Ionic Pathways following UV Photoexcitation of the (HI)2 van der

Oct 14, 2009 - 630090 Russia, NoVosibirsk State UniVersity, PirogoVa Street 2, NoVosibirsk 630090, Russia, and Institute of. Semiconductor Physics ...
0 downloads 0 Views 2MB Size
Download PDF
J. Phys. Chem. A 2010, 114, 3067–3073

3067

Ionic Pathways following UV Photoexcitation of the (HI)2 van der Waals Dimer† Konstantin V. Vidma,‡,§,| David H. Parker,*,‡ Georgii A. Bogdanchikov,§,| Alexey V. Baklanov,§,| and Sergei A. Kochubei⊥ Institute for Molecules and Materials, Radboud UniVersity Nijmegen, Heijendaalseweg 135, 6525 ED Nijmegen, The Netherlands, Institute of Chemical Kinetics and Combustion, Institutskaya Street 3, NoVosibirsk 630090 Russia, NoVosibirsk State UniVersity, PirogoVa Street 2, NoVosibirsk 630090, Russia, and Institute of Semiconductor Physics, Academician LaVrentieV AVe. 13, NoVosibirsk 630090, Russia ReceiVed: July 17, 2009; ReVised Manuscript ReceiVed: September 24, 2009

Photodissociation of the (HI)2 van der Waals dimers at 248 nm and nearby wavelengths has been studied using time-of-flight mass spectrometry and velocity map imaging. I2+ product ions with a translational temperature of 130 K and “translationally hot” I+ ions with an average kinetic energy of Et ) 1.24 ( 0.03 eV and angular anisotropy β ) 1.92 ( 0.11 were detected as dimer-specific ionic photofragments. Velocity map images of the I2+ and I+ species were found to be qualitatively similar to those observed in the case of photoexcitation of the (CH3I)2 dimer (J. Chem. Phys. 2005, 122, 204301). As in the case of the (CH3I)2 dimer, the absence of neutral I2-specific features in the ionic species images from (HI)2 allows us to eliminate neutral molecular I2 as a precursor of I+ and I2+. Similar to the case of (CH3I)2, we deduce that the observed I2+ ions are produced in their 2Π3/2,g ground electronic state as a result of photodissociation of the ionized dimer (HI)2+ + hν f I2+ + .... The formation of “translationally hot” I+ ions is attributed to photodissociation of nascent vibrationally excited I2+ with an average vibrational energy of 1.05 ( 0.10 eV. This vibrational excitation is explained by the nonequilibrium initial I-I distance in I2+ arising in photodissociation of (HI)2+ after prompt release of the light H atoms. On the basis of our ab initio calculated value for the I-I distance of (3.17 Å) in the (HI)2+ precursor dimer, the vibrational excitation of I2+ is expected to be 1.02 eV, which is in quantitative agreement with our experimentally deduced value. The interpretation of our results was supported by ab initio calculations of the structure and energy of neutral and ionized dimers of HI at the MP4(SDTQ)//MP2 level. Introduction The photochemistry of small (RI)n van der Waals clusters and particularly (RI)2 dimers has been extensively studied during last two decades. Concerning the variation of R, special attention has been paid to R ) CH3 and H because of the large amount of reference data that is available concerning the photochemistry of the corresponding single molecules CH3I and HI. An overview of the literature concerning studies of CH3I and HI clusters is given in our recent paper.1 The mechanism and dynamics of photochemical processes in such clusters are known to deviate substantially from the photochemistry of the individual molecules starting at low values of n, and even for n ) 2. As known from the literature2 the UV photoexcitation of the isolated CH3I or HI molecules leads to direct dissociation, giving rise to iodine atoms in the 2P1/2 and 2P3/2 states (from here on denoted as I and I* atoms, where ∆E(I*-I) ) 7603.2 cm-1 ) 0.94265 eV), as well as CH3 and H fragments, respectively. In turn, the analogous excitation of the (CH3I)n and (HI)n clusters and particularly of (CH3I)2 and (HI)2 leads to the opening of new photochemical channels, as observed in the formation of a number of new photoproducts. A detailed history of the observation of those products is also described in our recent paper.1 The majority of such observations have been presented for (CH3I)n clusters, but there were also a number of papers †

Part of the “Benoıˆt Soep Festschrift”. Institute for Molecules and Materials, Radboud University Nijmegen. § Institute of Chemical Kinetics and Combustion. | Novosibirsk State University. ⊥ nstitute of Semiconductor Physics. ‡

where the changes in the system photochemistry were observed for clustered HI.3-8 Fan et al.3 observed the formation of neutral I2 following excitation of clustered HI at the dissociation wavelengths of 193 and 248 nm. I2 molecules were detected by means of laserinduced fluorescence spectroscopy and they were found to be in the X0g+ ground electronic state with very low internal (rovibrational) energy. In the same paper, Fan et al.3 studied the excitation of clusters of other molecules such as CH3I, C2H5I, i- and n-C3H7I, CF3I, CH3Br, and C2H5Br, and for all of them they detected the formation of I2 or Br2 in the ground electronic state. Young4 observed the formation of ionic molecular iodine, + I2 , in the time-of-flight mass spectrum arising after photoexcitation of clustered HI in the 230-255 nm wavelength range. He suggested that formation of I2+ is produced in the two-photon ionization of neutral I2, whose formation was earlier reported by Fan et al.3 Burnett and Young5 detected the formation of rotationally hot HI after the irradiation of clustered HI at the wavelength of 280-285 nm. Young performed kinetic energy resolved time-of-flight measurements of the photodissociation of clustered HI.6 In that study he observed slight changes in the kinetic energy distributions of H+ ions and large changes in the kinetic energy of I+ ions, following the excitation of clustered HI at wavelengths around 240 nm. The main new feature was a signal of I+ ions with 0.924 eV kinetic energy and a high angular anisotropy. Such a high kinetic energy for I atoms or I+ ions cannot result from the photodissociation of individual molecules HI. Young

10.1021/jp9067679  2010 American Chemical Society Published on Web 10/14/2009

3068

J. Phys. Chem. A, Vol. 114, No. 9, 2010

proposed that the I+ ions are formed as a result of three-photon dissociative ionization of neutral I2 produced in the photodissociation of clustered HI.3 Randall and Donaldson7 observed emission from an ion-pair state of I2 following the excitation of clustered HI at 248 nm. They suggested that I2 in the ion pair state might be produced in a two-stage process where the first stage is a four-center reaction (HI)2 f H2 + I2*, with I2* designating production of the electronically excited A, A′, or B states. In the second stage I2* absorbs another photon of 248 nm to reach the ion-pair state, from which the fluorescence is observed. Zhang et al.8 observed and characterized the changes in the kinetic energy of H atoms originating from the photoexcitation of clustered and not clustered HI at 266 nm using the Rydberg tagging technique. Summarizing the previous studies, we conclude that the most spectacular and distinctive products of HI cluster photochemistry, which are different from the photochemistry of individual HI molecules are, ionic4 and neutral3 molecular iodine, and atomic iodine ions with high kinetic energy and high recoil anisotropy.6 An important question is whether this striking photochemistry takes place in the simple HI dimer, (HI)2, or only in larger clusters, (HI)n, with n > 2. Exotic mechanisms such as four-center reactions are relevant in the photochemistry of (HI)2. In our previous studies1,9 we characterized in detail analogous product channels for (CH3I)n clusters arising as a result of photodissociation around 250 nm. In those papers we have shown that neutral I2 originates only from the photodissociation of (CH3I)n clusters with n > 29 and that the observed I2+ ions arising from the (CH3I)2 dimer are not the result of neutral I2 photoionization but appear as a result of the photochemistry of (CH3I)2+ dimer ions.1 Another conclusion from our CH3I cluster work was that I+ ions with high kinetic energy and high angular recoil anisotropy are produced as a result of the photodissociation of nascent I2+.1 In the present paper we perform a detailed study of the UV photodissociation of (HI)2 clusters at dissociation wavelengths around 250 nm using of the velocity map imaging technique. Our approach is identical to that used in our previous study of clusters (CH3I)2.1 For interpretation of the experimental data, quantum chemical calculations of the structure and energy of neutral and ionic (HI)2 have been also carried out. Experimental and Computational Details Time-of-Flight Experiments. The experimental setups used in the present study are described in detail elsewhere.1 Here a brief overview of the experimental conditions will be given. The setup used in Novosibirsk is a molecular beam apparatus combined with a time-of-flight mass spectrometer (TOF MS) in the Wiley-McLaren arrangement.10 A supersonic molecular beam was produced by expanding a gas mixture of 3% of HI and 97% of Ar at 1 bar in a home-built pulsed solenoid valve with a 0.23 mm nozzle. Each time before preparing the gas mixture the sample of HI was purified from H2 and I2 contamination. The gas jet passed through a 2.5 mm skimmer mounted 60 mm downstream and was irradiated by the perpendicularly propagating light of a pulsed KrF excimer laser in the extraction region of the TOF MS. The extraction electric field of the TOF MS was directed perpendicular to the molecular and laser beams. Nascent photoions were extracted from the photoexcitation region and accelerated in the TOF direction to a microchannel plate (MCP) detector. The MCP current (mass spectrum) was digitized, stored, and processed by a computer.

Vidma et al. The home-built pulsed KrF excimer laser (248 nm) had a maximum pulse energy of about 1 mJ and a pulse duration of about 5 ns. The light was linearly polarized in the direction parallel to the static electric field in the extracting region of the TOF MS where photoexcitation took place. The laser beam was focused by a 53 cm focal distance lens, and the energy of the laser pulse was monitored by a UV-sensitive photodiode, which was mounted behind the output window of the chamber. Quartz filters were used to attenuate the pulse energy. The lasers, valve, and detection equipment operated at a 1 Hz repetition rate. Velocity Map Imaging Experiments. The Nijmegen velocity map imaging setup has been described in detail elsewhere.1,11,12 A brief overview of the experimental conditions will be given here. A supersonic molecular beam was created by a pulsed solenoid valve (general valve) with a nozzle diameter of 0.20 mm. The valve was mounted in the on-axis configuration, generating a molecular beam directed along the TOF axis.11 After passing through a 2 mm skimmer mounted 20 mm downstream from the nozzle, the beam passes through a 2 mm hole in the repeller electrode and into the velocity map imaging lens where the beam was irradiated by a pulsed laser beam propagating perpendicular to the molecular beam. Product photoions were extracted from the photoexcitation region and projected onto the surface of a two-dimensional (2D) imaging detector, which was gated at the proper arrival time for mass selection. The voltage applied to the electrodes was adjusted in order to project all ions of the same velocity to the same point on the 2D detector, independent of their point of origin (velocity mapping conditions).11 The obtained images were inverted in order to reconstruct the photoproduct speed and angular distribution with using the BASEX inversion algorithm.13,14 Tunable UV radiation for photolysis in the range of 247-255 nm was obtained by frequency doubling of the output of Quanta Ray PDL-2 pulsed dye laser (Coumarin 500) pumped by the third harmonic (355 nm) of a Nd:YAG laser (Continuum Surelite) in a BBO crystal. The resulting pulse energy of UV light was about 1 mJ, and the pulse duration was about 5 ns. The light was linearly polarized in the direction parallel to the surface of the 2D detector and was focused by a 35 cm focal length lens. Photodissociation of I2 molecules at 248.6 nm was performed for kinetic energy calibration of the images. The detected product ring corresponding to the second dissociation limit of I2 giving I and I* atoms was used. The kinetic energy TI atom of each of I atom produced in this process must be

TIatom )

(hV248.6nm - D0*) ) 0.6256 eV 2

(1)

In this formula hV248.6nm ) 4.9873 eV is the energy of the 248.6 nm photon, D0* ) 2.485042 eV is the second dissociation limit of I2.15 In the calibration experiments I atoms were ionized by the radiation of another laser tuned to 303.68 nm, resonant for (2 + 1) REMPI of I atoms.12 A gas mixture of HI (3%) and Ar (97%) for the supersonic expansion was prepared in a stainless steel bottle. The backing pressure used in the experiments was 1.5 bar. Two modes of pulsed valve operation were used in order to provide different conditions for cluster formation. Conditions unfavorable for clustering were provided by a short gas pulse, while favorable clustering conditions were provided with a longer gas pulse. The lasers, valve, and detection equipment operated at a 10 Hz repetition rate.

Photodissociation of (HI)2

J. Phys. Chem. A, Vol. 114, No. 9, 2010 3069

Figure 2. Dependence of the integral of the I+ and I2+ TOF peaks and their sum signal (I2+ + I+) on the energy of the 248.6 nm laser pulse, shown on a double logarithmic scale. The solid line indicates the linear fit of the total ionic signal (I2+ + I+) in the region before the saturation. The slope of the line is 1.93 ( 0.16. Figure 1. Calculated structures and energies of two stable isomers of neutral (HI)2. The plane of Cs symmetry coincides with the plane of the figure.

Results

Computational Details. For evaluation of intermolecular interactions in the van der Waals dimer of HI, an approach similar to that used for (CH3I)2 dimers in ref 16 has been applied. An expanded basis set with the addition of polarization and diffuse functions has been used. For the hydrogen atom the augcc-pVTZ basis functions suggested by Dunning17 have been used. For iodine atoms SDB-aug-cc-pVTZ basis functions elaborated by Martin and Sunderman18 have been applied. These last functions in conjunction with a relativistic core polarization potential, RECP, suggested by Bergner, Dolg et al.19 have been taken from their online version.20 The HI monomer geometry was optimized, and its vibrational energy was calculated at the MP2 level. This geometry of the monomer was then used to reveal the most stable relative configurations of HI molecules in the (HI)2 dimer. The two most stable configurations were then fully optimized at the MP2 level. The dimer vibrational energy has been calculated for all isomers with the use of a scaling factor (0.961), also used to fit the calculated value with the experimental one for the HI monomer. The same approach has been used for the calculations of the vertical and adiabatic ionization potentials of the ionic (HI)2+ dimer in its two most stable configurations. The binding energy (Ebind) values for two most stable dimer structures were calculated at the MP4(SDTQ)//MP2 level with the same basis set. Ebind was corrected by taking into account basis set superposition error (BSSE) by the counterpoise method21 and zero-point energy (ZPE), calculated within MP2 approach for optimized molecular structures. All the calculations have been carried out with Gaussian 03 package22 on the SGI Origin 3800 1024 node system of the SARA Computing and Networking Services in Amsterdam.

Calculated Structure of Neutral Dimers. Two most stable structures of the (HI)2 dimer found by ab initio calculations are shown in Figure 1 together with their calculated binding energy. The structure of this dimer generated in a supersonic jet has been experimentally studied by McIntosh et al.23 using rovibrationally resolved near-infrared spectroscopy at ∼4.5 µm. From their study the rotational constant of the ground state of the dimer was found to be equal to B0 ) 0.378438 (50) GHz. This value is close to the lowest two rotational constants (0.396 and 0.395 GHz) for the head-to-tail (ht) structure (Table 1) of the HI dimer. We thus consider the ht configuration to be dominant in our experiment as well. Experimental Results. The time-of-flight mass spectrum following 248.6 nm photoexcitation of clustered HI is qualitatively similar with the mass spectrum obtained in the experiments with clustered CH3I (Figure 1 of the ref 1) and shows two main features: I2+ and I+ ions. The dependences of the integrated I+ and I2+ peaks on the laser pulse energy are shown in Figure 2. They also look similar with the analogous dependencies obtained in the experiments with clustered CH3I (Figure 3 of the ref 1). At lower laser powers both ionic signals increase, but at higher laser powers the I2+ signal decreases due to photodissociation, while the I+ signal continues to rise. Photofragmentation of each I2+ ion gives one I+ ion; therefore for correct extraction of the power dependence of production of I2+ on the laser pulse energy we used the sum of the I+ and I2+ signals. The order of power dependence was found to be 1.93 ( 0.16. Young and coworkers4 found a higher power dependence for dissociation at 240 and 238 nm (and more extreme clustering conditions). Partial saturation of our signal is possible. A raw image of I2+ ions resulting from the photoexcitation of clustered HI at the wavelength of 248.6 is presented in the

TABLE 1: Parameters of ab Initio Calculated Geometry, Energy and Ionization Potentials of the (HI)2 Dimer ionization potentials binding energy, cm-1

rotational constants, GHz

adiabatic,eV

vertical, eV

226

hh-(HI)2 “Head to Head” Configuration, CS Symmetry 204.335; 0.506; 0.505. 9.31

10.26

224

ht-(HI)2 “Head to Tail” Configuration, CS Symmetry 202.412; 0.396; 0.395. 9.81

10.20

3070

J. Phys. Chem. A, Vol. 114, No. 9, 2010

Figure 3. Speed distribution of I2+ ions resulting from photodissociation of clustered HI at 248.6 nm. The dashed line represents the speed distribution extracted from the raw I2+ image shown in the inset. The solid line represents a fit of the experimental profile with a Maxwell curve. The best-fit provides a value for the I2+ translational temperature equal to 130 K. The double headed arrow next to the image represents the direction of the laser polarization. Darker areas correspond to higher signal levels.

Vidma et al.

Figure 5. Speed distribution of I+ ions arising from the photofragmentation of HI under conditions favoring clustering. Thick gray line is the I+ speed distribution extracted from the image shown in Figure 4a. The fit of this curve by the sum of five Gaussians is shown by the black solid line. Each of the five individual Gaussians is shown by a dotted line.

of two different channels. We number these channels as 3a and 3b, having in mind, however, that these two channels are not distinguishable by eye when looking at the raw image. The results of the fitting are represented in Figure 5, and the parameters for the best fit Gaussian curves are represented in Table 2. The angular distribution for each channel has been fit with the following function Figure 4. Raw I+ ion images obtained at 248.6 nm under conditions (a) favorable and (b) unfavorable for clustering of HI. Image (b) is enlarged by a factor of 4 in comparison with image (a). Image (a) is shown with a high sensitivity gray scale and (b) is shown in a lower sensitivity gray scale. Channels 1 and 2 are also present in image (a) but they are not visible because of the saturation of the intensity scale. Channels 3 and 4 are not visible in image (b) even at the highest sensitivity scale. Laser polarization is shown by the double-headed arrow.

inset of Figure 3. The angular distribution extracted from the inversion of this image was found to be isotropic and the speed distribution is represented in Figure 3. A fit of this distribution with a Maxwell curve is also presented in Figure 3. The best fit provided a value of 130 K for the I2+ translational temperature. In Figure 4 raw I+ images resulting from the photoexcitation of HI at 248.6 nm for conditions favorable (a) and unfavorable (b) for clustering are shown. The image obtained under conditions unfavorable for clustering shows two small rings: each ring corresponds to separate I+ formation channels. These are denoted as channels 1 and 2 and indicated by arrows in Figure 4b. The image obtained under conditions favorable for clustering also shows channels 1 and 2 (they are not visible in Figure 4a due to the saturation of the brightness scale around the central part of the image), as well as two new channels: a “blob” slightly stretched in the vertical direction (channel 3) and a large, slightly blurry ring with positive anisotropy (channel 4). Figure 5 represents the speed distribution extracted from the I+ ion image shown in Figure 4a. This distribution contains four peaks; each peak corresponds to one of the channels discussed above. The total distribution has been fit with the sum of five Gaussian curves: each of the channels 1, 2, and 4 was fit by one Gaussian and channel 3 was fit with the sum of two Gaussian curves. Two Gaussians provided a much better fit to the peak corresponding to channel 3 than one Gaussian. This might indicate that channel 3 is not a single channel but consists

(

( 23 cos θ- 21 ))

f)A1+β

2

(2)

the general expression for the angular distribution of fragment recoil following one-photon dissociation.24 In eq 2, θ is the angle between the photofragment recoil velocity and the direction of polarization of the excitation light, and β is the anisotropy parameter. The value of β for each channel is presented in Table 2 as well. All four channels show the same β dependence when the wavelength was scanned in the range of 248-250 nm. This indicates their nonresonant nature. Figure 6 represents the results of the calibration experiment on the comparative photodissociation of neat I2 and clustered HI at 249.61 nm, the wavelength for (2 + 1) REMPI of I(2P3/ 2). Figure 6a represents the image resulting from the photoexcitation of I2 (0.2 Torr) seeded in argon (1 bar), and Figure 6b represents the image resulting from the photoexcitation of clusters of HI (HI (3%) seeded in argon (1 bar)). The goal of this experiment was to examine processes that would occur with neutral I2 (should it be produced following photodissociation of clustered HI) and to compare them with the results of the actual excitation of clustered HI. The main signal in the image depicted in Figure 6a is a large, perpendicular ring which arises from one-photon dissociation of I2 to the second dissociation limit, giving I and I* atoms. The image resulting from the photoexcitation of clustered HI did not show any traces of any signal of this type. We use cold I2 as a precursor, in accord with the results of Fan et al.3 Randall and Donaldson7 observed vibrationally and even electronically excited I2 in their study of neat HI expansions, where very large clusters and even liquid droplets should be present. We observe that neutral I2 does not make any significant contribution to the signal of I+ resulting from the photoexcitation of small HI clusters.

Photodissociation of (HI)2

J. Phys. Chem. A, Vol. 114, No. 9, 2010 3071

TABLE 2: Parameters of Five Gaussian Curves That Provide the Best Fit of the Speed Distribution of I+ Ions Arising under Conditions Favorable for Clusteringa no. of channels 1 2 3a 3b 4 a

peak integral (% from the total)

υc, m/s

ω, m/s

I+ kinetic energy corresponding to the peak center, eV

anisotropy parameter, β

14 7 16 44 19

48 ( 4 111 ( 4 222 ( 5 429 ( 5 1372 ( 15

12 ( 4 22 ( 6 83 ( 4 218 ( 16 168 ( 9

0.0015 ( 0.0003 0.0081 ( 0.0006 0.032 ( 0.002 0.121 ( 0.003 1.24 ( 0.03

0.16 ( 0.02 1.67 ( 0.06 0.69 ( 0.08 0.92 ( 0.05 1.92 ( 0.11

Anisotropy parameters measured for each of five channels are also represented. Each Gaussian curve is expressed by the formula

G(υ) ) A exp(-(υ - υc)2 /2ω2)

Figure 6. (a) I+ image resulting from the photoexcitation of I2 at 249.61 nm, corresponding to the (2 + 1) REMPI wavelength of I(2P3/2). Ring A corresponds to the process: I2 + hV f I + I* with subsequent resonant ionization of I(2P3/2) atoms, and ring B corresponds to the two-photon ionization of I2 with subsequent photodissociation of product parent ions: I2+(2Π3/2) + hV f I+ + I (see Figure 10 in ref 1 for details). (b) Image of I+ ions resulting from the photodissociation of clustered HI at 249.61 nm.

Discussion Comparison with Results of Experiments on CH3I Clusters. First of all, we compare our results on HI obtained in the present study with the results of the analogous experiments on the photoexcitation of clustered CH3I at the same wavelength.1 Images of I2+ from the photoexcitation of clustered HI (inset of Figure 3) and from clustered CH3I (Figure 5 in ref 1) look qualitatively similar. Both images show an isotropic angular distribution. Photoexcitation of clustered HI results in I2+ with a lower translational temperature (130 K) compared with the photoexcitation of clustered CH3I (650 K). The I+ images originating from clustered HI (Figure 4a) look qualitatively similar to I+ images from the photoexcitation of clustered CH3I (Figure 6b of the ref 1). Both of those images contain two features due to cluster photochemistry: a blurred “blob” and a large ring with a positive anisotropy (channels 3 and 4 in the present study). In the case of clustered CH3I, the blob had a maximum of intensity at the kinetic energy of 0.17 eV and the large ring had 0.94 eV kinetic energy and an anisotropy parameter β of 1.55. As in paper,1 the ions giving rise to the outer ring are named “translationally hot” I+ ions. The dependencies of the intensities of the signals on the laser pulse energy obtained in the present study (Figure 2) are similar with those obtained in the experiments with clustered CH3I (Figure 3 in ref 1). Size of the (HI)n Cluster. In our study of CH3I clusters1 we paid special attention to the determination of the size of the (CH3I)n clusters that gave rise to our observed signals. On the basis of the results of our experiments as well as on the literature data, we concluded that under the expansion conditions used in many previous experiments the formation of quite large clusters is possible. We also proved that with our very low CH3I

concentrations, the origins of our observed I2+ and “translationally hot” I+ signals were from (CH3I)2 dimers. These signals remain qualitatively the same even at expansion conditions with very low concentrations of CH3I which do not favor formation of clusters higher than dimers. In the present HI cluster study (again in this case, with low HI concentrations) our signals of I2+ and I+ show a very similar behavior as in the experiments with clustered CH3I. Therefore, we can suggest that the mechanisms of the formation of those products are similar with the mechanisms deduced in the experiments with CH3I clusters. We assume, thus, that the origin of the presently observed ionic products is from the photodissociation of (HI)2 dimers. Mechanism for the Formation of I2+ Ions. Photoexcitation of I2 and Clustered HI at 249.61 nm. The results of the experiment on the comparative photoexcitation of clustered HI and pure I2 at 249.61 nm, represented in Figure 6, are the same as the results of the analogous experiments with (CH3I)2.1 As extensively discussed in ref 1, the absence of I2-specific features shown in Figure 6a in the image of ionic species resulting from the photoexcitation of (HI)2 shown in Figure 6b allow us to neglect molecular I2 as a precursor of I+ and I2+. Photodissociation of (HI)2+ as the Precursor of I2+. In analogy with our study of clustered CH3I,1 we suppose that the precursor of I2+ in the present study is the ionized dimer, (HI)2+. In ref 1 we concluded that I2+ arises from the photodissociation of (CH3I)2+, which is produced by two-photon ionization of (CH3I)2. The adiabatic IP of (HI)2 calculated in this work and presented in Table 1 has the values 9.31 and 9.81 eV, and the calculated vertical IP has the values 10.26 and 10.20 for the hh and ht geometries of (HI)2, respectively. These values allow us to expect that the energy sum of two 248.6 nm photons (9.97 eV) can cause ionization of the (HI)2 dimer in either configuration. As we discussed above, the dominant configuration of neutral dimer is ht. After “vertical” two-quantum ionization of ht-(HI)2 isomer we expect formation of ht-(HI)2+. We can reasonably assume that further rearrangement takes place as ht-(HI)2+ f hh-(HI)2+ on the time scale of laser pulse because the hh geometry of ion is more stable by 0.5 eV as compared with ht-(HI)2+. This value is equal to the difference in the calculated adiabatic IPs of these configurations. This allows us to expect the hh-(HI)2+ form to be a precursor of further photodissociation giving rise to I2+. In our experiments with clustered CH3I we deduced that the most probable process of formation of I2+ from the ionized dimer (CH3I)2+ is photodissociation to form two CH3 fragments and I2+. On the basis of the similarities of the properties of ions I2+ obtained in the studies of (CH3I)21 and (HI)2 (present study), we conclude that the mechanism of formation of I2+ in the present study is the same.

3072

J. Phys. Chem. A, Vol. 114, No. 9, 2010

The power dependencies of the production of I2+ on the laser power energy gives values close to 2 for both clustered HI and CH3I. The I2+ recoil angular distribution was found to be isotropic in both cases. The fact that translational temperature of I2+ in the case of clustered HI is lower than that for clustered CH3I can be qualitatively explained by momentum conservation; I2+ “pushes off” from much heavier methyl radical(s) for (CH3I)2 compared to light H atom(s) for (HI)2. In the experiments with clustered CH3I, the I2+ was found to be produced in the ground electronic (2Π3/2,g) state.1 Due to the observed similarity of the mechanisms in the present study, we expect that I2+ from (HI)2 is also produced in the 2Π3/2,g state. Mechanism of the Formation of I+ Ions. Channels 1 and 2. Channels 1 and 2 were observed in experiments with clustered as well as unclustered HI and can thus be attributed to the photochemistry of individual HI molecules. The images for these I+ formation channels are similar at both 248.60 and 249.61 nm photoexcitation, where 249.61 nm is a (2 + 1) REMPI wavelength for I* while 248.60 nm is a REMPI wavelength neither for I nor for I*. We can thus rule out one-photon dissociation of HI in the A-band continuum25,26 followed by REMPI of the nascent I atoms. For an alternative explanation, we have previously reported27 a velocity map imaging study of the photodissociation of HI by three-photon excitation at 238.66 nm. Similar to the results of that study, at 248.60 nm threephoton excitation of HI most likely occurs with the formation of HI+ in the predissociative A(2∑+1/2) electronic state and a low energy electron. According to the Franck-Condon principle the HI+ ion must be produced with a H-I distance about the same as in the ground state of neutral HI (1.609 Å28). The potential energy of the state A (2∑+1/2) at this H-I distance is close to the dissociation limit for the H(2S) + I+(3P1,0) products (channel 1) and H(2S) + I+(3P2), channel 2.29 Channel 4. The I+ formation channels 3a, 3b, and 4 appear only under conditions favorable for clustering; they can therefore be attributed to the photochemistry of clustered HI. Observations of I+ ions with energy and anisotropy similar to those of channel 4 have been reported in the literature for the (CH3I)21 and (HI)26 dimers. Young observed I+ with 0.924 eV kinetic energy and an anisotropy parameter of β g 1.5 ( 0.2 in experiments on photoexcitation of clustered HI at 240.02 nm.6 In our recent paper1 we observed the formation of “translationally hot” ions I+ with 0.94 eV kinetic energy and β ) 1.55 in experiments on the photoexcitation of clustered CH3I at 248.6 nm (outer ring in Figure 6b in ref 1). We suggest that the origin of channel 4 is one-photon dissociation of I2+sthe same mechanism as that proposed for the formation of “translationally hot” ions I+ in the photodissociation of CH3I clusters.1 The maximum of channel 4 corresponds to 1.24 ( 0.03 eV kinetic energy, corresponding to a total kinetic energy release (TKER) of 2.48 ( 0.06 eV under the assumption of I2+ dissociation. By analogy with the experiments on (CH3I)2 dimers, we expect that I2+ must be produced in the ground electronic state (2Π3/2,g). In order to identify the pathways giving rise to channel 4, we refer to the results of the experiment on the photodissociation of I2+ at 248.6 nm, which are represented in Figure 8 of ref 1. In that experiment we found only one ring that corresponds to the photodissociation of I2+ (2Π3/2,g). This ring was assigned as a result of photodissociation of I2+ (2Π3/2,g) to the one or several of the close-lying dissociation limits: [I+(3P1,0) + I(2P3/2)] and [I+(3P2) + I(2P1/2)]. These three channels correspond to a total kinetic energy release within the interval of 1.43 (

Vidma et al.

Figure 7. Calculated structure of two stable isomers of the ionic (HI)2+ dimer.

0.07 eV, under the assumption that we start with vibrationally unexcited I2+(2Π3/2,g). We suppose that one or several of these processes give rise to the channel 4 in the present experiment with clustered HI. We attribute the difference (1.05 ( 0.10 eV) between the observed value of TKER (2.48 ( 0.06 eV) and the value (1.43 ( 0.07 eV) expected for photodissociation of vibrationally unexcited I2+ (2Π3/2,g) is due to the vibrational excitation of I2+ produced from clusters. Vibrational Excitation of I2+. The value of 1.05 ( 0.10 eV vibrational excitation is in good agreement with the proposed mechanism of the formation of I2+ as a result of the photodissociation of ionized dimer (HI)2+. Because of the large difference in the masses of H and I, we can assume that in the photodissociation of (HI)2+, the light hydrogen atoms fly away rapidly leaving the heavy I atoms at the same positions as in the precursor ionized dimer. The calculated structures of the ionized dimer in the hh and ht configurations are shown in Figure 7. As discussed above, ionization of the neutral HI dimer should finally produce the hh configuration of (HI)2+. The calculated equilibrium distance between the iodine atoms in the hh-configuration has a value of 3.172 Å. Since this value is substantially longer than the equilibrium I-I distance in the ground state of I2+ (2Π3/2,g) (2.57 Å30), the ions must be produced with vibrational excitation. We calculated the value of this vibrational excitation using the Morse potential energy curve constructed for the 2Π3/2,g state of the I2+. This value was found to be of 1.02 eV, which overlaps within the experimental error with the value 1.05 ( 0.10 eV measured in the experiment. This extraction procedure is illustrated in Figure 8. Channel 3. Channels 3a and 3b correspond to I+ ions with 0.03 and 0.12 eV kinetic energy, respectively. Photodissociation of I2+ around 248 nm cannot provide ions I+ with such a small kinetic energy (Figure 9 in ref 1). The images of these channels do not have a well-pronounced shape, and therefore it is hard to establish unambiguously the mechanism of their formation. We can suggest two possible mechanisms: The first is the dissociative ionization of translationally hot molecular HI with the formation of I+ via channels 1 and 2. The translationally hot HI in this mechanism could be produced in the photodissociation of clustered HI. Another possible mechanism is the formation of I+ in the direct photodissociation of the (HI)2+ dimer. Conclusion The photodynamics governing formation of I+ and I2+ in the photofragmentation of the HI dimer, (HI)2, have been studied. Speed and angular distributions of nascent I+ and I2+ were determined by means of velocity map imaging. The images

Photodissociation of (HI)2

J. Phys. Chem. A, Vol. 114, No. 9, 2010 3073 as well as support by Russian Foundation of Basic Research (Grant No. 06-03-32542). The computing time on the SGI Origin 3800 system of SARA Computing and Networking Services in Amsterdam was provided by the Dutch National Computer Facility NCF. References and Notes

Figure 8. Scheme illustrating the possible source of vibrational excitation of I2+ arising from photofragmentation of the (HI)2+ dimer. The thin solid line represents a Morse potential energy curve constructed for the ground 2Π3/2,g state of the I2+ ion (see details in the text). The I-I distance coordinate for (HI)2+ at 3.17 Å in its ground head-tohead configuration and the potential energy of I2+ (8230 cm-1 ) 1.02 eV) corresponding to this distance are shown by projections on the axis. The Morse curve has been constructed with parameters values for the 2Π3/2,g state of I2+ of ωe ) 240 cm-1 and xeωe ) 0.77 cm-1, re ) 2.57 from ref 30, and De ) 21812 cm-1 is calculated using the values of IP I2+(2Π3/2,g) ) 75069 cm-1,31 IP of the iodine atom ) 84295.1 cm-1,32 and dissociation energy of I2 ) 12440.24 cm-1.15

obtained were qualitatively similar with those observed in experiments on (CH3I)2 (ref 1). Comparison of the velocity map images of I+ ions arising from I2 and HI clusters allowed us to neglect I2 as a precursor of ionic species arising from clusters. By analogy with the conclusions made in the (CH3I)2 study, we conclude here that I2+ is produced as a result of the photodissociation of (HI)2+ and that the “hot” I+ ions we observe are produced as a result of the photodissociation of I2+. The velocity distribution of I2+ was found to be isotropic and well fitted with a Maxwell distribution with a translational temperature of 130 K. The I+ image contains contributions from two channels attributed to the photochemistry of clusters. One of them provides I+ ions with an average kinetic energy of 1.24 ( 0.03 eV and anisotropy of β ) 1.92 ( 0.11. This channel was attributed to the photodissociation of I2+. The value of the kinetic energy of “hot” I+ allowed us to deduce that I2+ is produced in the 2Π3/2,g ground electronic state with an average vibrational energy of 1.05 ( 0.10 eV. The vibrational excitation is explained by the nonequilibrium initial I-I distance in I2+ arising in photodissociation of (HI)2+ after release of the light H atoms. The interpretation of our results was supported by ab initio calculations of the energy and structure of the neutral (HI)2 and ionized dimers (HI)2+. For neutral dimer (HI)2 two stable isomers were foundshead-to-head and head-to-tail isomers. The structure of the head-to-tail isomer corresponds to the rotational constants experimentally determined for the (HI)2 dimer by McIntosh et al. (ref 23). Photofragmentation of (HI)2 is concluded to be similar to that of (CH3I)2. The quantitative difference in energy distribution of the dimer-specific ionic features for these two cases is explained as being due to the difference in mass of the atoms involved. Acknowledgment. We gratefully acknowledge the financial support from Netherlands Organization for Scientific Research (NWO) under the programs Molecular Atmospheric Physics (MAP-09), NWO-CW, ECHO project number 700.55.025, and NWORussia-NetherlandsCooperativeResearchGrant047.009.001

(1) Vidma, K. V.; Baklanov, A. V.; Khvorostov, E. B.; Ishchenko, V. N.; Kochubei, S. A.; Eppink, A. T. J. B.; Chestakov, D. A.; Parker, D. H. J. Chem. Phys. 2005, 122, 204301. (2) Okabe, H. Photochemistry of Small Molecules; Wiley-Interscience: New York, 1978. (3) Fan, Y. B.; Randall, K. L.; Donaldson, D. J. J. Chem. Phys. 1993, 98, 4700. (4) Young, M. A. J. Phys. Chem. 1994, 98, 7790. (5) Burnett, J.; Young, M. A. Chem. Phys. Lett. 1994, 228, 403. (6) Young, M. A. J. Chem. Phys. 1995, 102, 7925. (7) Randall, K. L.; Donaldson, D. J. J. Phys. Chem. 1995, 99, 6763. (8) Zhang, J.; Dulligan, M.; Segall, J.; Wen, Y.; Wittig, C. J. Phys. Chem. 1995, 99, 13680. (9) Vidma, K. V.; Baklanov, A. V.; Zhang, Y.; Parker, D. H. J. Chem. Phys. 2006, 125, 133303. (10) Whiley, W. C.; McLaren, I. H. ReV. Sci. Instrum. 1955, 26, 1150. (11) Eppink, A. T. J. B.; Parker, D. H. ReV. Sci. Instrum. 1997, 68, 3477. (12) Eppink, A. T. J. B.; Parker, D. H. J. Chem. Phys. 1998, 109, 4758. (13) Dribinski, V.; Ossadtchi, A.; Mandelshtam, V. A.; Reisler, H. ReV. Sci. Instrum. 2002, 73, 2634. (14) Imaging in Molecular Dynamics: Technology and Applications; Whitaker, B. J., Ed.; Cambridge University Press: New York, 2003. (15) Davies, Mark R.; Shelley, John C.; Le RoyRobert J., Robert J. J. Chem. Phys. 1991, 94, 3479. (16) Bogdanchikov, G. A.; Baklanov, A. V.; Parker, D. H. Chem. Phys. Lett. 2003, 376, 395. (17) Dunning, T. N., Jr. J. Chem. Phys. 1989, 90, 1007. (18) Martin, J. M. L.; Sunderman, A. J. Chem. Phys. 2001, 114, 3408. (19) Bergner, A.; Dolg, M.; Kuechle, W.; Stoll, H.; Preuss, H. Mol. Phys. 1993, 80, 1431. (20) EMSL Basis Set Library: http:/www.emsl.pnl.gov:2080/forms/ basisform.html/. (21) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553. (22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li X.; Knox, J. E.; Hratchian H. P.; Cross, J. B.; Bakken V.; Adamo, C.; Jaramillo J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y., Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, ReVision C.02; Gaussian, Inc.: Wallingford, CT, 2004. (23) McIntosh, A. L.; Wang, Z.; Lucchese, R. R.; Bevan, J. W. Chem. Phys. Lett. 2000, 328, 153–159. (24) Zare, R. N.; Herschbach, D. R. Proc. IEEE 1963, 51, 173. (25) Gendron, D. J.; Hepburn, J. W. J. Chem. Phys. 1998, 109, 7205. (26) Manzhos, S.; Loock, H.-P.; Bakker, B. L. G.; Parker, D. H. J. Chem. Phys. 2002, 117, 9347. (27) Loock, H.-P.; Bakker, B. L.G.; Parker, D. H. Can. J. Phys. 2001, 79, 211. (28) Molecular Spectra and Molecular Structure. IV. Constants of Diatomic Molecules; Huber, K. P., Herzberg, G., Eds.; van Nostrand: New York, 1979. (29) Cormack, A. J.; Yencha, A. J.; Donovan, R. J.; Lawley, K. P.; Hopkirk, A.; King, G. C. Chem. Phys. 1997, 221, 175. (30) Mason, S. M.; Tuckett, R. P. Chem. Phys. Lett. 1989, 160, 575. (31) Cockett, M. C. R.; Goode, J. G.; Lawley, K. P.; Donovan, R. J. J. Chem. Phys. 1995, 102, 5226. (32) Ionization potentials of atoms and atomic ions. In Handbook of Chemistry and Physics; Lide, R., Ed.; CRC Press: Boca Ratan, FL, 1992; pp 10-211. Moore C. E. Atomic Energy Levels. National Standard Reference Data Series, National Bureau of Standards Circular 35; U.S. Government Printing Office: Washington, DC, 1971, Vol. III.

JP9067679