Direct and Indirect Detachment in the Iodide ... - ACS Publications

Nov 20, 2009 - Direct and Indirect Detachment in the Iodide-Pyrrole Cluster Anion: The Role of Dipole. Bound and Neutral Cluster States†. Foster Mba...
1 downloads 0 Views 775KB Size
Download PDF
J. Phys. Chem. A 2010, 114, 1539–1547

1539

Direct and Indirect Detachment in the Iodide-Pyrrole Cluster Anion: The Role of Dipole Bound and Neutral Cluster States† Foster Mbaiwa, Matthew Van Duzor, Jie Wei, and Richard Mabbs* Department of Chemistry, Campus Box 1134, Washington UniVersity in St. Louis, Missouri 63130 ReceiVed: September 4, 2009; ReVised Manuscript ReceiVed: October 29, 2009

Photoelectron imaging probes both molecular electronic structure and electron molecule interactions. In the current work images were recorded for detachment from the I- · C4H5N (I- · pyrrole) cluster anion at wavelengths between 360 and 260 nm. The direct detachment spectra show strong similarities to those of I-, although a strong solvent shift, broadening and some structure is observed. A nondirect, dissociative or autodetachment feature is also observed over a range of wavelengths. Ab initio calculations identify several local minima associated with neutral and anion isomers. Energy and Franck-Condon arguments are used to assess the role of these in the detachment process. The cluster anion structure is essentially an I- atomic anion in the presence of a neutral pyrrole molecule. The spectral structure arises due to interactions in the open shell neutral cluster residue resulting from detachment. The indirect detachment feature arises through the formation of an intermediate dipole bound cluster anion state which subsequently dissociates. The energy dependence of this channel (observed over a 0.6 eV range of photon energies) is discussed in terms of the wide amplitude motions associated with the van der Waals modes of the cluster anions. 1. Introduction Photodetachment spectroscopy is a powerful tool in the study of the electronic structure and dynamics of both anions and the resultant neutral species.1 Velocity mapped photoelectron imaging extends this technique2-4 to allow relatively straightforward, simultaneous measurement of the photoelectron spectrum and photoelectron angular distribution (PAD). A molecular anion photoelectron spectrum contains information on the energy eigenvalues of the parent negative ion and the residual neutral species. The presence of spectral structure is often associated with vibrational excitation in the residual neutral due to the change in molecular geometry accompanying the loss of the excess electron. The PAD, in the absence of significant interaction with the neutral residue, is determined by the nature of the parent orbital and the selection rules of the detachment process.5 Cluster anions represent particularly interesting cases. Studies of cluster anion photoelectron spectra and PADs yield insights into the nature of the binding interaction between the anion and neutral species and the interaction of the departing electron with the residual neutral framework of the cluster.5,6 In cluster anions based upon a negatively charged atomic moiety, such a structure can be associated with delocalization of the electron onto the accompanying “neutral” solvent molecules. Since the photoelectron angular distribution (PAD), in the absence of significant intracluster interactions, is determined by the nature of the parent orbital and the selection rules of the detachment process,5 comparison of the PADs from free and clustered anions can be a sensitive indicator of changes in the electronic structure of the anion upon binding. Monosolvated cluster anions, in which the excess electron is localized on an atomic anion moiety represent ideal environments for studies of the electron kinetic energy (eKE) dependence of electron-molecule interactions.7-9 In this context these †

Part of the “W. Carl Lineberger Festschrift”. * Corresponding author. E-mail: [email protected].

cluster anions can be conceptualized as an electron source (the anion) in the proximity of a target species. Photodetachment from the anion produces electrons in the vicinity of the target. The use of a tunable laser provides eKE control as a result of energy conservation. Strictly speaking, photoexcitation accesses excited states of the cluster arrangement that have similarities to the states associated with the neutral cluster-free electron interaction. Recently we have observed drastic changes in the photoelectron angular distribution (PAD) associated with Idetachment when clustered with a CH3I molecule.5,6 Furthermore, for I- · CH3I detachment, the variation of the PAD as a function of eKE shows interesting structure associated with the presence of a state analogous to the vibrational Feshbach resonance responsible for the temporary capture of a free electron by CH3I.5,10-14 In this paper we turn our attention to photodetachment from the I- · C4H5N (I- · pyrrole) cluster anion. The interaction of the pyrrole molecule with anions is particularly interesting, since these molecules are important subunits of the anion receptor molecules (calix[n]pyrroles)15-18 that form the basis of many colorimetric anion specific sensors (effectively anion indicators). We compare the cluster anion photodetachment spectra to those of I-, noting a splitting of the lowest binding energy feature and a solvent shift that shows the ion-molecule interaction to be much stronger than that for I- · H2O,7 despite the similar dipole moments of the two molecules.19,20 We also observe a nondirect detachment process that competes with direct detachment from the cluster. A similar feature was observed in the photodetachment of I- · aniline at 257 nm.8 In the latter case this process was ascribed to an intracluster electron transfer process creating a dipole bound cluster anion state that subsequently dissociates, liberating the electron in the process. Such a process could be reasonably expected to have a dependence on the photodetachment energy; for example, a dipole bound state typically lies just below a neutral electronic state. In the case of I- · aniline detachment, coincidence studies were performed by encompassing four photon energies between

10.1021/jp9085798  2010 American Chemical Society Published on Web 11/20/2009

1540

J. Phys. Chem. A, Vol. 114, No. 3, 2010

Mbaiwa et al.

3.60 and 4.82 eV.21 The indirect detachment signature was observed in each measurement but was found to be closely associated with the nearest neutral electronic state. This suggests an energy limit for the indirect detachment process to compete effectively with direct detachment. By using a tunable dye laser to photoexcite the I- · C4H5N cluster anion, we are able to probe the energy dependence of a similar charge transfer process and find that the indirect detachment channel is seen over a relatively broad energy range (0.6 eV) for a given channel, which is initially surprising given the weakness of the dipole binding interaction. Using the results of ab initio calculations for support, we will discuss these findings in the context of the cluster structure and dynamics subsequent to photoexcitation, in particular examining the role of the dipole bound state in the indirect detachment process. 2. Experimental Section The photoelectron velocity map imaging spectrometer used in this work is a design similar to that described in detail in ref 22.Inbrief,theinstrumentcomprisesanionsource,Wiley-McLaren time-of-flight (TOF) mass spectrometer,23 and a velocity mapped photoelectron imaging arrangement.2 The ambient vapor pressure of methyl iodide is mixed with 200 psig of argon. This mixture is flowed over pyrrole at a backing pressure of 5 psig and supersonically expanded through a pulsed nozzle (General valve, series 9, 0.76 mm orifice diameter). Anions are formed by secondary electron attachment after impact ionization of the Ar carrier gas with a continuous 1 kV electron beam which crosses the expansion a few nozzle diameters downstream. The anions are pulse extracted into the TOF tube, accelerated to 1.95 kV, and re-referenced to ground using a potential switch. The ions are steered toward a microchannel plate detector (MCP) some 2 m downstream by means of electrostatic deflectors and are spatially focused using an Einzel lens. The desired ion is selected by synchronizing a linearly polarized laser pulse to the arrival time. The frequency doubled (BBO crystal) output (4-14 mJ/pulse) from a tunable dye laser (Cobra-Stretch, Sirah Laser pumped at 10 Hz by the second or third harmonic of an INDI10 Nd: YAG laser, Spectra-Physics Inc.) produces photoelectrons at detachment wavelengths between 370 and 260 nm. The 3-D laboratory frame photoelectron distribution is projected onto a 2D position sensitive imaging detector (Burle Inc.) comprising a chevron type MCP (40 mm diameter, 10 µm pore, MgO coated input face) and a P20 phosphor screen using the three electrode velocity mapping arrangement described by Eppink and Parker.2 Our imaging experiments are performed in a perpendicular arrangement. The detector is located 20 cm from the point of intersection of the ion and laser beams along the normal to their mutual plane. The laser electric vector is polarized parallel to the plane of the detector. Images are recorded by means of a CCD camera (Imperx VGA 120) with a resolution of 640 × 480 pixels at a rate of 10 frames per second. Each frame corresponds to a millisecond exposure. Typically, 2500 frames were accumulated to obtain a single image. Photoelectron angular distributions and spectra are extracted from the accumulated images using the basis set expansion (BASEX) method of Dribinski et al.24 3. Results Photoelectron images were recorded for I- and I- · C4H5N at 5 nm intervals throughout the range 360-260 nm. No photoelectrons were observed at wavelengths above 360 nm. Representative images and the corresponding spectra of I- · C4H5N are shown in Figure 1a-d. Figure 1e shows an image corre-

Figure 1. Photoelectron images and spectra (as a function of electron binding energy) of I- · C4H5N at (a) 345 nm, (b) 325 nm, (c) 315 nm, (d) 260 nm, and (e) I- at 300 nm. In all images the direction of the laser polarization is as indicated by the arrow in (a).

sponding to free atomic iodide anion detachment at 300 nm, which also serves as the imaging detector calibration in these experiments. Photoelectron spectra were obtained by integrating the inverted images over all angles at individual radial values and converting the radii to electron velocities using the well characterized photoelectron spectroscopic properties of I-.25 The velocity domain photoelectron spectrum is then converted to the energy domain by the appropriate Jacobian transformation. The spectra in Figure 1 are presented as a function of electron binding energy (eBE) which is determined according to energy conservation (eKE ) hc/λ - eBE, with λ being the detachment photon wavelength) to facilitate comparison between results. At 260 nm, the shortest wavelength used in this study, three features are apparent in the I- · C4H5N photoelectron image and corresponding spectrum (Figure 1d). There are two narrow, welldefined peaks separated by 0.95 ( 0.02 eV. The eBE of these two features remains constant for all the cluster anion images, regardless of the photon energy. The 260 nm spectrum shows a strong resemblance to that obtained from free I- photodetachment, in particular the transition spacing is the same as the spin orbit splitting in the free iodine atom (0.94 eV), within the limit of our experimental error. The most obvious difference between the clustered and free I- photoelectron spectra is the increase in the eBE in I- · C4H5N, a solvation induced stabilization of 0.66 eV. The two anisotropic rings in the I- image of Figure 1e (and hence two peaks in the spectrum of I-) correspond to photodetachment from I-(1S0) resulting in the production of the I(2P3/ 2 2 2) and I( P1/2) spin-orbit states. Of these, the P3/2 state has the lower energy and so the outer ring in the image (higher eKE feature) corresponds to this detachment channel. The similarity of the 260 nm photoelectron spectrum of I- · C4H5N strongly indicates the localization of the excess electron on the I atom within the cluster and relatively little perturbation of the pyrrole moiety. This localization of the excess charge on the halogen

Detachment in the Iodide-Pyrrole Cluster Anion atom has been noted in other cluster anions of this type and is fairly common.5,6,8 For convenience of discussion we will refer to the two narrow features in the I- · C4H5N spectrum at 260 nm according to the analogous free I- transitions, although this labeling obscures some of the details of the cluster anion detachment, as discussed later. The images of Figure 1b,c show only the “2P3/2” direct detachment channel as the photon energy lies below the “2P1/2” detachment threshold. When the differences between the I- · C4H5N and I- spectra are examined in more detail, the diminishing resolution of the detector energy resolution with increasing eKE must be taken into account. This reduction is clearly seen in the I- spectrum of Figure 1e, where the fwhm of the Lorentzian line shape of the 2P3/2 feature is 0.085 eV, but the corresponding width of the 2P1/2 transition is only 0.023 eV. Thus, when considering the width of our spectral features, only transitions with similar eKEs are directly comparable. Applying this caveat, we note some subtle differences between the atomic and cluster anion spectra. In general, the transitions in the I- · C4H5N spectra are broader than those of I- for a given eKE. Additionally, comparing the different detachment channels of the cluster anion at similar eKEs shows that the “2P3/2” feature is consistently broader than the “2P1/2” transition and we can partially resolve structure at low eKE (e.g., the 315, 320, and 325 nm spectra of Figures 1 and 2). The splitting between the two spectral features contributing to the “2P3/2” peak can be estimated by fitting two Lorentzian functions of the same fwhm (∼0.08 eV) revealing a peak to peak separation of 0.06 eV as shown in Figure 2a. Figure 2b (which compares the I- 303 nm and I- · C4H4N 325 and 260 nm spectra, scaled relative to the individual spectral maxima, over the same eKE range) clearly shows that no similar structure is observed over the same eKE range in the “2P1/2” channel. An interesting feature appears in most of our I- · C4H5N images, a tight central spot. While its relative intensity varies with respect to the “2P3/2” and “2P1/2” features, it is nevertheless present in all cluster anion images recorded between 360 and 310 nm (at 5 nm intervals) and reappears at 270 nm. This therefore corresponds to a near zero eKE feature in most of our images, and the eBE corresponding to this feature appears to depend upon the energy of the detachment photon. This is a signature of an indirect or autodetachment mechanism mediated by an excited intermediate state. The relative amount of these low energy electrons (compared to the “2P3/2” and “2P1/2” channels) varies with photon energy and reaches a maximum close to the “2P3/2” threshold. Intriguingly, these low energy electrons are still produced at photon energies significantly below the “2P3/2” vertical detachment energy. The long wavelength threshold for these electrons appears to be 360 nm, a photon energy of 3.45 eV. In contrast, the observed “2P3/2” direct detachment spectral maximum corresponds to an eBE of 3.72 eV. From the sample images of I- · C4H5N in Figure 1, it is clear that the photoelectron angular distributions (PADs) associated with the “2P3/2” transitions are not necessarily isotropic and that the degree of anisotropy has some dependence on the eKE. The PAD is characterized by the anisotropy parameter β, which can be found by fitting the angular distributions to the equation,

I(θ) ) σ/4π[1 + βP2(cos θ)] where P2(cos θ) is the second-order Legendre polynomial and θ is the angle between the photoelectron velocity vector (in the plane of the detector) and the direction of the laser polarization vector. The value of β is constrained to lie between -1 (a sine

J. Phys. Chem. A, Vol. 114, No. 3, 2010 1541

Figure 2. (a) Photoelectron spectrum (open circles) of I- · C4H5N at 320 nm. Dash-dot lines represent Lorentzian functions used to fit the spectrum and the solid line is a convolution of these. (b) Photoelectron spectra comparing the “2P3/2” feature of I- · C4H5N at 325 nm (open circles), “2P1/2” feature of I- · C4H5N at 260 nm (solid line) and the 2 P1/2 feature of I- at 303 nm (dash-dot line) at the same eKE.

squared distribution) and +2 (a cosine squared distribution). The variation of β as function of eKE is shown in Figure 3 for both I- and I- · C4H5N. The trend for the value of β for I- · C4H5N is obviously very similar to that of I- and shows the familiar Bethe-Cooper-Zare behavior26-28 expected for detachment from an atomic p orbital. 4. Ab Initio Calculations Three major features of the above results stand out and require closer consideration: (i) the strength of the anion-neutral binding interaction (reflected in the solvent shift), which is larger than encountered in other systems where the neutral moiety has a dipole moment similar to that for pyrrole; (ii) the nature of broadening of the “2P3/2” and “2P1/2” transitions and in particular the observed splitting in the near threshold direct detachment spectra of the “2P3/2” channel;

1542

J. Phys. Chem. A, Vol. 114, No. 3, 2010

Mbaiwa et al.

Figure 3. β vs eKE for I- · C4H5N (“2P3/2” feature, open squares), I(2P3/2 feature, filled triangles), and I- (2P1/2 feature, open triangles).

(iii) the origin of the low eKE feature and in particular its persistence over a surprisingly large photon energy range. To aid in the explanation of these observations, we have performed a series of ab initio calculations that we present below. These calculations were performed using the Q-CHEM29 program through the University of Southern California’s Center for Computational Studies of Electronic Structure and Spectroscopy of Open-Shell and Electronically Excited States. 4.1. Cluster Anion and Neutral Geometry Optimization. Cluster anions often have several close lying isomers, while consideration of the photodetachment transition energies also requires consideration of the resultant neutral species. Cluster anion and neutral cluster geometries were determined using the MP2 method and the following basis sets. For C, H, and N atoms we use the aug-cc-pvdz basis set while the crenbl pseudopotential and basis set as modified by Combariza et al.30 is used for I, supplementing the basis set by the addition of six diffuse sp functions. As in ref 31, these have exponents ranging from 7.35 × 10-3 to 2.36 × 10-6 with a ratio of 5 between subsequent members of this series. The results of the geometry optimization are shown in Figure 4. Initiating the optimization for the closed shell cluster anion using different I atom locations (A-E) reveals only two isomers. These lie 0.644 eV apart with the more stable isomer of I- · C4H5N (A1) retaining the C2V symmetry of the pyrrole subunit with a collinear N-H-I segment. The H-I distance is 2.465 Å and there is very little perturbation of the pyrrole moiety. For example, in isomer A1 the N-H bond experiences the largest change upon cluster anion formation, a difference of 0.025 Å or an elongation of just 2.5%. Isomer A2 has the I atom on the opposite side of the pyrrole unit, and we argue later that there is no evidence of this isomer contributing to our photodetachment images. For the open shell, neutral cluster a similar approach reveals four minima associated with the ground electronic state potential. The global minimum corresponds to structure N1, in which the I atom is located above the plane of the pyrrole ring (but not directly above the C2 axis) and toward the opposite end of the molecule to the N atom. In the second neutral isomer N2, the iodine is displaced from the C2 axis of pyrrole resulting in a 140° N-H---I angle. The third neutral isomer (N3) is nearly isoenergetic with N2 and corresponds to a structure similar to that of the cluster anion A1 isomer. The I atom again lies along

Figure 4. Neutral (N1-N4) and cluster anion (A1-A2) MP2 optimized ab initio geometries. Approximate starting points for neutral and cluster anion optimization are indicated as A-E (A lies above the plane of the pyrrole moiety). Erel is the isomer energy relative to A1.

the C2 axis, but at a greater distance than in the cluster anion (2.815 Å). Neutral isomer N4 has a structure similar to that of the anion A2 isomer. The relative energies of these minima to the A1 anion isomer (neglecting zero point energy) are also shown in Figure 4, while the dipole moments associated with each neutral cluster isomer at their equilibrium geometries are µN1 ) 3.45 D, µN2 ) 2.37 D, µN3 ) 2.58 D, and µN4 ) 1.89 D. Verification that the calculated structures correspond to true stable isomers (as opposed to transition states) requires calculation of the vibrational frequencies. For each isomer in Figure 4, the vibrational frequencies associated with all 27 modes are real. As shown in the Franck-Condon calculations presented later, the van der Waals modes (ν1-ν3) are of greatest relevance. In particular, the ν3 mode (of A1 and N3) corresponds to motion along the (N)H-I direction while mode ν2 leads to displacement of the I atom from the plane containing the pyrrole moiety. 4.2. Detachment Energies and Neutral Electronic States. The vertical detachment energy corresponds to the difference between the anion and higher lying neutral state energies at the anion equilibrium geometry. Experimentally, this is determined using the spectral maximum, with the provision that the photon energy exceeds the VDE. Calculation predicts the VDE from A1 to be 3.77 eV (neglecting zero point energy), in very close agreement with our experimentally measured 3.72 eV in the “2P3/2” channel. Vertical detachment from the A2 isomer requires 3.11 eV, but since we do not see any detachment at photon energies 2.0 D).38 Their conclusion was that the indirect detachment is mediated by a dipole bound anion state. Subsequent dissociation of the dipole bound cluster reduces the dipole moment and leads to electron loss.8,21 It is also interesting to note that the indirect detachment features observed by Bowen et al.21 were associated with different close lying neutral states as the detachment wavelength changed. In the case of the iodidepyrrole cluster detachment, the loss of the indirect detachment signal at 310 nm and subsequent reappearance at 270 nm is consistent with this finding. At 310 nm indirect detachment due to the dipole bound state associated with the “2P3/2” channel is no longer favorable compared to direct detachment. However, as we increase the photon energy, the dipole bound surface lying just below the neutral “2P1/2” or III surface becomes accessible. The observation of indirect detachment in the I- · C4H5N cluster anion suggests that this phenomenon might be relatively common in cluster anions. For mediation by a dipole bound state the requirement is that vertical excitation produces a highly polar neutral framework in which the individual substituents have dipole moments lower than the critical value of 2.0 D. Our ab initio calculated dipole moment for the N3 cluster framework at the equilibrium geometry of A1 is ∼2.7 D. This is sufficient to support a dipole bound state although the closeness of the neutral cluster dipole moment to the critical value suggests that the dipole bound anion is energetically very close to the neutral cluster. Our calculation bears out this prediction, yielding a difference between the dipole bound and neutral potential curves of ∼0.14 meV at this geometry, as illustrated in Figure 7a. In the current case the indirect detachment process may be represented schematically as, I- · C4H5N + hν f(I · C4H5N)*f I + C4H5N + e-. Calculations for the C2V, N3 isomer indicate the presence of a shallow minimum along the (N)H-I coordinate (Figure 5). However, the cluster anion and dipole bound cluster anion minima occur at different (N)H-I distances. Vertical excitation accesses the repulsive wall of the dipole bound species (Figure 6) at energies predominantly close to or above the 0.14 eV dissociation limit. Subsequent motion along the intermoiety axis lengthens the physical bond and therefore reduces the dipole moment. Eventually, the dipole moment falls

1546

J. Phys. Chem. A, Vol. 114, No. 3, 2010

below the critical value required to support the dipole bound state and the electron is lost (corresponding to the region where the dipole bound and neutral diabats reverse their energy ordering). Given the small difference between the dipole bound anion and neutral curves, we might expect to observe indirect detachment over a very narrow energy window, just below the direct detachment threshold. This is clearly not the case and the autodetachment feature is observed up to 0.35 eV above, and 0.27 eV below the “2P3/2” threshold. Understanding the persistence of the indirect detachment feature above threshold is relatively simple. The potential curves of Figure 6 show that the repulsive region of the dipole bound state potential lies in the region covering the range of probable cluster anion ground state zero point (N)H-I separations. Hence the overlap between the continuum nuclear wave functions of the dipole bound state and the bound nuclear wave functions of the cluster anion ground state is likely to be good for a relatively large range of photon energies. Crude estimation of this range can be made by comparing the difference between the dipole bound anion curve and the cluster anion ground state curve. The vertical arrows in Figure 6b correspond to the turning points of the cluster anion ground state zero point level in a simplistic diatomic-like picture. These show that photon excitation from 3.74 to 3.89 eV can feasibly access dipole bound states immediately below the I and II neutral states. The upper limit of this estimate is within 200 meV of the photon energy at which the autodetachment feature disappears above the “2P3/2” channel threshold. Given the very approximate nature of the above approach, which neglects hot band excitation and geometries other than those of C2V symmetry, the production of dipole bound cluster anions up to 350 meV above threshold seems reasonable. The persistence of the indirect detachment feature at photon energies down to 270 meV below the A1 vertical detachment energy is more puzzling. Within the diatomic approximation of Figure 6, the neutral N3 isomer is energetically inaccessible from the A1 curve for photon energies 0.4 eV, lending plausibility to the arguments advanced above. 6. Conclusion We have made photodetachment imaging measurements of the I- · C4H5N cluster anion over a range of wavelengths between 370 and 260 nm. Comparison of the photoelectron spectra and angular distributions with those of free I- detachment, and ab initio calculations show that the cluster anion is essentially composed of an atomic iodide anion moiety in the proximity of a relatively unperturbed neutral pyrrole molecule. Physical interaction with both the dipole and quadrupole moments of the pyrrole molecule stabilize the anionic moiety, leading to a shift in the photoelectron spectrum that is large compared to other molecules with a similar dipole moment. Structure is observed in the “2P3/2” cluster anion spectral transition. However, our results and calculations confirm that this is not due to delocalization of the electron onto the pyrrole moiety, which would cause vibrational excitation of the neutral molecule. Instead, the explanation lies with the effect of the pyrrole moiety on the open shell iodine atom within the residual neutral cluster. In addition to the direct detachment features there is a low energy, autodetachment feature that we ascribe to the production of an intermediate, dipole bound cluster anion state, which subsequently undergoes dissociation. A similar feature has been observed previously in detachment from the I- · aniline cluster anion. We suggest that this may be a relatively common phenomenon in clusters containing molecules whose dipole moments are close to the critical 2.0 D. The wavelength range of the current measurements allow us to assess the energy dependence of the dipole bound channel and reveal that it occurs over a surprisingly wide range of excitation energy. We conjecture that this is due to the wide amplitude motion associated with the van der Waals modes of the cluster anion which allows excitation to a wide range of excited state geometries.

Detachment in the Iodide-Pyrrole Cluster Anion Acknowledgment. We thank the National Science Foundation (Grant CHE-0748738) for support of this work. We are also grateful for the opportunity to use the USC iopenshell environment. References and Notes (1) Ervin, K. M.; Lineberger, W. C.; Ervin, K. M.; Lineberger, W. C. In AdVances in Gas Phase Ion Chemistry; Babcock, L. M., Ed.; JAI Press Inc.: Greenwich, CT, 1992; Vol. 1, p 121. (2) Eppink, A. T. J. B.; Parker, D. H. ReV. Sci. Instrum. 1997, 68, 3477. (3) Mabbs, R.; Grumbling, E. R.; Sanov, A. Chem. Soc. ReV. 2009, 38, 2169. (4) Sanov, A.; Mabbs, R. Int. ReV. Phys. Chem. 2008, 27, 53. (5) Mabbs, R.; van Duzor, M.; Mbaiwa, F.; Wei, J. JPCS, in press. (6) van Duzor, M.; Mbaiwa, F.; Wei, J.; Mabbs, R. J. Chem. Phys., submitted for publication. (7) Mabbs, R.; Surber, E.; Sanov, A. J. Chem. Phys. 2005, 122, 054308. (8) Bowen, M. S.; Beccucci, M.; Continetti, R. E. J. Phys. Chem. A 2005, 109, 11781. (9) Dessent, C. E. H.; Bailey, C. G.; Johnson, M. A. J. Chem. Phys. 1995, 103, 2006. (10) Cyr, D. M.; Bishea, G. A.; Scarton, M. G.; Johnson, M. A. J. Chem. Phys. 1992, 97, 5911. (11) Dessent, C. E. H.; Bailey, C. G.; Johnson, M. A. J. Chem. Phys. 1996, 105, 10416. (12) Hotop, H.; Ruf, M.-W.; Fabrikant, I. I. Phys. Scr. 2004, T110, 20. (13) Schramm, A.; et al. J. Phys. D 1999, 32, 2153. (14) Weber, J. M.; et al. Eur. Phys. J. D 2000, 11, 247. (15) Miyaji, H.; et al. Chem. Commun. 1999, 1723. (16) Gale, P. A.; Sessler, J. L.; Kral, V. Chem. Commun. 1998, 1. (17) Gale, P. A.; Twyman, L. J.; Handlin, C. I.; Sessler, J. L. Chem. Commun. 1999, 1851. (18) Cafeo, G.; et al. Chem. Commun. 2000, 1207.

J. Phys. Chem. A, Vol. 114, No. 3, 2010 1547 (19) CRC Handbook of Chemistry and Physics, 89th ed.; CRC Press: Boca Raton, FL. (20) McClellan, A. L., Tables of Experimental Dipole Moments; W. H. Freeman: San Francisco, CA, 1963. (21) Bowen, M. S.; Beccucci, M.; Continetti, R. E. J. Chem. Phys. 2006, 125, 133309. (22) Surber, E.; Mabbs, R.; Sanov, A. J. Phys. Chem. A 2003, 107, 8215. (23) Wiley, W. C.; McLaren, I. H. ReV. Sci. Instrum. 1955, 26, 1150. (24) Dribinski, V.; Ossadtchi, A.; Mandelshtam, V. A.; Reisler, H. ReV. Sci. Instrum. 2002, 73, 2634. (25) Moore, C. E. Natl. Bur. Std. Circular 1958, 467, 245. (26) Bethe, H. A. Bethe, H. A. In Handbuch Der Physik; Geiger, H., Scheel, W.; Eds.; Springer: Berlin, 1933; Vol. 24, p 483 (27) Cooper, J.; Zare, R. N. J. Chem. Phys. 1968, 48, 942. (28) Cooper, J.; Zare, R. N. J. Chem. Phys. 1968, 49, 4252. (29) Shao, Y. Q-Chem, 3.2 ed.; Q.Chem Inc.: Pittsburgh, PA, 2007. (30) Combariza, J. E.; Kestnet, N. R.; Jortner, J. J. Chem. Phys. 1994, 100, 2851. (31) Chen, H. Y.; Sheu, W. S. J. Am. Chem. Soc. 2000, 122, 7534. (32) Sanov, A.; Faeder, J.; Parson, R.; Lineberger, W. C. Chem. Phys. Lett. 1999, 313, 812. (33) Evans, D. H.; Keesee, R. G.; Castleman, A. W., Jr. J. Chem. Phys. 1987, 86, 2927. (34) Sutter, D. H.; Flygare, W. H. J. Am. Chem. Soc. 1969, 91, 6895. (35) Carles, S.; Lecomte, F.; Schermann, J. P.; Desfrancois, C. J. Phys. Chem. A 2000, 104, 10662. (36) Mozhayskiy, V. A.; Krylov, A. I. Ezspectrum, http://Iopenshell.Usc.Edu/Downloads. (37) Younkin, J. M.; Smith, L. J.; Compton, R. N. Theor. Chim. Acta 1976, 41, 157. (38) Crawford, O. H.; Garret, W. R. J. Chem. Phys. 1977, 66, 4968. (39) Bavia, M.; Bertinelli, F.; Taliani, C.; Zauli, C. Mol. Phys. 1976, 31, 479.

JP9085798