Photodissociation of Pyrene Cations: Structure and ... - ACS Publications

Article pubs.acs.org/JPCA

Photodissociation of Pyrene Cations: Structure and Energetics from C16H10+ to C14+ and Almost Everything in Between Brandi West,† Francesca Useli-Bacchitta,‡,§ Hassan Sabbah,‡,§ Valérie Blanchet,∥,⊥ Andras Bodi,# Paul M. Mayer,† and Christine Joblin*,‡,§ †

Chemistry Department, University of Ottawa, Ottawa K1N 6N5, Canada Université de Toulouse, UPS-OMP, IRAP, 31062 Toulouse, France § CNRS, IRAP, 9 Av. colonel Roche, BP 44346, F-31028 Toulouse, Cedex 4, France ∥ Laboratoire des Collisions Agrégats Réactivité, Université de Toulouse-CNRS, 31062 Toulouse, France # Molecular Dynamics Group, Paul Scherrer Institut, Villigen 5232, Switzerland ‡

S Supporting Information *

ABSTRACT: The unimolecular dissociation of the pyrene radical cation, C 16 H 10 +• , has been explored using a combination of computational techniques and experimental approaches, such as multiple photon absorption in the cold ion trap Piège à Ions pour la Recherche et l’Etude de Nouvelles Espèces Astrochimiques (PIRENEA) and imaging photoelectron photoion coincidence spectrometry (iPEPICO). In total, 22 reactions, involving the fragmentation cascade (H, C2H2, and C4H2 loss) from the pyrene radical cation down to the C14+• fragment ion, have been studied using PIRENEA. Branching ratios have been measured for reactions from C16H10+•, C16H8+•, and C16H5+. Density functional theory calculations of the fragmentation pathways observed experimentally and postulated theoretically lead to 17 unique structures. One important prediction is the opening of the pyrene ring system starting from the C16H4+• radical. In the iPEPICO experiments, only two reactions could be studied, namely, R1 C16H10+• → C16H9+ + H (m/z = 201) and R2 C16H9+ → C16H8+• + H (m/z = 200). The activation energies for these reactions were determined to be 5.4 ± 1.2 and 3.3 ± 1.1 eV, respectively.

1. INTRODUCTION Polycyclic aromatic hydrocarbon (PAH) molecules have long been thought to play a very active role in the chemistry of the interstellar medium. For this reason, these molecules have been the focus of many laboratory studies to try and further our understanding of them and their role.1 Of particular interest is their photostability against dissociation, as PAH molecules absorb efficiently the UV radiation from surrounding stars.2 Previously, we have investigated the photoelectron spectra of naphthalene, anthracene, pyrene, and the dihydro forms of naphthalene and anthracene3 and carried out detailed studies on their thermochemical properties.4,5 Pyrene is the first example of a pericondensed PAH in our list of species. A previous study on the dissociation of pyrene conducted by Ling et al. shows three dissociation channels, two primary channels (H loss and H2 loss), and one secondary channel (2H loss).6 H2 loss has been reported to have the lower appearance energy (15.2 ± 0.2 versus 16.2 ± 0.2 eV for H loss at a residence time of 24 μs) though it is quite apparent from the breakdown curves that H loss is the dominant channel. In the present work, we are reinvestigating the dissociation of the pyrene cation using both the Piège à Ions pour la Recherche et l’Etude de Nouvelles Espèces Astrochimiques (PIRENEA) setup, which consists of a cold ion cyclotron © 2014 American Chemical Society

resonance (ICR) cell to simulate the interstellar environment, and the imaging photoelectron photoion coincidence (iPEPICO) experiment available at the Swiss Light Source (SLS). In PIRENEA, dissociation is achieved by heating the trapped ions with a Xe arc lamp. In iPEPICO experiments, absorption of tunable VUV radiation by neutrals is used to produce internal energy selected parent ions, which then dissociate as a function of their energy. In chemical models of interstellar PAHs, subsequent dissociation of photofragments has to be considered as well. For instance, carbon clusters C24/C24+ are predicted to be formed from the full dehydrogenation of coronene C24H12.2 The PIRENEA experimental results were therefore used to create a fragmentation map outlining all the different pathways connecting the fragments from the pyrene radical cation (C16H10+•) down to the bare 14 carbon clusters (C14+•) as well as all side channels that occurred. The data collected from the iPEPICO experiments were used to create breakdown curves for H loss and 2H loss, in which the fractional ion abundances of parent and fragment ions in threshold photoionization are Received: June 27, 2014 Revised: August 13, 2014 Published: August 18, 2014 7824

dx.doi.org/10.1021/jp506420u | J. Phys. Chem. A 2014, 118, 7824−7831

The Journal of Physical Chemistry A

Article

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.12 The mass spectrum corresponding to this ring is subtracted from the center TOF distribution to obtain the threshold ionization mass spectra. The photon energies used in this experiment range from 17 to 22 eV with steps ranging in size from 0.4 up to 20.4 eV and of 0.2 eV above 20.4 eV. A differentially pumped gas filter was used to filter out higher harmonic radiation below the ionization energy of neon with a suppression factor of higher than 106. The higher harmonic contamination of the VUV light above 21.5 eV is efficiently suppressed by the 1200 lines/mm laminar grating.9 It was necessary to use a deconvolution procedure to the threshold photoionization mass spectra to extract the fractional abundances of the overlapping parent, H loss, and 2H loss peaks. This is based on the idea of finding the weighted center of the TOF distribution and has been used previously for our work on naphthalene.5 For example, the molecular ion M+• has a peak with a certain TOF center and a full width at 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, that is, the peak width. In the present case, this involves either of the combinations [M − H]+, M+•, 13 C-M+• or [M − 2H]+•, [M − H]+, M+•.2 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 diagram. 2.3. Computational Methods. 2.3.1. Ab Initio Calculations. Ionic structures were calculated for pyrene and all potential fragments, as well as the neutral structure for pyrene. These calculations were completed using the Gaussian 09 suite of programs.13 Geometries were optimized using the B3-LYP/ 6-311G++(d,p) level of theory, and these optimized geometries were then used to calculate the total energy including the zero point energy, vibrational frequencies, and rotational constants for all structures. For each ion, the lowest energy structures were chosen as precursors for subsequent pathways. For example, for the determination of C16H9+, there were three likely structures (shown in Figure 1) corresponding to the three

plotted as a function of photon energy. By using RRKM, the breakdown curves were fit, and the energy of activation energy (E0) and entropy of activation (ΔS‡) were extracted. These results will be compared with those of Ling et al., who used tunable VUV radiation but no internal energy selection, in combination with a residence time up to 10 ms to study the dissociative photoionization of pyrene.6 Theoretical structures have been calculated based on energetic and chemical intuition garnered from our previous work on naphthalene.5 These calculated structures are combined with the PIRENEA results to give a relevant picture of the intermediates and products pyrene forms as the fragmentation progresses through 22 separate reactions.

2. EXPERIMENTAL METHODS Pyrene was obtained from Sigma−Aldrich and used without further purification. In the iPEPICO experiments, the solid sample was introduced directly, while in the PIRENEA experiments, the sample was first dissolved in toluene (1 mg/ mL), and then, three drops were applied to the sample boat and allowed to dry for several hours before being introduced into the instrument. 2.1. PIRENEA. The PIRENEA setup located at IRAP in Toulouse is a home-built Fourier transform ICR mass spectrometer with the additional characteristics of a cold environment generated by a set of cryogenic shields to approach the conditions of low temperatures (10−50 K) and extreme isolation that prevail in the interstellar medium. The full experimental setup has been described previously.7 In short, PAH ions are produced through a laser desorption/ionization technique on a solid sample using the fourth harmonic, λ = 266 nm, output of a Nd/YAG laser (Minilite II, Continuum). The ions are then trapped through the conjugated action of an axial magnetic field (5 T) and a static electrical potential. The 13C isotopic species are selectively ejected by providing enough energy into their cyclotron motion. The experiments start then with pure 12C16H10+ ions. For the reported studies, a 150 W Xe arc lamp (Photomax model 60100, Oriel) was employed to induce photodissociation of the studied ions. This method is not used to simulate the interstellar radiation field, but as described earlier by Boissel et al.,8 this is an easy way to study dissociation close to the threshold by building internal energy into the ions, which is achieved by the competition between multiple photon absorption and radiative cooling. The detection scheme consists of exciting the trapped ions with a dipolar radio frequency signal containing the cyclotron frequencies of the ions. This will drive them into a coherent motion leading to an image current that is Fourier transformed to retrieve the mass/charge ratio of the observed ions. The experimental procedure followed in this work consists of production and isolation of the ions of interest, irradiation with the light source, and mass analysis. 2.2. 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.9−11 The neutral gas molecules are ionized by monochromatic VUV synchrotron radiation. The ions are then directed into 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 nonzero

Figure 1. Molecular structure of pyrene with the three unique hydrogen sites indicated in dark gray. 7825



Article

continuous irradiation, most of the peaks are not primary daughter species but produced by sequential fragmentation. Dehydrogenation of the parent ion by up to six hydrogen atoms is observed together with the formation of smaller hydrocarbons and carbon cluster ions starting with C14+. Apart from dehydrogenated species of the parent ion, large carbon clusters, including C14+, C12+, and C10+, appear to be major peaks in the mass spectrum. Two methods have been employed to characterize the fragmentation cascade. In the first method, only the species of interest are kept in the ICR cell by selectively ejecting the other species. The remaining ions are then subjected to irradiation by the Xe lamp to determine their fragmentation pattern. An example of such an experiment is presented in Figure 3, where fragment cations with m/z = 200

distinct hydrogen sites, with the lowest energy isomer used for subsequent steps. The vibrational frequencies and rotational constants were used as inputs into the RRKM calculations (for C16H10, C16H10+•, and C16H9+). 2.3.2. RRKM Calculations. For some of the dissociation channels, it was possible to fit experimental data with RRKM. In order to determine their 0 K activation energy (E0) and the entropy of activation (ΔS‡) from the experimental breakdown curves, the dissociation rate for each channel, k(E), needs to be determined using the following equation:14 k(E) =

σN ‡(E − E0) hρ(E)

(1)

where σ is the reaction degeneracy, h is Planck’s constant, N‡(E − E0) is the number of states of the transition state at a given energy (E − E0), and ρ(E) is the density of states of the reactant ion at that same energy (E). For the loss of H from pyrene (yielding C16H9+), σ has a value of 10, since calculations show that losing any of the hydrogen atoms is equivalent. For the next H loss to C16H8+•, the symmetry value used is 2 as it will be shown that, at this stage, only hydrogen atoms neighboring the absent hydrogen site can be considered. As in our previous work,5 the iPEPICO program of Sztáray et al. was used for fitting the iPEPICO breakdown curves for the first two consecutive hydrogen losses, C16H9+ and C16H8+•.15 In the iPEPICO TOF distributions, the experimental rate information is contained in the daughter ion peak shapes. Since the effect of the peak broadening due to metastable H loss is negligible compared with the instrumental peak width, the experiments did not yield direct rate information. This is in contrast with previous PAH experiments, in which experimental rates to C2H2 loss daughter ions, products of parallel dissociation channels, could be used to determine relative rates for m/z = 1 and 2 losses, as well.5 As a result, activation energies and entropies have a much higher uncertainty in the present case.

Figure 3. (a) Isolation of C16H8+ ions (m/z = 200) after ejection of all other species. (b) PIRENEA mass spectrum of C16H8+• after 6 s of irradiation with the Xe arc lamp.

(C16H8+•) are isolated and subjected to Xe lamp irradiation. In the second method, specific fragment ions are continuously ejected during lamp irradiation eliminating daughter species produced by their fragmentation. Table 1 shows all cationic

3. RESULTS AND DISCUSSION 3.1. PIRENEA. Figure 2 displays the photodissociation spectrum recorded for the pyrene cation after 10 s of irradiation with the Xe arc lamp. As the experiment is performed under

Table 1. List of the Fragments Detected in the Photodissociation Cascade of Pyrene Cation

Figure 2. PIRENEA mass spectrum of pyrene cation 12C16H10+• after 10 s of irradiation with the Xe arc lamp. n = 8, 10, 12, and 14 is the total carbon number per molecule. 7826

m/z

ion

m/z

ion

201 200 199 198 197 196 176 174 173 172 171 170 169 168 152 150 149

C16H9 C16H8 C16H7 C16H6 C16H5 C16H4 C14H8 C14H6 C14H5 C14H4 C14H3 C14H2 C14H C14 C12H8 C12H6 C12H5

148 146 144 132 123 122 121 120 100 99 98 97 85 84 74 62 61

C12H4 C12H2 C12 C11 C10H3 C10H2 C10H C10 C8H4 C8H3 C8H2 C8 H C7 H C7 C6H2 C5H2 C5 H



Article

publication.16 Finally, the channel of diacetylene loss was found to be a minor channel and only observed for C16H10+•, C16H8+•, and C16H4+•. Branching ratios could be unambiguously derived for the first two only. 3.2. Calculations. The next step was to determine the likely structures of species present in the fragmentation map focusing the analysis down to C12H2+• for one path and C14+ for another path. Figure 4 reports all the species involved in the map and their associated structure numbers. All structures are shown in Figure 5a−c, while the calculated dissociation energies can be found in Table 2. 3.2.1. Dehydrogenation of C16Hn+. The main “trunk” of the dissociation tree for pyrene consists of consecutive hydrogen losses. There are nine potential reactions that make up the C16Hn+ pathway and are listed with structure numbers in parentheses as follows:

species, with their respective chemical formulas, detected in the study of the pyrene cation dissociation cascade. Three main channels of dissociation have been determined for pyrene cations, hydrogen loss, acetylene loss, and diacetylene loss. Hydrogen loss is the major dissociation channel and can involve both H and H2 channels for a given species. Under continuous irradiation of the Xe lamp, the production of C16H8+•, for instance, could be due to the loss of H2 from the parent C16H10+ or to H loss from C16H10+ followed by H loss from C16H9+, the latter path being referred to as sequential H loss. To disentangle the two paths, C16H9+ product is extracted continuously from the cell during irradiation to preclude any dissociation path that would involve this intermediate. With this continuous selective ejection activated, no C16H8+• is detected: sequential H loss is therefore the dominant process in the production of C16H8+•. Although this was not tested for all species of the fragmentation cascade, we expect dehydrogenation to proceed by sequential H loss rather than by H2 loss in the conditions of the PIRENEA experiments, that is, close to the dissociation threshold. Branching ratios of the three dissociation channels (H, C2H2, C4H2) were determined for pyrene cations and some fragments. This is performed by measuring the peak intensity of the fragments related to each dissociation channel. Among the results summarized in Figure 4, one important observation is that some fragments show significant acetylene group loss branching ratios. For instance, C16H5+ has two dissociation channels with equal branching ratios, hydrogen loss and acetylene loss. The study of this branching ratio on a larger sample of parent PAH species is the subject of another

C16H10+•(M+•) → H + C16H 9+(1)

(R1)

C16H 9+(1) → H + C16H8+•(2)

(R2)

C16H10+•(M+•) → H 2 + C16H8+•(2)

(R3)

+•

+

C16H8 (2) → H + C16H 7 (3)

(R4)

C16H 7+(3) → H + C16H6+•(4)

(R5)

C16H8+•(2) → H 2 + C16H6+•(4)

(R6)

C16H6+•(4) → H + C16H5+(5)

(R7)

C16H5+(5) → H + C16H4 +•(6)

(R8)

+•

+•

C16H6 (4) → H 2 + C16H4 (6)

(R9)

Reaction R1 corresponds to the first H loss, for which only three sites need to be considered, as shown in Figure 1. The calculations resulted in a maximum difference in energy of only 0.25 eV between the sites. This may be an indication that, because of the aromaticity of the molecule, all the H atoms are equivalent at this stage. On the basis of the end position being lowest in energy, C16H9+ is depicted as 1 with a dissociation energy of 5.10 eV. For R2, it was determined that the initial position has a more significant effect on the energy, with the neighboring site being 1 eV lower in energy than either of the other two possible locations. Therefore, C16H8+• was assigned as 2 with a dissociation energy of 3.50 eV. We have seen that the dominant peak M − 2 in the PIRENEA results (Figure 2) is produced by consecutive H loss (R1 followed by R2) and not by H2 loss (R3). Still, there have been other studies where R3 is observed.6 The assignment for R4 follows the same trend as for R2 with the H atom being removed from the already dehydrogenated ring resulting in 3 with a dissociation energy of 4.48 eV. For R5, the results of dehydrogenation continue along this trend where the empty sites will pair. There is a slight change in that the pre-existing empty sites from 3 undergo hydrogen scrambling to get the resulting structure 4 with two dehydrogenated carbons on two neighboring rings. The resulting C16H6+• has an energy of dissociation of 4.11 eV, slightly lower than that of the R4 reaction. The next hydrogen loss R7 has no empty site with which to pair, and therefore, there is an increase in the energy requirement, with a dissociation energy of 4.45 eV similar to that of the R4 reaction and resulting in structure 5.

Figure 4. Schematic of fragmentation map of pyrene cations. The letters (A−C) are to relate position of structure maps on the overall map in Figure 5. Solid arrows indicate H loss, dashed arrows represent C2H2 loss, and dash/dot arrows are C4H2 loss. The numbers in parentheses are the structure number in order to link the structure image to the text. The dashed lines around structures 2 and 4 represent the division of the map for the more detailed structure diagrams shown in Figure 5a−c. Experimentally measured branching ratios are noted above reaction arrows in %. Not shown is the formation of C12H8+• (C4H2 loss) from C16H10+• with a branching ratio of 0.2%. 7827



Article

Table 2. Calculated Dissociation Energies for Reactions R1−R22 in the Fragmentation Map of Pyrene Cations no. R1 R2 R3 R4 R5 R6 R7 R8 R9 R10 R11

reaction C16H10+• → C16H9+ + H C16H9+→ C16H8+• + H C16H10+•→ C16H8+• + H2 C16H8+•→ C16H7+ + H C16H7+→ C16H6+• + H C16H8+•→ C16H6+• + H2 C16H6+•→ C16H5+ + H C16H5+→ C16H4+• + H C16H6+•→ C16H4+• + H2 C16H10+•→ C14H8+• + C2H2 C14H8+•→ C12H6+• + C2H2

ΔE (eV)

no.

reaction

5.10

R12

3.50

R13

4.12

R14

4.48

R15

4.11

R16

4.09

R17

4.45

R18

4.08

R19

4.03

R20

6.02

R21

3.58

R22

C16H8+•→ C14H6+• + C2H2 C14H6+•→ C12H4+• + C2H2 C16H5+→ C14H3+ + C2H2 C16H4+•→ C14H2+• + C2H2 C16H8+•→ C12H6+• + C4H2 C16H4+•→ C12H2+• + C4H2 C14H6+•→ C14H5+ + H C14H5+→ C14H4+• + H C14H3+→ C14H2+• + H C14H2+•→ C14H+ + H C14H+→ C14+• + H

ΔE (eV) 6.00 3.61 3.94 3.35 9.64 8.75 4.82 4.95 3.49 3.29 3.58

This is in direct agreement with literature, where the alternate structure was determined to be the most stable for C16H4+•.17 When each steps of the two dissociation pathways (H2 versus H loss) are compared, there is a clear difference between them, as seen in Figure 6. For the H loss channel, there is a clear

Figure 6. Dissociation energies for dehydrogenation of C16Hn+ species. Ling et al. points correspond to ref 6.6 Error bars correspond to error of iPEPICO values.

Figure 5. (a) Fragmentation map with structures and dissociation energies for pyrene fragments down to C16H8+•; (b) fragmentation map with structures and dissociation energies for pyrene fragmentation from C16H8+• to C16H6+•; (c) fragmentation map with structures and dissociation energies for pyrene fragmentation from C16H6+• to C14+•.

alternating pattern in energies when losing an odd or even numbered hydrogen, with the odd H loss being more favorable. This feature is expected as the lowest energy structures with an even number of empty hydrogen sites always have them paired. This trend is less pronounced as the molecule becomes more dehydrogenated. In the case of H2 loss, the trend is very linear. There are no alternating energies. The idea of paired “holes” explains this as well. When two hydrogens are lost successively, this pairing happens automatically; therefore, there should be

The final reaction for the dehydrogenation “trunk” would be R8 resulting in the formation of C16H4+•, which has the alternate structure 6 proposed by Lee et al.17 While the alternate structure was calculated for all fragments from C16H7+ onward, this is the only one for which the dissociation energy, 4.08 eV, is significantly lower than for the pyrene structure. 7828



Article

(3.94 eV compared to 4.08 eV). Reaction R14 is also the first instance where all polycyclic structure is lost, yielding structure 11. This trend of facile acetylene loss continues with R15 when we no longer see any dehydrogenation and the energy of acetylene loss is even lower with only 3.35 eV and the resulting structure 12. Unlike for the dehydrogenation reactions, there is no obvious trend one can derive from acetylene loss alone. The trend observed has more to do with the competition between acetylene loss and dehydrogenation. As seen in Figure 7, at

no variance. The absence of an observable H2 loss suggests that this reaction has a reverse energy barrier and/or an unfavorable entropy of activation, making it kinetically noncompetitive on the experimental time scale. The competition of the H/H2 channels in the dehydrogenation of PAHs is debated in the literature on the basis of energetic calculations,18,19 but addressing completely this question would require molecular dynamics simulations, which have still to be done. Finally, the comparison of our values with those derived for R1−R3 by Ling et al.6 can be seen in Table 3; these values fall within acceptable error, 0.5 eV for R1 and approximately 0.6 eV for R2 and R3. Table 3. Comparison of E0 (eV) and Δ‡S1000K (J K−1 mol−1) Values E0a Δ‡S1000Ka calcd E0b lit E0c lit Δ‡S1000Kc

R1

R2

R3

5.4 ± 1.2 57 ± 89 5.1 4.6 44.8

3.3 ± 1.1 16 ± 84 3.5 4.1 55.6

− − 4.12 3.52 −53.1

a

Fit experimental iPEPICO data from this work. bCalculated dissociation energies; this work assuming there is no reverse barrier in the dissociation, i.e., E0 = ΔE (Table 2). cValues from Ling et al.6

3.2.2. Acetylene Loss. The next category of fragmentation to be discussed is the acetylene loss. In this work, there are six primary or consecutive fragments formed by acetylene losses: C16H10+•(M+•) → C14 H8+•(7) + C2H 2

(R10)

C14 H8+•(7) → C12H6+•(8) + C2H 2

(R11)

C16H8+•(2) → C14 H6+•(9) + C2H 2

(R12)

C14 H6+•(9) → C12H4 +•(10) + C2H 2

(R13)

C16H5+(5) → C14H3+(11) + C2H 2

(R14)

+•

+•

C16H4 (6) → C14H 2 (12) + C2H 2

Figure 7. Histogram comparing the energy required for different molecules to undergo dehydrogenation (light gray) and acetylene loss (dark gray). For C16H4+•, it should be noted that no dehydrogenation was observed.

early stages of dehydrogenation, there is very little competition between the two dissociations; in both cases, acetylene loss is at least 0.90 eV higher in energy. This changes dramatically at C16H5+, where acetylene loss competes with dehydrogenation. For C16H4+•, dehydrogenation is no longer observed, and acetylene loss is even lower in energy. In both instances of consecutive acetylene losses, the first acetylene loss is quite endoergic (around 6.00 eV), and the second acetylene loss has a drastically reduced dissociation energy of around 3.60 eV. This indicates that, once the pyrene structure is destroyed (structures 7 and 9), the molecule loses its planar structure and consequently the barrier to overcome to lose further hydrocarbon fragments is lowered. It seems that the first carbon loss is the main limiting step. All these theoretical insights provide a very nice rationale to account for the PIRENEA results, which led to the fragmentation cascade and branching ratios summarized in Figure 4. 3.2.3. Dehydrogenation of C14Hn+ Species. Similarly to the parent species, the fragments undergo various stages of dehydrogenation after acetylene loss. In the data presented here, there are two separate dehydrogenation channels starting from C14H6+• and C14H3+. The reactions are as follows:

(R15)

Reaction R10 is the first acetylene loss, to make C14H8+•, which results in the nonplanar structure 7 with a dissociation energy of 6.02 eV. This dissociation energy is larger than the competing reaction R1 (5.10 eV) and in agreement with the branching ratio of 1.3%/98.5% for R10 compared to R1 measured at low energies in the PIRENEA setup (Figure 4). Acetylene loss, due to this typical energy difference, does not appear to be able to compete with dehydrogenation on a large scale until the molecule is quite dehydrogenated as discussed further for the 50% branching ratio of the R14 channel. C14H8+• undergoes another acetylene loss (R11) that has a much lower energy of dissociation of 3.58 eV resulting in structure 8. The next case of acetylene loss, which also happens to consist of two consecutive losses, occurs at C16H8+• for R12 and R13. The fragment structures calculated for these reactions, 9 and 10, respectively, look very similar to 7 and 8; they were also determined to have very similar energetics with R12 having a dissociation energy of 6.00 eV and R13 a dissociation energy of 3.61 eV. Reaction R14, which is the acetylene loss from C16H5+, shows a very different behavior. The loss of acetylene was calculated to be comparable in energy to dehydrogenation, R8

C14 H6+•(9) → H + C14 H5+(14) +

+•

+

+•

C14 H5 (14) → H + C14H4 (15)

7829

(R18) (R19)

C14 H3 (11) → H + C14H 2 (12)

(R20)

C14 H 2+•(10) → H + C14H+(16)

(R21)


The Journal of Physical Chemistry A C14 H+(16) → H + C14 +•(17)

Article

(R22)

+•

Starting from C14H6 , the molecule undergoes two consecutive dehydrogenation processes. Reaction R18 is the first reaction, which results in 14 with a calculated dissociation energy of 4.82 eV. This is substantially higher in energy than the competing reaction R13 by 1.19 eV; therefore, this is likely a minor reaction channel. Reaction R19, which results in 15 gives a similar energy of 4.95 eV. Reactions R20−R22 result in a separate reaction channel leading to the C14+• carbon cluster. Reaction R20 consists of the first dehydrogenation of this reaction with structure 12 accessed by an energy of 3.49 eV. It should be noted that this is the same structure resulting from R15. This could be an indication that it does not matter in which order the dehydrogenation and loss of acetylene occurs relative to each other. Reactions R21 and R22 continue to lose hydrogen atoms without drastic change in the carbon backbone, yielding structures 16 and 17, respectively. The dissociation energies are very similar as well, corresponding to 3.29 and 3.58 eV, respectively. The C14Hn+ fragments show that consecutive dehydrogenation reactions progress with very little change in the dissociation energies of each subsequent loss until complete dehydrogenation. This explains why the dominant peaks of the mass spectrum in Figure 2 are the carbon clusters. 3.2.4. Diacetylene Loss. There are two remaining reactions that have yet to be discussed: C16H8+•(2) → C12H6+•(8) + C4 H 2

(R16)

C16H4 +•(6) → C12H 2+•(13) + C4H 2

(R17)

Figure 8. Experimental iPEPICO breakdown curve for the pyrene radical cation over the photon energy range 17−22 eV. The reaction numbers for each product channel has been included in parentheses. Calculated fits are overlaid. Derived energetic and entropic parameters can be found in Table 3.

methods discussed; for this reason, R3 is also listed. All the iPEPICO data were fit by RRKM theory. For R1, the fit values obtained were E0 = 5.4 ± 1.2 eV and ΔS‡ = 57 ± 89 J K−1 mol−1. The activation energy is slightly higher than the calculated value (ΔE = 5.1 eV) but in agreement taking the uncertainty associated with the iPEPICO data. This experimental error is as well large enough to account for the 0.8 eV of difference with the calculated value reported by Ling et al. (4.60 eV). The entropy of activation can also be compared, with fairly good agreement between the two; the literature reported the ΔS‡ value to be 44.7 J K−1 mol−1.6 Reaction R2 was also compared between all three methods. The iPEPICO fitting resulted in an E0 value of 3.04 ± 1.09 eV and a ΔS‡ value of 16 ± 84 J K−1 mol−1. When comparing with the literature, the entropy of activation is quite different from the value reported (55.6 J K−1 mol−1), same with the activation energy (4.1 eV) though both are bracketing the calculated result (3.5 eV).6 This discrepancy is again observed when comparing the calculated and literature values for R3, with a literature value of 3.52 eV while the calculated is 4.12 eV. Overall, the comparison between experimental and theoretical values shows that there is no significant reverse barrier in the dissociation, that is, E0 = ΔE (Table 2). The values we could derive for ΔS‡ are unfortunately not significant enough to further constrain the reaction mechanisms.

both of which result from the loss of diacetylene. There was another case of C4H2 loss originating from the pyrene radical cation (to generate C12H8+•). This channel only accounted for 0.2% of the parent loss, as seen in Figure 4, and as a result was not included in the cascade of calculated structures presented here. These reactions have much higher dissociation energies than all others observed that lends some doubt to how prevalent they are, especially for R16 where this same fragment is generated via R10 and R11. The final reaction channel to be discussed is R17, which is the only route for the generation of C12H2+• resulting in 13. The energy for this channel is also quite high, 8.75 eV, similar to the other diacetylene loss. This indicates how unfavorable this channel is for pericondensed PAHs, as compared to catacondensed species, such as naphthalene.5 3.3. iPEPICO. iPEPICO data was collected over a large photon energy range, from 17 to 22 eV. Three reaction channels were observed; R1 and R2 can be seen in Figure 8. Reaction R10, while present, consists of less than 10% of the fragment ions in agreement with the PIRENEA results (Figure 4). There is no way to restrict the fitting parameters at such low daughter ion intensity, and therefore, the curve can be fit with too wide a range of values for the fit to be meaningful. Because of the low density of experimental data points, there is a wide range of possible values for all fits, though the ones shown in Table 3 yielded the best fit by a small margin. The theoretical fitting of the breakdown curve is shown in Figure 8. The estimated values for E0 and ΔS‡ are listed in Table 3 along with the calculated values for those reaction channels and the literature values from Ling et al.6 This table compares reaction channels that are common with at least two of the three

4. CONCLUSIONS The unimolecular dissociation of the pyrene radical cation has been investigated using the experimental techniques of PIRENEA and iPEPICO, as well as by computation. A total of 22 reactions have been investigated by PIRENEA, and 17 unique structures were calculated to rationalize the experimental observations. The dominant fragments occur from a series of consecutive H atom loss from the pyrene radical cation until C16H4+•. There is an alternating pattern in the dissociation energy due to the pairing of empty H sites when possible. The pyrene structure is well preserved until this final structure C16H4+• at which point the remaining hydrogen atoms are contained on a single benzene ring, and the remaining bare carbons form a single ring. Once the PAH structure is lost, it is 7830



Article

never restored. When C16H4+• undergoes acetylene loss, this structure and all subsequent structures are a simple ring with any remaining hydrogens on adjacent carbon atoms. The other fragmentation routes stem from the few paths resulting from acetylene loss. As the molecule gets more dehydrogenated, the acetylene loss channels become more energetically competitive. Reactions R1 and R2 were investigated further by comparison between the iPEPICO results, calculations as well as literature values. The E0 values for the reactions are 5.4 ± 1.2 and 3.3 ± 1.1 eV, respectively, which matches calculations quite well (5.10 eV for R1 and 3.50 eV for R2). Finally, this work provides further insights into the chemical evolution of PAHs in astrophysical environments. The photoprocessing of PAHs has been proposed as an explanation to account for the increased abundance of small hydrocarbons in UV-irradiated regions of the interstellar medium. 20 Our results provide a first quantitative view on the fragmentation cascade of pyrene and its potential to inject C2H2 and carbon clusters into the chemistry of these environments.

■

(3) 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. (4) West, B.; Joblin, C.; Blanchet, V.; Bodi, A.; Sztáray, B.; Mayer, P. The Dynamics of Hydrogen and Methyl Radical Loss from Ionized Dihydro-Polycyclic Aromatic Hydrocarbons: A Tandem Mass Spectrometry and Imaging-PEPICO Study of Dihydronaphthalene and Dihydrophenanthrene. J. Phys. Chem. A 2014, 118, 1807−1816. (5) 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. (6) Ling, Y.; Gotkis, Y.; Lifshitz, C. Time-Dependent Mass Spectra and Breakdown Graphs. 18. Pyrene. Eur. J. Mass Spectrom. 1995, 1, 41−49. (7) Useli-Bacchitta, F.; Bonnamy, A.; Malloci, G.; Mulas, G.; Toublanc, D.; Joblin, C. Visible Photodissociation Spectroscopy of PAH Cations and Derivatives in the PIRENEA Experiment. Chem. Phys. 2010, 371, 16−23. (8) Boissel, P.; de Parseval, P.; Marty, P.; Lefèvre, G. Fragmentation of Isolated Ions by Multiple Photon Absorption: A Quantitative Study. J. Chem. Phys. 1997, 106, 4973−4984. (9) 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., Sect. A 2009, 610, 597−603. (10) 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. (11) 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. (12) Bodi, A.; Hemberger, P. Imaging Breakdown Diagrams for Bromobutyne Isomers with Photoelectron-Photoion Coincidence. Phys. Chem. Chem. Phys. 2014, 16, 505−515. (13) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; et al. Gaussian 09, Revision D.01; Gaussian Inc.: Wallingford, CT, 2009. (14) Baer, T.; Hase, W. L. Unimolecular Reaction Dynamics, Theory and Experiments; Oxford University Press: New York, 1996. (15) Sztáray, B.; Bodi, A.; Baer, T. Modeling Unimolecular Reactions in Photoelectron Photoion Coincidence Experiments. J. Mass Spectrom. 2010, 45, 1233−1245. (16) Sabbah, H.; Useli-Bacchitta, F.; Joblin, C. Photodissociation Pathways of Isolated Polycyclic Aromatic Hydrocarbon cations in the PIRENEA setup. Submitted for publication. (17) Lee, S.; Gotts, N.; von Helden, G.; Bowers, M. T. Structures of CnHx+ Molecules for n ≤ 22 and x ≤ 5: Emergence of PAHs and Effects of Dangling Bonds on Conformation. J. Phys. Chem. A 1997, 101, 2096−2102. (18) Jolibois, F.; Klotz, A.; Gadéa, F.-X.; Joblin, C. Hydrogen Dissociation of Naphthalene Cation: A Theoretical Study. Astron. Astrophys. 2005, 444, 629−634. (19) Paris, C.; Alcamí, M.; Martín, F.; Díaz-Tendero, S. Multiple Ionization and Hydrogen Loss from Neutral and Positively-Charged Coronene. J. Chem. Phys. 2014, 140, 204307. (20) Pety, J.; Teyssier, D.; Fossé, D.; Gerin, M.; Roueff, E.; Abergel, A.; Habart, E.; Cernicharo, J. Are PAHs Precursors of Small Hydrocarbons in Photo-Dissociation Regions? The Horsehead Case. Astron. Astrophys. 2005, 435, 885−899.

ASSOCIATED CONTENT

S Supporting Information *

Table S1 containing vibrational frequencies used in the RRKM calculations described in this paper, and the complete citation for ref 13. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Phone: 33-561-558-601; e-mail: [email protected]. Present Address

⊥ Valérie Blanchet: CELIA, Univ. Bordeaux-CNRS, Talence, France.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work has been supported by the French National Program Physique et Chimie du Milieu Interstellaire, the program “Molécules et grains: du laboratoire à l’Univers” of the MidiPyrénées Observatory and by the University Paul Sabatier. The authors are grateful to Michel Armengaud and Patrick Frabel for their technical involvement on the PIRENEA setup. B.W. acknowledges the support of the French Agence Nationale de la Recherche (ANR), under grant GASPARIM “Gas-phase PAH research for the interstellar medium”, ANR-2010-BLANC0501. P.M.M. thanks the Natural Sciences and Engineering Research Council of Canada for continuing financial support. 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 funding from the European Community’s Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 226716.

■

REFERENCES

(1) Joblin, C.; Tielens, A. G. G. M. PAHs and the Universe: A Symposium to Celebrate the 25th Anniversary of the PAH Hypothesis, EAS Publications Series; Cambridge University Press: Cambridge, U.K., 2011; Vol. 46. (2) Montillaud, J.; Joblin, C.; Toublanc, D. Evolution of Polycyclic Aromatic Hydrocarbons in Photodissociation Regions. Astron. Astrophys. 2013, 552, A15. 7831


Photodissociation of Pyrene Cations: Structure and ... - ACS Publications

Recommend Documents