Among Reaction Products. III: The Method of Measured Relaxation Applied to H + Cl(2).

The method of measured relaxation is described for the determination of initial vibrational energy distribution in the products of exothermic reaction. Hydrogen atoms coming from an orifice were diffused into flowing chlorine gas. Measurements were made of the resultant ir chemiluminescence at successive points along the line of flow. The concurrent processes of reaction, diffusion, flow, radiation, and deactivation were analyzed in some detail on a computer. A variety of relaxation models were used in an attempt to place limits on k(upsilon), the rate constant for reaction to form HCl in specified vibrational energy levels: [equation]. The set of k(upsilon) obtained from this work is in satisfactory agreement with those obtained by another experimental method (the method of arrested relaxation described in Parts IV and V of the present series).

Energy Distribution Among Reaction Products: H + SCl(2) ? HCl + SCl.

The method of arrested relaxation has been applied to the study of infrared chemiluminescence arising from HCl(upsilon'>0) formed in the reaction H + SCl(2) ? HCI + SCl. The main findings are (a) that the mean fraction of the available energy entering vibration in the new bond is similar to that for the reaction H + Cl(2) ? HCl + Cl, the value being f (upsilon') approximately 0.43 (assuming the energy available for distribution among the products to be E'(tot) = 48 kcal mole(-1)); (b) that the breadth of the distribution over the populated vibrational levels, upsilon' = 1-5, substantially exceeds that for H + Cl(2) ? HCl + Cl, indicating altered reaction dynamics; (c) that the rotational distribution in the newly formed HCl is double-peaked, suggesting two distinct types of reaction dynamics (neither closely resembling that for H + Cl(2)), one of which gives rise to HCl with modest rotational excitation and one to HCl with a high degree of rotational excitation; (d) the mean fraction of E'(tot) entering product rotation (averaged over all reactive encounters) is f (R') asymptotically equal to 0.19; (e) the group of reactive encounters that result in less rotationally excited HCl are distinguished from the group that result in highly rotationally excited HCl, by differing correlation with upsilon'; this is further evidence that two distinct types of dynamics are involved in the reaction H + SCl(2) ? HCl + SCl.

Formation of Vibrationally Excited OH by the Reaction H + O(3).

An earlier study [Chem. Phys. Lett. 1, 619 (1968)] concluded that the reaction H + O(3) ? OH + O(2) forms OH predominantly in the highest accessible vibrational levels, upsilon = 8 and 9. We have extended this earlier work (1) by using fourier transform spectroscopy which is capable of giving more precise values for the relative vibrational populations at low intensities, (2) by recording emission down to lower background pressures (1 x 10(-4) Torr), and (3) by treating the vessel walls so as to remove OHdagger (vibrationally excited OH in it ground (2)II electronic state) more effectively. This involved using a room temperature vessel coated with silica gel. Under these conditions (provided that the values available for the radiational lifetime of OHdagger are correct) vibrational relaxation of OHdagger should have been largely arrested. We conclude that the relative rate constants for formation of OHdagger in levels upsilon are k(upsilon = 6) < 0.4, k(upsilon = 7) asymptotically equal to 0.4, k(upsilon = 8) asymptotically equal to 0.8, and k(upsilon = 9) = 1.00.

Infrared-Emission Studies of Electronic-to-Vibrational Energy Transfer. IV: Hg + HF.

Electronic-to-vibrational transfer has been observed for the process Hg* (6(3)P(1,0)) + HF(upsilon=0) ? Hg(6(1)S(0)) + HFdagger(upsilon=1-6). (The asterisk symbolizes electronic excitation, the dagger indicates vibrational excitation.) The observations were made by recording ir emission from HFdagger(v=1-6). A set of relative rate constants k(upsilon) were obtained from the observed stationary-state concentrations N(upsilon). The k(upsilon) decrease rapidly with increasing upsilon, up to upsilon = 6. From a comparison of the total HF emission intensity with total CO emission (due to Hg* + CO) a rough absolute cross section Q(E,V) ~ 5 A(2) was obtained for electronic-to-vibrational transfer in Hg* + HF collisions.