Dynamics of Hydrogen and Methyl Radical Loss ... - ACS Publications

Article pubs.acs.org/JPCA

Dynamics of Hydrogen and Methyl Radical Loss from Ionized Dihydro-Polycyclic Aromatic Hydrocarbons: A Tandem Mass Spectrometry and Imaging Photoelectron−Photoion Coincidence (iPEPICO) Study of Dihydronaphthalene and Dihydrophenanthrene Brandi West,† Christine Joblin,‡,§ Valerie Blanchet,∥ Andras Bodi,⊥ Bálint Sztáray,# and Paul M Mayer*,† †

Chemistry Department, University of Ottawa, Ottawa, Canada K1N 6N5 Université de Toulouse, UPS-OMP, IRAP, F-31028 Toulouse cedex 4, France § CNRS, IRAP, 9 Av. colonel Roche, BP 44346, F-31028 Toulouse cedex 4, France ∥ Laboratoire des Collisions Agrégats Réactivité, Université Toulouse-CNRS, 31000 Toulouse, France ⊥ Molecular Dynamics Group, Paul Scherrer Institut, Villigen 5232, Switzerland # Chemistry Department, University of the Pacific, Stockton, California 95211, United States ‡

S Supporting Information *

ABSTRACT: Ionized 1,2-dihydronaphthalene (C10H10+) and 9,10-dihydrophenanthrene (C14H12+) are homologous dihydrogenated polycyclic aromatic hydrocarbons containing adjacent sp3 carbon sites. Tandem mass spectrometry involving kiloelectronvolt collision induced dissociation was employed to aid in the structural characterization of the products of the main dissociation channels, loss of H (and subsequent H and H2 losses in dihydronaphthalene) and CH3. Evident from both the CID mass spectra and the imaging photoelectron− photoion coincidence (iPEPICO) breakdown curves is the fact that there are two competitive routes to the loss of H. For 1,2dihydronaphthalene these give activation energies of 2.22 ± 0.10 and 2.44 ± 0.05 eV, whereas only 2.37 ± 0.12 eV was obtained for 9,10-dihydrophenanthrene. The two parallel H-loss chaneels are believed to be the result of isomerization taking place to the methylindene ion and the 9-methylfluorene ion for 1,2-dihydronaphthalene and 9,10-dihydrophenanthren, respectively. Each newly formed isomer dissociates by H loss (one of the two competing H-loss reactions) and, of course, methyl loss. Methyl radical loss has similar kinetics for the two systems, E0 = 2.57 ± 0.12 eV, Δ‡S = 18 ± 11 J K−1 mol−1 for ionized dihydronaphthalene and E0 = 2.38 ± 0.15 eV, Δ‡S = −3 ± 15 J K−1 mol−1 for ionized dihydrophenanthrene, but as can be seen, the E0 and Δ‡S are slightly lower for the latter. The final bond rupture step in both H and CH3 loss is expected to be accompanied by a positive Δ‡S, thus the low energy H loss and CH3 loss originate from the isomer ion in both cases, with the entropy of activation being dominated by the isomerization step. In contrast, sp3-H loss from the dihydro-PAHs differ by little in both systems (E0 = 2.44 eV in ionized dihydronaphthalene and 2.37 eV in ionized dihydrophenanthrene and the Δ‡S values are 27 and 18 J K−1 mol−1, respectively). The presence of a second sp3 carbon site has decreased the C−H bond dissociation energy relative to protonated naphthalene and protonated phenanthrene, possibly to facilitate the restoration of the unaltered PAH ion. The calculated dihedral angle is −34.3° in C10H10+• whereas C14H12+• has an angle of −49.6°, indicating that to restore the planar nature of the molecules, which is required for all reaction channels investigated, there is more rearrangement needed for 9,10dihydrophenanthrene. Energetics and entropic values associated with H and H2 loss from [M − H]+ ions from ionized dihydronaphthalene were determined to be 2.72 eV, 9 ± 17 J K−1 mol−1, and 2.85 eV, 9 ± 7 J K−1 mol−1, respectively.

1. INTRODUCTION

only PAHs but also modified PAHs, such as hydrogenated species. Our group has previously studied the energetics of the naphthalene cation.12 It is the simplest example of a PAH molecule and as such has been studied quite extensively.13−15 It is for this reason that naphthalene would be the base structure of the first hydrogenated molecule studied. Hirama et al. determined that there are three sites equally likely to adsorb the

Molecular H2 is the most prolific molecular species found in the interstellar medium (ISM).1 Whereas it has been long recognized that H2 is formed on dust grains,2 investigation of the involved mechanisms is still a very active subject of research.3,4 One popular theory is that polycyclic aromatic hydrocarbons (PAH) may play a role in the formation of H2 by acting as catalysts.5 Atomic hydrogen would adsorb onto the surface of the molecules, thus hydrogenating the PAHs, prior to the desorption of H2.6−910,11 This reaction channel demonstrates the importance of investigating the energetics of not © 2014 American Chemical Society

Received: January 14, 2014 Revised: February 12, 2014 Published: February 12, 2014 1807

dx.doi.org/10.1021/jp500430g | J. Phys. Chem. A 2014, 118, 1807−1816

The Journal of Physical Chemistry A

Article

first superhydrogen atom; this represents the only three sites present on the naphthalene structure as seen in Figure 1a. Of all

2. EXPERIMENTAL METHODS 1,2-Dihydronaphthalene and 9,10-dihydrophenanthrene were obtained from Sigma-Aldrich (Sigma-Aldrich, Oakville, Canada) and used without further purification. The 1,2-dihydronaphthalene sample was introduced into the modified VG ZAB mass spectrometer via a liquid inlet, whereas the 9,10dihydrophenanthrene was introduced using a solid probe at ambient temperature. Both were introduced into the iPEPICO apparatus after pumping to remove air. MIKES, CID, and MI-CID Experiments. All MIKES, CID, and MI-CID experiments were performed on a modified VGZAB three sector instrument.17 The geometry consists of a magnetic sector, followed by two electrostatic analyzers (BEE geometry) (VG Analytical, Manchester, U.K.) The full details of the experiment have been explained previously.12 In short, the sample is introduced at a pressure ranging from 10−5 to 10−4 mbar, is ionized through electron ionization and accelerated to 6−8 kV. The ion of interest is then mass selected in the magnetic sector and directed toward the first electrostatic analyzer (ESA). In the case of a CID experiment helium gas is introduced into the field free region under single collision conditions (5 × 10−7 mbar.) to induce fragmentation, whereas for the other experiments (MIKES, MI-CID) the ion travels through unhindered. For CID and MIKES experiments, the first ESA scans to detect all fragments. For the MI-CID experiments, the first ESA is fixed on a specific energy to select one ion to travel toward the next field free region. Helium collision gas is introduced to induce fragmentation; these fragments are then detected by the second ESA. Spectra were recorded using ZABCAT program which was developed by Mommers Technologies Inc. (Ottawa, Canada).18 iPEPICO. All iPEPICO experiments were performed on the VUV beamline at the Swiss Light Source (Paul Scherrer Institut, Villigen, Switzerland). The experimental setup and a detailed description of the experiment are presented elsewhere.19−21 Briefly, monochromatic VUV synchrotron radiation is used as a photoionization source to ionize gas molecules. Ions are directed toward a time-of-flight mass spectrometer (TOF) while the ejected electrons are extracted in the opposite direction toward an imaging multichannel plate (MCP) detector with each event time and position stamped. Threshold electrons are focused onto the center of the MCP, and kinetic energy electrons detected in a small ring region around this center spot give a good representation of the hot electron background of the threshold signal. The mass spectrum corresponding to this ring is subtracted from the center TOF distribution to obtain the threshold ionization mass spectra. The time-of-flight mass spectrometer has two acceleration regions, one of which has a low draw out potential, which means that ions dissociating on the microsecond time scale can do so in this region. The results are ion time-of-flight distributions that are asymmetric. These asymmetric distributions can be modeled to extract absolute unimolecular decay rate constants.22 In the case that more than one product ion is generated, all are formed with the same overall decomposition rate constant of the precursor, and thus rate information for all channels can be gleaned from the asymmetric time-of-flight of just one of the products, and the relative rate constants are obtained as branching ratios of the fragment ion peaks. The photon energies used in this experiment range from 10 to 17 eV with steps ranging in size from 0.04 eV in the region where the relative abundances are changing rapidly, to 0.08 eV

Figure 1. Optimized structures calculated at B3LYP 6-311G++(d,p) level of theory, the structures are (a) C10H10+·, (b) C14H12+·, (c) C9H7+ and (d) C13H9+. Naphthalene structure (1a) with different carbon sites numbered in order of most favorable site for hydrogen addition.

three sites, position 1 is the most favorable for the first hydrogen addition.7 Going one step further, if we also consider the addition of a second hydrogen atom, it has been shown in larger PAHs that the carbon atoms on the outer edge immediately adjacent to the substituted carbon (site 2 in Figure 1a) is the most favorable location.9 This paper also shows the results of 9,10-dihyrophenanthrene, whose structure is similar to that of 1,2-dihydronaphthalene. As, to date, no energetics information has been found relating to the dissociation of either molecule. We are presenting them together to determine if their similar structures translate to similar dissociation properties as well. We have previously obtained the threshold photoelectron spectrum of 1,2-dihydronaphthalene16 and herein we explore its cation’s dissociation reactions and those of 9,10-dihydrophenanthrene cation via mass-analyzed ion kinetic energy spectrometry (MIKES), collision induced dissociation (CID) and metastable ion collision induced dissociation (MI-CID) experiments as well as imaging photoelectron−photoion coincidence (iPEPICO) experiments. The latter technique was performed at the VUV beamline of the Swiss Light Source (SLS), which selects ions as a function of their internal energy with 2 meV photon resolution and measures dissociation rates within the 103−107 s−1 range. This experimental setup includes a dual acceleration zone. Ions dissociating with the above range of rate constants will do so in the second, low field, region of the acceleration zone resulting in asymmetric time-of-flight profiles for fragment ions. By fitting the breakdown curves as well as the asymmetric time-of-flight profiles, one can achieve kinetic modeling of the dissociative ionization processes. This allows an accurate determination of barrier heights and entropies of activation that have an impact on dissociation and are useful parameters for the atmospheric, combustion or interstellar chemistry models to postulate pathways of aromatic formation by hydrogen-abstraction and C2H2 addition; as well as the possible catalytic role of PAHs in the ISM. The remainder of the paper is organized as follows. A description of the two experimental setups is given in section 2. Next, the computational approach is described in section 3. The results are presented and discussed in section 4. 1808



Article

Figure 2. MIKES spectra for various degrees of dehydrogenation of 1,2-dihydronaphthalene: (a) 1,2-dihydronaphthalene (m/z = 130), (b) 1,2dihydronaphthalene after loss of one hydrogen atom (m/z = 129), (c) 1,2-dihydronaphthalene after the loss of 2 hydrogen atoms (m/z = 128) with an inset showing the same spectrum magnified to show the presence of peaks at m/z = 102 and 78, and (d) CID mass spectrum of source generated C9H7+ from 1,2-dihydronaphthalene (m/z 115).

where μ = [∫ tTOF(t) dt]/[∫ TOF(t) dt], σ2 = [∫ (t − t0)2TOF(t) dt]/[∫ TOF(t) dt], t0 refers to the moment center, and σp2 refers to the precursor ion. After deconvolution of the three peaks in question, 13C contributions were taken into account prior to the calculation of final relative abundances for use in the breakdown diagrams.

in the lower and upper energies where the ion ratios are more constant. This energy range is lower that the double ionization around 21 eV.23 Due to the low resolving power of the TOF mass spectrometer, it was necessary to use a deconvolution procedure to extract the fractional abundances of the overlapping precursor, H-loss and 2H-loss peaks. This is based on the idea of finding the weighted center of the TOF distribution. For example, the molecular ion M+• has a peak with a certain time-of-flight (TOF) center and full width halfmaximum. As the [M − H]+ peak grows with increasing photon energy, the center of the TOF distribution of the peak cluster shifts toward the TOF of the [M − H]+ ion, the shift being proportional to the relative abundance of the two peaks. This procedure can be generalized for a cluster of three peaks, making use of the second moment of the TOF distribution around the peak center. In the present case, this involves either of the combinations [M − H]+, M+• (with [13C − M − H]+), 13 C−M+• or [M − 2H]+•, [M − H]+ (with [13C − M − 2H]+), M+•. If we consider three peaks with relative abundances a, b, and c, with arrival times of t1, t2, and t3, and each with the same fullwidth-half-maximum (due to the low mass of H the difference in the peak widths due to kinetic energy release broadening of the fragment ions relative to the precursor ion should be negligible), one can derive the following expressions for a, b, and c:12

3. COMPUTATIONAL METHODS Ab Initio Calculations. Ionic structures were calculated for 1,2-dihydronaphthalene, 9,10-dihydrophenanthrene, and all potential fragments, as well as the neutral structures for the precursor molecules. These ab initio calculations were completed using Gaussian 09 suite of programs.24 Geometry optimizations and vibrational calculations were performed using the B3LYP/6-311++G(d,p) level of theory. The harmonic vibrational frequencies and rotational constants for neutral and ionic 1,2-dihydronaphthalene and 9,10-dihydrophenanthrene were used as input for subsequent RRKM calculations. RRKM Calculations. To determine the 0 K activation energy (E0) and the entropy of activation (Δ‡S) from the experimental breakdown curves, the rate of each dissociation pathway, k(E), needs to be determined. k(E) was calculated according to the usual equation,25 k(E) =

σN ‡(E − Eo) hρ(E)

(4)

where σ is the symmetry of the dissociation channel or reaction degeneracy, h is Plank’s constant, N‡(E − E0) is the number of internal (rotational and vibrational) states of the transition state at its internal energy (E − E0) and ρ(E) is the density of states of the reactant ion at an internal energy E. For 1,2dihydronaphthalene σ takes the values of 2, 1, 1, 2, and 1 for R1a, R1b, R2, R3, and R4 respectively, whereas for 9,10dihydrophenanthrene the values are 4 and 1 for R5 and R6 (see results and discussions for reaction definitions). The calculated vibrational frequencies and rotational constants were used to

a = (σ 2 − σp2 − μ(t 2 + t3) + t 2t3)/((t1 − t 2)(t1 − t3)) (1)

b = (σ 2 − σp2 − μ(t1 + t 2) + t1t 2)/((t3 − t 2)(t3 − t1)) (2)

c = (σ 2 − σp2 − μ(t1 + t3) + t1t3)/((t 2 − t1)(t 2 − t3)) (3) 1809



Article

structures calculated, as seen in Table 1. Thus, included in Figure 3 is the isomerization of DHN to ionized methylindene

calculate the reactant ion density of states according to the Beyer and Swinehart direct count algorithm.26 The transition states were not calculated; instead, an appropriate vibrational mode was removed to simulate the critical transition coordinate, such as a C−H stretch removed for the formation of C10H7+ (compounds 7 and 9 from Figure 3) from 1,2dihydronaphthalene, or a mode corresponding to a change in H−H distance (and H−C distance) for H2 loss. These frequencies are listed in Table S1 of Supporting Information. The transition state sum-of-states were varied as part of the fitting process by scaling the frequencies of the transitional modes by a common factor. Δ‡S1000K was calculated at 1000 K according to the standard statistical mechanics equations. No restriction has been imposed on the degree of tightness or looseness of the transitions states. Theoretical Model Fitting. The miniPEPICO program of Sztáray et al. was used for fitting the experimental breakdown curves and ion time-of-flight distributions.22 In brief, this program employs RRKM theory to fit experimental breakdown curves for competing dissociation channels based on vibrational frequencies and rotational constants for the precusor neutral (to calculate the initial internal energy distribution of the formed ion) and dissociating ions (M+• and (M − H)+), which in this work consist of C10H10+• and C10H9+ for 1,2dihydronaphthalene and only C14H12+• for 9,10-dihydrophenanthrene, the experimental conditions such as temperature (in these cases 298 and 313 K), the experimental data, and the physical aspects of the iPEPICO apparatus, as well as the barrier heights of each dissociation step. The experimental data required consist of the percent abundance of each fragment at different energies, which in this case ranged from 10 to 17 eV, as well as the TOF spectra at different energies. Photon energies were converted to ion internal energies for the RRKM modeling using the reported ionization energy of both compounds, namely, 8.010 ± 0.010 eV for 1,2-dihydronaphthalene16 and 7.81 eV for 9,10-dihydrophenanthrene. The value for 9,10-dihydrophenanthrene was obtained from an extrapolation of the linear portion of the photoionization efficiency curve near the threshold for ionization (see the Supporting Information, Figure S1).

Table 1. Comparison of Energies Calculated at the B3-LYP/ 6-311++G(d,p) Level of Theory for the Formation of Four Structures of C9H7+ from 1,2-Dihydronaphthalene (Reaction R2)

(R1)

C10H10+• → C9H 7+ + CH3• (m /z = 115)

(R2)

There are also two sequential channels which consist of hydrogen loss from C10H9+, reactions R3 and R4. C10H 9+ → C10H8+• + H• (m /z = 128)

(R3)

C10H 9+ → C10H 7+ + H 2 (m /z = 127)

(R4)

ΔE (eV)

indenyl tropylacetylene tropylcyclobutadiene cyclononapentaene

2.155 3.084 3.988 6.060

(2), which itself can undergo H (R1b) and CH3 loss (R2). In Figure 2d, the CID mass spectrum of m/z 115 demonstrates that the primary fragment appeared at m/z 89, which corresponds to a loss of C2H2 from the indenyl cation. This assignment of the indenyl structure is in agreement with work done by Dass and Gross who determined that all C10H10+• isomers produce the same fragment ions after rearranging to a common structure.27 While fitting the iPEPICO data, we observed that there are two competing channels for reaction R1, a low energy channel and a high energy channel. If the isomerization is considered, then there are two plausible sources for H-loss, the hydrogens attached to the sp3-carbons on the dihydro structure (R1a), resulting in protonated naphthalene (3), and the solitary hydrogen attached to the sp3-carbon on the indene ring of methylindene (R1b), giving structure 5. In the iPEPICO section, it will be shown that all the sequential fragments come from protonated naphthalene (3). Starting from protonated naphthalene (3), losing a hydrogen atom would likely yield the naphthalene structure 6. In the case of H2 loss, the predicted structure is 7, reported previously by West et al.12 as the structure of the naphthalene radical cation after the loss of a hydrogen atom. Both of these channels result in the restoration of the planar nature of the molecule and the elimination of sp3 carbon sites. MI-CID results indicate that, when source generated ions are used, there is sufficient energy to see evidence of fragmentation other than from 3. Starting from 5, the second sequential Hloss reaction would be the loss of a sp3-H from the methyl group (R3b) to make 1-methyleneindene (8) due to the lower energy requirement. In the case of H2 loss from 5, both hydrogen atoms would come from the methyl carbon (R4b) to make 9. Figure 4 shows a comparison of the MI-CID mass spectra of C10H7+ ions from 1,2-dihydronaphthalene (7 and/or 9) and from naphthalene (7). All of the features characteristic of structure 7 appear in both traces, with additional peaks present (around 115) assigned to 9. This indicates the likelihood of a mixture of C10H9+ ion structures being formed in the ion source of the mass spectrometer for 1,2dihydronaphthalene. So, both 3 and 5 are formed in the decomposition of high internal energy molecular ions. 1,2-Dihydronaphthalene: iPEPICO Spectrometry. iPEPICO experiments were conducted for photon energies ranging from 10 to 17 eV. All fragments observed in the tandem mass spectrometry experiments for C10H10+• and C10H9+ were observed, with the sequential channels growing in at approximately 14.3 eV. Figure 5 shows the experimental breakdown curves for all reaction channels.

4. RESULTS AND DISCUSSION 1,2-Dihydronaphthalene: Tandem Mass Spectrometry. Parts a−c of Figure 2 show MIKES spectra for the dissociation of 1,2-dihydronaphthalene (DHN, C10H10+•) and for its sequential dissociation after the loss of the first hydrogen atom (C10H9+) and second H atom (C10H8+). C10H10+• exhibits two distinct primary dissociation channels, which correspond to reactions R1 and R2 C10H10+• → C10H 9+ + H• (m /z = 129)

C9H7+ isomer

The most probable structures for reactions R1−R4 are shown in Figure 3. For reaction R2, the product ion structure assigned was that of indenyl cation 4. Indenyl was determined to be the most stable structure as compared to other polycyclic and open 1810



Article

Figure 3. Fragmentation model for 1,2-dihydronaphthalene, outlining most probable structures for reactions 1−4.

As stated previously, the primary channels come from two isomers; therefore, not only was it necessary to fit the energetics of dissociation, but also the energy barrier between the two wells had to be estimated at least. A variety of barriers were used to test the sensitivity of the model in deriving the various dissociation energies, and the results are shown in Figure S2 in the Supporting Information. When the isomerization barrier height was varied between 1.5 and 2.0 eV, there was very little change in E0 for all the affected channels. The E0 values reported in Table 2 are for the average of these values as there is no way to say for certain where exactly the isomerization barrier lies. The E0 values for these two competing H loss processes are 2.44 ± 0.10 and 2.22 ± 0.05 eV for R1a and R1b, respectively. This could indicate that, solely on the basis of E0, the sequential channels are likely to come from 1-methylindene (R1b). However, when Δ‡S is factored in, 27 ± 14 J K−1 mol−1 (R1a) and −11 ± 5 J K−1 mol−1 (R1b), it can be seen that reaction R1a quickly becomes the dominant fragmentation channel. This relationship is illustrated in Figure 7; at the appearance energy of the sequential channels, at approximately 14 eV, there is very little contribution from R1b. The energetics extracted for R2 were 2.57 ± 0.12 eV and 18 ± 11 J K−1 mol−1, for E0 and Δ‡S, respectively. Due to the assumption that only structure 3 dissociates further, the assignment of the sequential channels becomes quite straightforward. As was alluded to earlier, this assumption allowed for the fitting of these channels using a single-well approximation. For R3, a literature value (calculated at the B3LYP/6-311++G(d,p) level of theory) for the loss of a hydrogen atom from protonated naphthalene (protonated in the 1 position) was found to be 2.71 eV, which is in excellent

Figure 4. Comparison of the CID mass spectrum of m/z = 127 (C10H7+) generated from metastable H2 loss from source generated C10H9+ ions from 1,2-dihydronaphthalene (broken gray curve, compounds 7 and 9) and from metastable H loss from source generated naphthalene ions (solid black curve, compound 7 only).

Due to the complex nature of the primary channels, they were fit first while ignoring the sequential channels. The breakdown curves corresponding to the three primary channels were fit together with the asymmetric TOF peak shapes for C9H7+ (see Computational Methods), the latter to aid in the determination of Δ‡S (the tightness of the transition state). As will be shown later in the section, the sequential channels were fit using a single well model where only R1a is responsible for all sequential dissociations. The fitted breakdown curves can be seen in Figure 5, whereas the fitting of the asymmetric TOF peaks for C9H7+ can be seen in Figure 6. TOF fitting was only used for the primary channels because all channels exhibit a Gaussian peak shape above 13.4 eV, when the rate constant for dissociation of the precursor ion is greater than 106 s−1. The resulting E0 and Δ‡S are shown in Table 2. 1811



Article

Figure 5. Experimental iPEPICO breakdown curve for the 1,2-dihydronaphthalene radical cation over the photon energy range 10−17 eV. The reaction number for each product channel has been included in parentheses. Calculated fits are overlaid. Derived energetic and entropic parameters can be found in Table 2.

agreement with the present RRKM value of 2.72 ± 0.19 eV.28 This supports the current results that R3 is only produced from the protonated naphthalene cation. The loss of H2 from protonated naphthalene was calculated at the same level of theory to be 3.02 eV. This is similar to the present RRKM value 2.85 ± 0.10 eV. The RRKM Δ‡S values for R3 and R4 are 9 ± 17 and 9 ± 7 J K−1 mol−1. It was not possible to find any literature values with which to compare; however, as these channels result in the loss of H and H2 presumably by bond cleavage reactions, a slightly positive Δ‡S is not unexpected. 9,10-Dihydrophenanthrene: Tandem Mass Spectrometry. Figure 8 shows the MIKES spectrum for the dissociation of 9,10-dihydrophenanthrene (DHP, C14H12+•). Similar to the case for 1,2-dihydronaphthalene, only two fragments are present;: one resulting from the loss of hydrogen (C14H11+) and one from the loss of a methyl radical (C13H9+). These fragments correspond to reactions R5 and R6: C14 H12+• → C14 H11+ + H• (m /z = 179)

(R5)

C14 H12+• → C13H 9+ + CH3• (m /z = 165)

(R6)

DHP to 9-methylfluorene must take place during the unimolecular chemistry of DHP. To test this assumption, the MI-CID mass spectra for m/z 165 were compared between the methyl loss from 9,10-dihydrophenanthrene and the hydrogen loss from fluorene. These results can be seen in Figure 10, which shows perfect agreement between the two mass spectra. This indicates that m/z 165 from both molecules results in the same structure. Because fluorene only lost a hydrogen atom, it is likely that the resulting structure would be very similar to that of the precursor ion. As was observed with ionized dihydronaphthalene, there are two possible channels for H atom loss. Thus, reaction R5 has been split into two competing channels, a channel due to loss of H from an sp3 carbon on DHP (R5a) to form 12 and one due to loss of H from 9-methylfluorene (R5b) to form 13. 9,10-Dihydrophenanthrene: iPEPICO Spectrometry. iPEPICO experiments were conducted for photon energies ranging from 10 to 17 eV. All fragments observed in the tandem mass spectrometry experiments for C14H12+• were observed; there were no consecutive reaction channels observed in the given energy range. Figure 11 shows the experimental and fitted breakdown curves for the three reaction channels. TOF fitting of the C13H9+ peak was used to aid in the assignment of the reactions channels, similarly to 1,2dihydronaphthalene (see Figure S3 in the Supporting Information). There is a hint of a second, fast component on the TOF distributions as shown in Figure 11, which would indicate that an isomerization pathway opens up at or close to the dissociation threshold. The data quality is insufficient to attempt fitting a two-well model to model the dissociation. The RRKM E0 and Δ‡S values are shown in Table 3.

The predicted structures for these fragments are shown in Figure 9. No literature was found for the fragmentation of ionized 9,10-dihydrophenanthrene, so the structures are based on those predicted for ionized 1,2-dihydronaphthalene as both molecules have similar starting structures and, as will be discussed later, the behaviors of the two systems are comparable. Reaction R6 was assigned the fluorenyl structure (11), partially on the basis of the results from calculations for 1,2-dihydronaphthalene where the indenyl structure was determined to be the most stable. Thus, the isomerization of 1812



Article

Figure 6. Representative TOF fits, for 1,2-dihydronaphthalene, calculated during the RRKM fitting of experimental iPEPICO data. The region shown is the C9H7+ region, as this peak was the only asymmetric TOF peak that was not obscured due to its proximity to other peaks. As the photon energy increases, it can be seen that the peak becomes increasingly Gaussian in shape.

Table 2. Calculated Reaction Endothermicities and Fitted 0 K Activation Energies and Entropies of Activation for Reactions 1−4 of 1,2-Dihydronaphthalene B3-LYP/6311++G(d,p)

(R1a) C10H10+• → • C10H9+ + H (R1b) C10H10+• → • C10H9+ + H (R2) C10H10+• → • C9H7+ + CH3 (R3) C10H9+ → • C10H8+• + H (R4) C10H9+ → •C10H7+ + H2

RRKM fit

ΔE (eV)

E0 (eV)

Δ‡S1000K (J K−1 mol−1)

1.99

2.44 ± 0.10

27 ± 14

2.39

2.22 ± 0.05

−11 ± 5

2.24

2.57 ± 0.12

18 ± 11

3.07

2.72 ± 0.19

9 ± 17

3.01

2.85 ± 0.10

9±7

Figure 7. Experimental breakdown diagram for the H loss fragmentation channel for 1,2-diydronaphthalene (R1) over the photon energy range of 10−17 eV. Calculated fits explicitly showing R1a and R1b are overlaid as well as their summation to fit the data.

When fitting reaction R5, it quickly became apparent that it was not necessary to fit the H loss using two channels as was done with 1,2-dihydronaphthalene. Two distinct reactions were necessary, however, to fit R6 correctly, and this was taken to be evidence of the need to incorporate an isomerization reaction to methylfluorene. As done previously, a range of isomerization barriers were used for the fitting to determine how sensitive the channels are to isomerization (Figure S4, Supporting Information). For R5, we obtained an E0 value of 2.37 ± 0.12 and a Δ‡S value of 18 ± 19 J K−1 mol−1. Although there are no literature data for comparison, these values are between those found for R1a and R1b in 1,2-dihydronaphthalene. The

final reaction channel for 9,10-dihydrophenanthrene is reaction R6 at E0 2.38 ± 0.15 eV and Δ‡S −3 ± 15 J K−1 mol−1.

4. CONCLUSION Comparing the two molecules presented in this paper, one can see that both size and ring geometry play a role in the energetics of the dissociation channels investigated. Although molecular size has little effect on the C−H bond dissociation 1813



Article

Figure 8. MIKES spectrum of 9,10-dihydrophenanthrene. Only two metastably generated fragments are observed, corresponding to the loss of hydrogen (m/z 179) and the loss of methyl radical (m/z 165).

Figure 10. Comparison of m/z = 165 generated by the methyl loss from (broken gray curve) 9,10-dihydrophenanthrene and by the hydrogen loss from fluorene (C13H10+) (solid black curve).

energy of ionized PAHs, their energy remained between 4.60 and 4.65 eV with increasing size from benzene through naphthalene to phenanthrene;29 as soon as the molecule is protonated, the size effect becomes apparent with the bond dissociation energy decreasing as the ring system expands, starting at 3.29 eV for protonated benzene and decreasing to 2.71 eV for protonated naphthalene and decreasing further to 2.57 eV for protonated phenanthrene.28 This trend can be extrapolated even further to the dihydro-PAHs when the C−H bond dissociation energies determined in this work are

considered; 2.44 ± 0.10 for 1,2-dihydronaphthalene (R1a) and 2.37 ± 0.10 eV for 9,10-dihydrophenanthrene (R5). The presence of a second sp3 carbon site has further decreased the C−H bond dissociation energy to facilitate the restoration of the unaltered PAH ion. It is interesting to note that at no point is the loss of H2 observed from the 9,10-dihydrophenanthrene cation, and it is observed only as a secondary channel from 1,2-dihydronaphthalene cation. This is in keeping with what has previously been reported in the literature where the loss of H2 is mainly observed from neutral and protonated species, such as

Figure 9. Fragmentation model for 9,10-dihydrophenanthrene, outlining the most probably structures for reactions R5 and R6. 1814



Article

frequencies for the processes modeled with eq 1 for both ions. This information is available free of charge via the Internet at http://pubs.acs.org

■

AUTHOR INFORMATION

Corresponding Author

*Corresponding Author E-mail: [email protected]. Phone: 1-613-562-5800 ext 6038. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work has been supported by the program “Molécules et grains: du laboratoire à l′Univers” of the Midi-Pyrénées Observatory and by the University Paul Sabatier. P. M. Mayer thanks the Natural Sciences and Engineering Research Council of Canada for continuing financial support. B. Sztáray is gratefully acknowledging the support of the National Science Foundation (CHE-1266407). The iPEPICO experiments were carried out at the VUV beamline of the Swiss Light Source of the Paul Scherrer Institut. The research leading to these results has received f unding f rom the European Community’s Seventh Framework Programme (FP7/2007−2013) under grant agreement no. 226716.

Figure 11. Experimental breakdown curve for the 9,10-dihydrophenanthrene radical cation over the photon energy range of 10−17 eV. The reaction number for each product channel has been included in parentheses. Calculated fits are overlaid. Derived energetic and entropic parameters can be found in Table 3.

Table 3. Calculated Reaction Endothermicities and 0 K Activation Energies and Entropies of Activation for Reactions R5 and R6 of 9,10-Dihydrophenanthrene B3-LYP/6-311+ +G(d,p)

(R5a) C14H12+• → • C14H11+ + H (R6) C14H12+• → • C13H9+ + CH3

■

RRKM fit

ΔE (eV)

E30 (eV)

Δ‡S (J K mol−1)

2.23

2.37 ± 0.12

18 ± 19

1.98

2.38 ± 0.15

−3 ± 15

(1) Carruthers, G. R Rocket Observation of Interstellar Molecular Hydrogen. Astrophys. J. Lett. 1970, 161, L81−L85. (2) Gould, R. J. a.; S., E. E. The Interstellar Abundance of the Hydrogen Molecule. I. Basic Processes. Astrophys. J. Lett. 1963, 138, 393−407. (3) Lemaire, J. L.; Vidali, G.; Baouche, S.; Chehrouri, M.; Chaabouni, H.; Mokrane, H. Competing Mechanisms of Molecular Hydrogen Formation in Conditions Relevant to the Interstellar Medium. Astrophys. J. Lett. 2010, 725, L156. (4) Cazaux, S. M.; Spaans, S.; Allouche, M. A.When Sticking Influences H2 Formation. Astron. Astrophys. 2011, 535, A27−A36. (5) Pahs and the Universe: A Symposium to Celebrate the 25th Anniversary of the Pah Hypothesis; Joblin, C., Tielens, A. G. G. M., Eds.; EAS Publications Series; EAS: Versoix, Switzerland, 2011; Vol. 46. (6) Cassam-Chenaï, P.; Pauzat, F.; Ellinger, Y. Is Stripping of Polycyclic Aromatic Hydrocarbons a Route to Molecular Hydrogen? In Molecules and Grains in Space, 50th International Meeting of Physical Chemistry; Nenner, I., Ed.; American Institute of Physics Press: Mont Sainte-Odile, France, 1994; Vol. 312, pp 543. (7) Hirama, M.; Tokosumi, T.; Ishida, T.; Aihara, J.-i. Possible Molecular Hydrogen Formation Mediated by the Inner and Outer Carbon Atoms of Typical Pah Cations. J. Chem. Phys. 2004, 305, 307− 316. (8) Bauschlicher, Charles W., Jr. The Reaction of Polycyclic Aromatic Hydrocarbon Cations with Hydrogen Atoms: The Astrophysical Implications. Astrophys. J. Lett. 1998, 509, L125. (9) Rauls, E.; Hornekær, L. Catalyzed Routes to Molecular Hydrogen Formation and Hydrogen Addition Reactions on Neutral Polycyclic Aromatic Hydrocarbons under Interstellar Conditions. Astrophys. J. 2008, 679, 531. (10) Vala, M.; Szczepanski, J.; Oomens, J.; Steill, J. D. H2 Ejection from Polycyclic Aromatic Hydrocarbons: Infrared Multiphoton Dissociation Study of Protonated 1,2-Dihydronaphthalene. J. Am. Chem. Soc. 2009, 131, 5784−5791. (11) Thrower, J. D.; Jørgensen, B.; Friis, E. E.; Baouche, S.; Mennella, V.; Luntz, A. C.; Andersen, M.; Hammer, B.; Hornekær, L. Experimental Evidence for the Formation of Highly Superhydrogenated Polycyclic Aromatic Hydrocarbons through H Atom Addition and Their Catalytic Role in H2 Formation. Astrophys. J. 2012, 752, 3.

−1

protonated 1,2-dihydronaphthalene,10 9,10-dihydrophenanthrene,30 and protonated PAHs, but not from radical cations.28 As a result, it may be unlikely that small PAHs catalyze H2 formation in ionic environments, in which H + H addition is more likely to result in isomerization and methyl group formation. No direct comparisons can be made for the secondary channels from ionized 1,2-dihydronaphthalene (R3 and R4) because these channels were not observed in the iPEPICO spectra of ionized 9,10-dihydrophenanthrene in the studied energy range. The energetics measured, 2.72 ± 0.19 eV (R3) and 2.85 ± 0.10 eV (R4), are in good agreement with calculated values found in the literature.28 And though there were no literature values found for Δ‡S, the values of 9 ± 17 J K−1 mol−1 (R3) and 9 ± 7 J K−1 mol−1 (R4) are reasonable for bond cleavage reactions involving little to no reverse energy barrier.

■

REFERENCES

ASSOCIATED CONTENT

S Supporting Information *

Full authorships for ref 24. Description of the deconvolution process. Figure S1 indicating how the IE was selected for 9,10dihydrophenanthrene, Figure S2 showing the effect of barrier energy on the RRKM fits for 1,2-dihydronaphthalene, Figure S3 showing the RRKM fitted TOF distributions for the dissociation channels of ionized 9,10-dihydrophenanthrene, and Figure S4 showing two additional RRKM fits for 9,10dihydrophenanthrene. Table S1 contains the vibrational 1815



Article

(12) West, B.; Joblin, C.; Blanchet, V.; Bodi, A.; Sztáray, B.; Mayer, P. M. On the Dissociation of the Naphthalene Radical Cation: New Ipepico and Tandem Mass Spectrometry Results. J. Phys. Chem. A 2012, 116, 10999−11007. (13) Ho, Y. P.; Dunbar, R. C.; Lifshitz, C. C-H Bond Strength of Naphthalene Ion - a Reevaluation Using New Time-Resolved Photodissociation Results. J. Am. Chem. Soc. 1995, 117, 6504−6508. (14) Laskin, J.; Futrell, J. Internal Energy Distributions Resulting from Sustained Off-Resonance Excitation in Fourier Transform Ion Cyclotron Resonance Mass Spectrometry. Ii. Fragmentation of the 1Bromonaphthalene Radical Cation. J. Phys. Chem. A 2000, 104, 5484− 5494. (15) Gotkis, Y.; Oleinikova, M.; Naor, M.; Lifshitz, C. TimeIndependent Mass Spectra and Breakdown Graphs. 17. Naphthalene and Phenanthrene. J. Phys. Chem. 1993, 97, 12282. (16) Mayer, P. M.; Blanchet, V.; Joblin, C. Threshold Photoelectron Study of Naphthalene, Anthracene, Pyrene, 1,2-Dihydronaphthalene, and 9,10-Dihydroanthracene. J. Chem. Phys. 2011, 134, 244312. (17) Holmes, J. L.; Mayer, P. M. Combined Mass Spectrometric and Thermochemical Examination of the C2H2N Family of Cations and Radicals. J. Phys. Chem. 1995, 99, 1366−1370. (18) Traeger, J. C.; Mommers, A. A. A Data Acquisition System for Mass-Analyzed Ion Kinetic Energy Spectra Using a Personal Computer. Org. Mass Spectrom. 1987, 22, 592−596. (19) Johnson, M.; Bodi, A.; Schulz, L.; Gerber, T. Vacuum Ultraviolet Beamline at the Swiss Light Source for Chemical Dynamics Studies. Nucl. Instrum. Methods Phys. Res. A 2009, 610, 597−603. (20) Bodi, A.; Johnson, M.; Gerber, T.; Gengeliczki, Z.; Sztáray, B.; Baer, T. Imaging Photoelectron Photoion Coincidence Spectroscopy with Velocity Focusing Electron Optics. Rev. Sci. Instrum. 2009, 80, 034101. (21) Bodi, A.; Sztaray, B.; Baer, T.; Johnson, M.; Gerber, T. Data Acquisition Schemes for Continuous Two-Particle Time-of-Flight Coincidence Experiments. Rev. Sci. Instrum. 2007, 78, 084102. (22) Sztáray, B.; Bodi, A.; Baer, T. Modeling Unimolecular Reactions in Photoelectron Photoion Coincidence Experiments. J. Mass Spectrom. 2010, 45, 1233−1245. (23) Ruehl, E.; Price, S. D.; Leach, S. Single and Double Photoionization Processes in Naphthalene between 8 and 35 eV. J. Phys. Chem. 1989, 93, 6312−6321. (24) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Vreven, T.; Kudin, K. N.; Burant, J. C.; et al. Gaussian 03, Revision C.02; Gaussian, Inc.: Wallingford, CT, 2003. (25) Baer, T.; Hase, W. L. Unimolecular Reaction Dynamics, Theory and Experiments; Oxford University Press: New York, 1996. (26) Beyer, T.; Swinehart, D. R. Number of Multiply-Restricted Partitions [A1] (Algorithm 448). ACM Commun. 1973, 16, 379. (27) Dass, C.; Gross, M. L. A Mass-Spectrometry Mass-Spectrometry Investigation of the Nature of [C10H10]+, [C9H7]+ and [C10H8]+ Gas-Phase Ions. Org. Mass Spectrom. 1983, 18, 542−546. (28) Kapinus, V. A. Photophysical Properties of Protonated Aromatic Hydrocarbons; California Institute of Technology: Pasadena, CA, 2005. (29) Fujiwara, K.; Harada, A.; Aihara, J.-i CH Bond Dissociation Energies of Polycyclic Aromatic Hydrocarbon Molecular Cations: Theoretical Interpretation of the (M-1)+ Peak in the Mass Spectra. J. Mass Spectrom. 1996, 31, 1216−1220. (30) Szczepanski, J.; Oomens, J.; Steill, J. D.; Vala, M. T. H2 Ejection from Polycyclic Aromatic Hydrocarbons: Infrared Multiphoton Dissociation Study of Protonated Acenaphthene and 9,10-Dihydrophenanthrene. Astrophys. J. 2011, 727, 12.

1816


Dynamics of Hydrogen and Methyl Radical Loss ... - ACS Publications

Recommend Documents