Dissociation of the Anthracene Radical Cation: A Comparative Look at

Article pubs.acs.org/JPCA

Dissociation of the Anthracene Radical Cation: A Comparative Look at iPEPICO and Collision-Induced Dissociation Mass Spectrometry Results Brandi West,† Alicia Sit,† Sabria Mohamed,† Christine Joblin,‡,§ Valerie Blanchet,∥ Andras Bodi,⊥ 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 ∥ Université de Bordeaux - CNRS - CEA, CELIA, F33405 Talence, France ⊥ Molecular Dynamics Group, Paul Scherrer Institut, Villigen 5232 Switzerland ‡

S Supporting Information *

ABSTRACT: The dissociation of the anthracene radical cation has been studied using two different methods: imaging photoelectron photoion coincidence spectrometry (iPEPCO) and atmospheric pressure chemical ionization−collision induced dissociation mass spectrometry (APCICID). Four reactions were investigated: (R1) C14H10+• → C14H9+ + H, (R2) C14H9+ → C14H8+• + H, (R3) C14H10+• → C12H8+• + C2H2 and (R4) C14H10+• → C10H8+• + C4H2. An attempt was made to assign structures to each fragment ion, and although there is still room for debate whether for the C12H8+• fragment ion is a cyclobuta[b]naphthalene or a biphenylene cation, our modeling results and calculations appear to suggest the more likely structure is cyclobuta[b]naphthalene. The results from the iPEPICO fitting of the dissociation of ionized anthracene are E0 = 4.28 ± 0.30 eV (R1), 2.71 ± 0.20 eV (R2), and 4.20 ± 0.30 eV (average of reaction R3) whereas the Δ‡S values (in J K−1 mol−1) are 12 ± 15 (R1), 0 ± 15 (R2), and either 7 ± 10 (using cyclobuta[b]naphthalene ion fragment in reaction R3) or 22 ± 10 (using the biphenylene ion fragment in reaction R3). Modeling of the APCI-CID breakdown diagrams required an estimate of the postcollision internal energy distribution, which was arbitrarily assumed to correspond to a Boltzmann distribution in this study. One goal of this work was to determine if this assumption yields satisfactory energetics in agreement with the more constrained and theoretically vetted iPEPICO results. In the end, it did, with the APCI-CID results being similar.

1. INTRODUCTION

There has been some work previously conducted on the anthracene radical cation and its isomer, ionized phenanthrene. Ling et al. completed a thorough study of their fragmentation, which included both experimental results outlining the dissociation and theoretical calculations to determine the molecular structure for the fragment ions.4,5 In one study, photoionization results were used to determine the main dissociation channels for the two C14H10+• ions, H-loss, 2H and H2 loss, acetylene loss as well as larger organic neutral losses (C3H3 and C4H2), and these reactions were also observed as sequential losses from larger fragments. They were able to measure 0 K activation energies and entropies of activation for all reactions using RRKM fitting.4 Mass-analyzed ion kinetic energy spectrometry (MIKES) results were also discussed in this paper, primarily focusing on the dished top peak shape observed for C12H8+•. This result led to the conclusion that the

Polycyclic aromatic hydrocarbons (PAHs) have been under investigation for quite some time as molecules of interstellar interest. This interest has spurred many laboratory studies on their formation and stability in simulated interstellar (and circumstellar) environments to attempt to get a better understanding of their potential roles in astrochemistry.1 Our laboratories have contributed to this discussion in a recent paper dealing with the photodissociation of the naphthalene radical cation by imaging photoelectron−photoion coincidence spectroscopy (iPEPICO) and tandem mass spectrometry, which established the fragmentation pathways and energetics for this species.2 The next molecules studied were variants of naphthalene, 1,2-dihydronaphthalene and 9,10-dihydrophenanthrene, to determine the effect of additional hydrogen atoms on the dissociation energetics.3 The work presented here tackles the next acene in size after naphthalene, anthracene, which possesses a cata-condensed structure consisting of three fused benzene rings. © 2014 American Chemical Society

Received: June 2, 2014 Revised: September 22, 2014 Published: September 22, 2014 9870

dx.doi.org/10.1021/jp505438f | J. Phys. Chem. A 2014, 118, 9870−9878

The Journal of Physical Chemistry A

Article

the CID breakdown curves to to see if APCI-CID is consistent with iPEPICO for studying these ions, and the suitability of the internal energy distribution approximation being made in our model, the basic question being: just how similar to a Boltzmann distribution is the postcollision internal energy distribution of the ions in this study? Numerous qualitative and quantitative approaches have been described to address this issue depending on the nature of the experiment.12 The most successful have been for slow-heating methods (such as CID performed in ion traps),13−20 CID and surface-induced dissociation in FT-ICR21−27 and single− collision threshold measurements.28−30 A number of studies have explored the relationship between Teff and Ecom under multiple collision conditions in a quadrupole collision cell.31−34 Although there is no evidence for the linear scaling of the effective temperature with Ecom shown in eq 1, it may still be a reasonable first-order approximation thanks to the small absolute range of Ecom exhibited in the breakdown diagram. Knyazev and Stein35,36 have recently demonstrated that postcollision ion internal energy distributions can indeed be non-Boltzmann in character. In their reported case for the small benzylammonium ion, the P(E) of the ions at Elab = 10 eV (Ecom = 2.7 eV) appears more like a pure Boltzmann distribution shifted up along the internal energy axis (Figure 6 in ref 35). The discrepancy between the modeled distribution and a thermal distribution increased with increasing collision energy. This would clearly draw into doubt our use of a thermal distribution to approximate the P(E) of ionized anthracene. In the end, this assumption in our model will be tested by determining if a suitable fit to the experimental data can be obtained, and if that fit results in acceptable kinetic values of E0 and Δ‡S, when compared to the iPEPICO results.

acetylene loss channel is associated with a tight transition state and a large reverse energy barrier, which would explain the kinetic energy release observed. Another observation was that both anthracene and phenanthrene possessed this same dished top peak, the authors describing them as indistinguishable, and therefore, it was assumed likely that ionized anthracene first isomerized to ionized phenanthrene prior to C2H2 loss. Their other study focused on elucidating the structure of the C12H8+• fragment through computational modeling.5 Six different structures were proposed, their threshold dissociation energies were calculated at two different levels of theory (B3LYP/ccpVDZ and B3PW91/cc-pVDZ), and the energies were then compared to the experimental results. On the basis of this comparison, it was determined that the likely structure of C12H8+• would be that of ionized biphenylene, which supports the assumption that both anthracene and phenanthrene radical cations would have to isomerize to a common structure, in this case ionized phenanthrene, prior to dissociation. On the basis of the results found by Ling and Lifshitz, Johansson et al. did further high energy collision-induced dissociation (CID) experiments as well as calculations to determine the reaction surface starting from the anthracene radical cation to three of the structures predicted: taking acetylene from an outer ring to form ionized ethynylnaphthalene, the formation of ionized biphenylene via phenanthrene ions, and also calculating the formation of acenaphthylene ions (the lowest energy structure from Ling et al.) from ionized phenanthrene.6 Johansson et al. stated that that because acenaphthylene had the lowest reaction barrier, it should be the most heavily weighted structure for the appearance energy of this reaction, but all the structures proposed were close enough in energy that they might also contribute. As the molecular weight of the PAHs increases, their vapor pressures generally decrease, making it progressively more difficult to explore the dissociative photoionization of large PAHs with techniques such as iPEPICO, though molecules as large as C42H18 have been studied by high temperature evaporation.7 Atmospheric pressure chemical ionization (APCI) is an alternative approach to the production of gasphase ions of nonpolar compounds such as anthracene. When APCI is coupled with a tandem mass spectrometer, energy resolved collision-induced dissociation mass spectrometry can be employed, and modeled, to derive similar information: fragmentation pathways and unimolecular reaction energetics and entropies, by modeling the experimental data with statistical rate theories such as RRKM theory.8−11 Once the RRKM k(E) rate curves have been obtained, branching ratios can be calculated as a function of ion internal energy. Our previous work modeling energy-resolved CID data employed a simple model in which the postcollision ions are assigned an effective temperature depending on the center-of-mass collision energy, and thus a “thermal” internal energy distribution, according to the relationship: Teff = Ti + αEcom

2. EXPERIMENTAL SECTION Anthracene was purchased from Sigma-Aldrich (Sigma-Aldrich, Oakville Canada) and was used without any further purification in MIKES and iPEPICO experiments. In APCI experiments, anthracene was dissolved in chloroform to produce a 100 μg/ mL solution. 2.1. MIKES. All mass-analyzed ion kinetic energy spectrometry (MIKES) experiments were performed on the modified VG-ZAB three-sector (BEE) instrument (VG Analytical, Manchester, U.K.) described previously.2 In brief, anthracene was introduced into the EI source via a direct insertion probe. A pressure of ∼2 × 10−6 Torr, measured by an ion gauge positioned in the source, was obtained for all MIKES experiments. As indicated, anthracene radical cations were generated using electron impact and then ejected from the source with acceleration voltages ranging from 6 to 8 kV, depending on the experiment. The ion of interest is selected by mass in the magnetic sector and directed through a field free region toward the first electrostatic analyzer. While drifting, the ions with sufficient internal energy will spontaneously dissociate and the resulting ion fragments and remaining anthracene radical cations will be detected on the basis of their kinetic energy. 2.2. iPEPICO. The imaging photoelectron photoion coincidence (iPEPICO) spectrometry experiments were conducted on the VUV beamline of the Swiss Light Source (SLS).37−39 In the experiment, anthracene is photoionized by monochromatic synchrotron radiation. The ions are then extracted toward a time-of-flight (TOF) mass spectrometer, and the ejected electrons are velocity map imaged on an

(1)

where Ti represents the initial temperature and α describes the relationship between the center-of-mass collision energy (Ecom) and the increase in the effective temperature (Teff). This is clearly the “Achilles heel” of the present model and thus we previously only attempted to derive relative energetics and entropies for the systems under study.8−11 One of our goals in this study is to compare the results of the modeling of iPEPICO fragmentation data with those from the modeling of 9871



Article

at 17 (as set in the Masslynx software) to provide baseline separation of all mass spectral peaks. Collision-induced dissociation (CID) was carried out using argon as a collision gas (argon at a pressure of ∼3.2 × 10−3 mbar) and an energy range (Elab) of 5−47 eV. During CID experiments, the second quadrupole resolution was held at 12 to increase sensitivity.

imaging multichannel plate (MCP) detector. The electrons are time and position stamped at the detector and the corresponding precursor or fragment photoions are detected in delayed coincidence. Threshold electrons account for the majority of the signal at the center of the MCP, whereas kinetic energy, “hot” electrons are detected according to their off-axis momentum. The mass spectrum based on electrons detected in a ring around the center spot is used to account for contamination from hot electrons. This is the result of kinetic energy electrons without any off-axis momentum component being detected in the center, threshold spot. The ring signal can then be subtracted from the center signal to obtain threshold photoionization mass spectra.40,41 The TOF mass spectrometer consists of two acceleration regions; one region possesses a low draw out potential that allows ions dissociating on the microsecond time scale to do so while being accelerated. The result is asymmetric time-of-flight peaks, which can be modeled to extract unimolecular decay rate constants as a function of photon energy.42 When multiple product ions are generated, they are all formed with the same overall effective decomposition rate constant of the precursor ion, and the fractional fragment ion abundances, i.e., the branching ratios, can be used to apportion absolute dissociation rates to each of the decay channels. This way, dissociation rate constants are obtained for all channels on the basis of the asymmetric peak profile of just one fragment ion. Due to the close proximity of M+•, (M − H)+, and (M − 2H)+•, and their respective 13C contributions, it is necessary to use a deconvolution procedure to determine their relative abundances to construct the breakdown diagrams. This process has been described in our previous paper,2 where a generic example of three signals, a, b, and c, with the same peak shapes and arrival times of t1, t2, and t3 respectively, has been assumed to derive the fractional abundances as follows:

3. COMPUTATIONAL METHODS 3.1. Ab Initio Calculations. All calculations used in the current work were conducted using the Gaussian 09 suite of programs43 using the B3-LYP/6-31G+(d) level of theory. Geometry optimizations, along with vibrational frequencies and rotational constants, were calculated for anthracene (M) and the anthracene radical cation (M+•), as well the observed fragments [M − H]+, [M − 2H]+•, and [M − C2H2]+•. The vibrational frequencies and rotational constants for M, M+•, and [M − H]+ were used in all the RRKM calculations. CBS-QB3 calculations were also carried out to shed more light on the C2H2-loss transition states. 3.2. RRKM Calculations. As in previous works, the 0 K activation energy (E0) and the entropy of activation (Δ‡S) are determined from fitting the experimental breakdown curve using RRKM theory, where the rate of each dissociation pathway, k(E), is calculated via the following formula,44 k(E) =

σN ‡(E − E0) hρ(E)

(5)

σ represents the reaction degeneracy, h is Planck’s constant, N‡(E − E0) is the number of internal states for the transition state at internal energy (E − E0) and ρ(E) is the density of states for the reactant ion at internal energy (E). For anthracene, σ values of 10 and 9 were used in the case of H loss and 2H loss as calculated energies indicate that all the hydrogens are equivalent; therefore, losing any one (or losing any second one after the initial loss) is equally likely. Choosing a σ value was more difficult in the case of acetylene loss as there are arguments for two different values, 1 versus 6. As detailed later in the Results and Discussion, both values were used to account for uncertainties regarding the reaction coordinate. Finally, for the loss of C4H2, a value of 2 was chosen as either end ring can be removed to give the naphthalene structure proposed. As stated in the previous section, vibrational and rotational information was extracted from the calculated structures; these values are then used to calculate the density of states for the reactant ions (M+• and [M − H]+) via the Beyer and Swinehart direct count algorithm.45 As in our previous work, the number of states of the transitions were approximated on the basis of the harmonic vibrational frequencies of the precursor ion. From the vibrational frequencies from the respective precursor ion (M+• for [M − H]+, [M − C2H2]+•, and [M − C4H2]+•; [M − H]+ for [M − 2H]+•), an appropriate mode was deleted to account for the reaction coordinate. An example of this would be for the case of H loss from the precursor ion, where the vibrational frequency removed corresponded to that of a C−H bond stretch; this stretching motion, if taken to the extreme, would theoretically be the breaking of a C−H bond−resulting in hydrogen atom loss. Of the remaining 3N − 7 modes for the transition state, the five modes with the lowest frequencies were scaled to adjust the entropy of activation. It has been shown for large species that transitional frequency scaling alone does not suffice to bring RRKM rates into agreement with experimental

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

b = (σ 2 − σp2 − μ(t1 + t 2) + t1t 2)/((t3 − t 2)(t3 − t1)) (3)

c = (σ 2 − σp2 − μ(t1 + t3) + t1t3)/((t 2 − t1)(t 2 − t3)) (4)

with μ = ∫ tTOF(t) dt/∫ tTOF(t) dt, σ2 = ∫ (t − t0)2TOF(t) dt/∫ TOF(t) dt, and t0 being the moment center and σp2 referring to the precursor ion. Once all the signals have been deconvoluted, their 13C contributions can be accounted for and the breakdown diagrams can be constructed. For the current molecule, the photon energy range used was 15−22 eV. The experimental data points were 0.1 eV apart in the regions where there is a high degree of change in ion abundance, the region of interest, whereas the step size was 0.2 eV outside of this range. 2.3. APCI-CID Mass Spectrometry. Atmospheric pressure chemical ionization (APCI) experiments were performed on a Micromass Quattro LC (Waters Micromass, Manchester, U.K.) triple quadrupole mass spectrometer equipped with a Z-spray source. Thirty microliters of sample solution was injected into the APCI source via a mobile phase of pure chloroform at a flow rate of 0.3 mL/min. The source block temperature was set at 150 °C while the probe temperature was kept at 400 °C. The corona and cone voltages were held steady at 3.4 kV and 30 V, respectively. The first quadrupole resolution was held constant 9872



Article

Figure 1. Reaction scheme for the unimolecular dissociation of the anthracene radical cation. Reactions R1−R3 were observed via MIKES and iPEPICO experiments whereas reactions R1, R2, and R4 were observed only in APCI experiments.

ones.46,47 Consequently, foregoing transition state calculations and thereby simplifying the model considerably appears to be a reasonable approximation in RRKM models of molecules of comparable sizes. 3.3. Fitting of the Statistical Model. The minimalPEPICO42 program was used to fit the iPEPICO experimental breakdown diagrams and TOF distributions. In short, the program combines the physical parameters of the iPEPICO experimental set up at the SLS with temperature (413 K in the current experiment, for the initial neutral molecule internal energy distribution) and the RRKM k(E) values for each channel to calculate experimental branching ratios for the ion dissociation as a function of photon energy, which are then compared to the experimental breakdown curves (13−22 eV) and asymmetric fragment ion TOF peaks, where applicable (in this case the peaks for C12H8+· were used up to hν = 18 eV). Photon energies were converted to ion internal energies using the ionization energy of anthracene (7.415 eV) reported by Mayer et al.48 The barriers to dissociation and the activation entropies are then optimized to obtain the best fit to experiment. Full details for the APCI-CID model have been reported previously.9 The three main experimental parameters for this model are the time-scale of the dissociation, the internal energy distribution of the collision complex (dependent on Ecom and gas density), and k(E). The time scale is known on the basis of the length of the collision cell (10 cm) and the translational energy, T (in this case T = Elab). The energy range was converted to center of mass energy (Ecom) using the following equation, ⎛ ⎞ MAr Ecom = E lab⎜ ⎟ ⎝ MAr + M ion ⎠

P(E ,Ecom) =

ρ(E)e−E / RTeff Q (Ecom)

(7)

where ρ(E) is the vibrational density of states (same as what was used for RRKM) and Q(Ecom) is the vibrational partition function at an effective temperature, Teff, described above. Theoretical branching ratios are then derived and fit to the experimental CID breakdown diagrams. In fitting the APCI-CID experiments, the iPEPICO results were used as an initial guess for E0 and Δ‡S for [M − H]+ and [M − C2H2]+•. Once an acceptable α value was obtained, these values were then adjusted to get the best fit with the experimental data.

4. RESULTS AND DISCUSSION 4.1. MIKES Mass Spectrometry. MIKES experiments revealed three reactions (R1−R3) for the unimolecular dissociation of the anthracene radical cation and its H-loss product ion: C14 H10+• → C14 H 9+ + H

(R1)

C14 H 9+ → C14 H8+• + H

(R2)

C14 H10+• → C12H8+• + C2H 2

(R3)

These results were in keeping with what was found previously for the naphthalene radical cation2 as well as those found by Ling and Lifshitz.4 The peaks for both C14H9+ and C14H8+• were Gaussian in shape, whereas our observed peak for C12H8+• was broader with a dished top. This peak shape is consistent with the results of Ling and Lifshitz who observed a similar metastable peak shape in their experiments.4 Both results are indicative of a reverse energy barrier for this channel and that the transition state for reaction R3 would likely correspond to a rearrangement reaction, and thus an entropically unfavorable “tight” transition state. Assigning product ion structures for reactions R1 and R2 was quite straightforward. The fragmentation map, with all proposed structures, is shown in Figure 1. In the previous study of the naphthalene radical cation, it was shown that due to hydrogen scrambling, all hydrogen sites are equally likely to be dehydrogenated.2 For this reason, structures 2 and 3 were assigned to C14H9+ and C14H8+•. The structure assignment for C12H8+• has undergone more debate. As stated, on the basis of the shape of the metastable peak, and the

(6)

where MAr and Mion are the respective masses of argon and the anthracene radical cation. The microcanonical rate constant is given by RRKM theory as outlined earlier, having two adjustable parameters, E0 and Δ‡S. As mentioned in the Introduction, the postcollision internal energy distribution of the dissociating ions has been approximated by the relationship in eq 1, with α and the initial temperature (Ti) also being adjustable parameters that can be varied to obtain the best fit between the theoretical and experimental dissociation diagrams.10 The resulting internal energy distribution of the anthracene ions is given by 9873



Article

Scheme 1. Structures of C12H8 Isomeric Ions

Figure 2. iPEPICO breakdown diagram and fitting illustrating the effect of using structure 4 (black broken line) versus structure 5 (solid gray line) for the calculation of C12H8+•.

indication of a reverse activation energy barrier,4 it is likely that the structure will involve some degree of rearrangement. However, Ling and Lifshitz derived a Δ‡S of over 50 J K−1 mol−1 from their kinetic modeling, indicative of a loose transition state. Ling, Martin, and Lifshitz calculated six potential product ions, covering both open and closed ring structures, at two different levels of theory (B3LYP/cc-PVDZ and B3PW91/cc-pVDZ).5 They compared the calculated dissociation energies with experimental values to determine which structure was the most probable. The results determined that, although acenaphthylene (6) was the lowest energy structure, biphenylene (4) was closest to the experimental dissociation energies. Cyclobuta[b]naphthalene (5), an alternate product ion to that for C2H2 loss from ionized naphthalene, was excluded because its calculated energy was too high. These results inspired another study, conducted by Johansson et al., which explored the potential energy surface for acetylene loss from the anthracene radical cation.6 The product structures explored were 2-ethynylnaphthalene (7), biphenylene (4), and acenaphthylene (6), with cyclobuta[b]naphthalene (5) not being considered due to the conclusions drawn by Ling et al. (Scheme 1).5 All calculations were conducted at the B3-LYP/6-311++G(2d,p) level of theory. The results for 2-ethynylnaphthalene (7), which had already been ruled out due to the dished peak shape from Ling’s MIKES study,4 indicate that the highest lying transition state toward its formation is at 4.99 eV. For the biphenylene ion structure (4), where anthracene first isomerizes to phenanthrene, the highest point on the potential energy surface was calculated to be 5.64 eV, whereas the energy requirement for isomerization was 4.89 eV. For the formation of acenaphthylene (6), which was only

calculated from the phenanthrene precursor ion, the maximum energy was 4.95 eV. On the basis of these results, there are a few problems with choosing a final structure. The formation of both acenaphthylene (6) and biphenylene (4) requires the isomerization of the anthracene radical cation to the phenanthrene radical cation. Because the formation of acenaphthylene was calculated to be more than 0.5 eV lower in energy than required for biphenylene, the likelihood of forming the latter should be quite low. Another problem, which will become apparent in the next section, is that all three energies calculated by Johansson et al. (4.99, 5.64, and 4.95 eV)6 are significantly higher than what was extracted from the RRKM modeling of the iPEPICO results (∼4.2 eV). In the end, two structures are presented here as possible structures for C12H8+•, biphenylene (4) and cyclobuta[b]naphthalene (5). These structures were chosen on the basis of the results of the naphthalene radical cation study where it was determined that acetylene loss in this system resulted in the formation of benzocyclobutene, and both these proposed structures are analogous.2 4.2. iPEPICO. iPEPICO experiments were conducted over a photon energy range of 15−22 eV. The fragments observed correspond to reactions R1−R3 from the MIKES results, the consecutive H loss (R2) appears at around 17 eV, as seen in Figure 2. Similarly to what was observed for the naphthalene radical cation, there was a rise in intensity observed for reaction R2, which could be attributed to acetylene loss from C14H9+.2 However, the change is quite subtle and, therefore, it was not possible to fit this channel reliably and reaction R2 is thus not included in this discussion. 9874



Article

Table 1. 0 K Activation Energies (E0/eV) Determined by iPEPICO and APCI-CID Experiments as Well as Comparison to Literature Values C14H10 → C14H9+ + H C14H9+ → C14H8+• + H C14H10+· → C12H8+• + C2H2 C14H10+· → C10H8+• + C4H2 +•

a

(R1) (R2) (R3) (R4)

iPEPICOa

iPEPICOb

APCI-CIDa

APCI-CIDb

literaturec

4.28 ± 0.29 2.71 ± 0.19 4.21 ± 0.30

4.28 ± 0.30 2.71 ± 0.20 4.19 ± 0.30

4.55 ± 0.10

4.55 ± 0.10

4.18 ± 0.10 4.87 ± 0.30

4.17 ± 0.06

4.38 2.85 4.50 4.46

± ± ± ±

0.1 0.2 0.1 0.1

Experiments assuming structure 5. bExperiments assuming structure 4. cLiterature values obtained by Ling and Lifshitz.4

Table 2. Entropy of Activation (Δ‡S/J K−1 mol−1) Determined by iPEPICO and APCI-CID Experiments as Well as Comparison to Literature Values C14H10 → C14H9+ + H C14H9+ → C14H8+• + H C14H10+· → C12H8+• + C2H2 C14H10+· → C10H8+• + C4H2 +•

a

(R1) (R2) (R3) (R4)

iPEPICOa

iPEPICOb

APCI-CIDa

APCI-CIDb

12 ± 15 0 ± 15 7 ± 10

13 ± 15 1 ± 15 22 ± 10

13 ± 2

13 ± 2

−1 ± 6 23 ± 8

12 ± 3 23 ± 8

literaturec 24.7 −8.4 53.6 8.4

± ± ± ±

4 8 4 4

Experiments assuming structure 5. bExperiments assuming structure 4. cLiterature values obtained by Ling and Lifshitz.4

the two models are essentially identical (4.19 ± 0.3 eV to form structure 4, compared to 4.21 ± 0.3 eV to form 5). There is a minor difference suggested when the entropies of activation are compared, 7 ± 10 J K−1 mol−1 for structure 5 and 22 ± 10 J K−1 mol−1 for structure 4. Ling et al. derived a Δ‡S value of 53.6 ± 4 J K−1 mol−1, which is significantly higher than our results.4 The discrepancy continues when the activation energies are compared as the value obtained by Ling (4.50 ± 0.1 eV) is at the upper error limit for both values obtained here. There is a well-known interplay between E 0 and Δ ‡ S in fitting experimental data, and the added constraint of fitting the asymmetric TOF peaks in our data further constrains the possible values for these two parameters, adding confidence to our results. C2H2 loss from ionized naphthalene to form ionized benzocyclobutadiene was found to have E0 and Δ‡S values of 4.12 ± 0.05 eV and 0 ± 2 J K−1 mol−1,2 respectively, which agree better with the present results forming ion 5 (especially the entropy). This comparison, despite the size difference, is being made due to the results of the study conducted by West et al., which compares 1,2-dihydronaphthalene and 9,10dihydrophenanthrene.3 It was concluded in this study that size has very little effect on the threshold energetic of similar PAH-like cations. A tighter transition state is also more consistent with a large reverse energy barrier as suggested by the shape of the metastable fragment ion peak in the MIKES experiment. All of this suggests that the final product from the loss of C2H2 from ionized anthracene is actually the cyclobuta[b]naphthalene ion (5). We investigated the loss of C2H2 from both ionized anthracene and phenanthrene at the CBS-QB3 level of theory. The key transition states, shown in Figure S2 of Supporting Information, are practically isoenergetic relative to ionized anthracene (+5.3 eV), and so the branching ratio for the two possible products, ionized cyclobuta[b]naphthalene and biphenylene, will largely be determined on the entropic nature of the two transition states, and thus the ratio of the sum-of-states of the two structures. This ratio approaches a value of 2 at the internal energies accessed in these experiments (i.e., the transition state leading from ionized anthracene to products is slightly looser), suggesting, again, that the principle product from the loss of C2H2 from ionized anthracene is ionized cyclobuta[b]-

A technique similar to that used in the naphthalene radical cation fitting was employed here: the primary channels were fit first, using both the experimental breakdown curves and the asymmetric TOF peaks for C12H8+•, followed by the fitting for the sequential dehydrogenation channel, reaction R2.2 The results of the fitting can be seen in Figure 2, with the calculated curves superimposed on the experimental data. The asymmetric TOF fits for C12H8+• can be seen in Figure S1 of the Supporting Information to demonstrate the quality of the fitting. The kinetic data, E0 and Δ‡S, extracted from these fits are summarized in Tables 1 and 2. For reaction R1, the values obtained for E0 and Δ‡S, 4.28 ± 0.29 eV and 12 ± 15 J K−1 mol−1, respectively, are in quite good agreement with the results for the equivalent channel from the naphthalene radical cation (4.20 ± 0.04 eV and 2 ± 2 J K−1 mol−1).2 This agrees with previous observations that all PAHs of similar design (in this case cata-condensed) will have similar chemical properties.49 The positive value for Δ‡S is also in keeping with the nature of the reaction being a simple bond cleavage which should require minimal molecular rearrangement. These values are also found to be in good agreement with those in the literature, 4.38 ± 0.1 eV and 24.7 ± 4 J K−1 mol−1 from Ling and Liftshitz, respectively, with good overlap between the two sets when uncertainties are taken into account.4 The values obtained here for reaction R2, E0 of 2.71 ± 0.19 eV and Δ‡S of 0 ± 15 J K−1 mol−1, are again in keeping with the results found by Ling (2.85 ± 0.2 eV and −8.4 ± 8 J K−1 mol−1). Unlike reaction R1, these values are not quite as close to those determined for the naphthalene radical cation (3.20 ± 0.13 eV and −19 ± 11 J K−1 mol−1 for E0 and Δ‡S, respectively).2 Our previous work, which looked at 1,2dihydronaphthalene and 9,10-dihydrophenanthrene radical cations, showed that when deviations from the unadulterated PAHs bring about structural changes that affect the C−H bond dissociation energies.3 The highest energy dissociation reaction observed using iPEPICO was reaction R3, the acetylene loss channel. The derived fit was independent of the reaction symmetry number (two are shown in Figure 2 corresponding to different proposed product ion structures, 4 or 5) The curves are indistinguishable below16.5 eV, as are their asymmetric TOF fits for the C12H8+• peak. The derived activation energies for 9875



Article

Figure 3. APCI/CID breakdown diagram and fitting illustrating the effect of using structure 4 (black broken line) versus structure 5 (solid gray line) for the calculation of C12H8+•. Pressure and the corresponding α value are also given.

naphthalene radical cation (8). The anthracene dissociative photoionization activation energies can be compared to those for ionized naphthalene. In C10H8+, C4H2 loss required only 0.15 eV more energy than C2H2 loss, whereas in this case it requires an extra 0.69 eV. The difference between these values, 0.54 eV, is likely due to a larger barrier for H-transfer due to the more delocalized charge in anthracene. The last reaction to consider is reaction R3. Again, the results are essentially independent of symmetry number with E0 ∼ 4.17 ± 0.06 eV and Δ‡S ∼ 12 ± 3 J K−1 mol−1, both in agreement with the iPEPICO results. Admittedly, our feeling at the start of the study was that the assignment of an effective temperature to the postcollision ions would not result in adequate fits to the experimental data. However, surprisingly, this turned out not to be the case. Our previous work has been focused on the relative energetics and entropics for dissociating systems involving noncovalent interactions, and this is the first time the simple model described above has been tested against reliable iPEPICO data. It appears that whatever the postcollision internal energy distribution is in these experiments, the shape of the resulting distribution function at least appears similar in effect to a Boltzmann distribution.

naphthalene, regardless of how easy ionized anthracene and phenanthrene interconvert prior to dissociation. 4.3. APCI-CID Mass Spectrometry. Collision-induced dissociation (CID) experiments were conducted over a center-of-mass collision energy range of 2.5−9.0 eV. In addition to reactions R1 and R3, loss of the larger C4H2 moiety was also observed: C14 H10+· → C10H8+· + C4 H 2

(R4)

For this study we limited the analysis to the primary reactions from the molecular ion and did not explore the sequential reaction, R2. The intensity of the peak due to [M − 2H]+• was simply added into the [M − H]+ intensity. The breakdown curves generated for reactions R1, R3, and R4 are shown in Figure 3 with the extracted E0 and Δ‡S summarized in Tables 1 and 2, respectively. Looking at Figure 3, it can be seen that the CID experiments produced well-defined breakdown curves, which made fitting the data possible with a fair degree of confidence. The α value was determined to be 400 K eV−1, which allows for the determination of the effective temperature of the postcollision ions at each center-of-mass collision energy. Reaction R1 gave E0 and Δ‡S values of 4.55 ± 0.10 eV and 13 ± 4 J K−1 mol−1, respectively. These values are within the uncertainty of the iPEPICO data reported herein. When the results are compared to literature values, it can be seen that the activation energy determined by Ling is equally close to the CID results as to the iPEPICO ones. The Δ‡S values deviate more when compared to literature values, but they are still in keeping with iPEPICO results. Reaction R4 can only be compared to literature results as there was insufficient ion intensity for this channel to produce a fit in the iPEPICO experiment. The activation energy was determined to be 4.87 ± 0.30 eV, which is in good agreement with the value determined by Ling of 4.46 ± 0.1 eV.4 There is, again, a larger difference when Δ‡S values are compared: CID results yielded a value of 23 ± 10 J K−1 mol−1 compared to 8.4 ± 4 J K−1 mol−1.4 Nevertheless, both results indicate that the transition state is loose, which helped us assign the product structure as the

5. CONCLUSION The dissociation of the anthracene radical cation has been investigated using two different techniques, imaging photoelectron photoion coincidence spectrometry and atmospheric pressure chemical ionization−collision induced dissociation mass spectrometry. The combination of these methods has allowed for a detailed picture of the fragmentation channels. The results from the iPEPICO fitting for reactions R1−R3 are E0 (in eV) values of 4.28 ± 0.30 (R1), 2.71 ± 0.20 (R2), and 4.20 ± 0.30 (average of R3) whereas the Δ‡S values (in J K−1 mol−1) of 12 ± 15 (R1), 0 ± 15 (R2), and either 7 ± 10 (using structure 5 for reaction R3) or 22 ± 10 (using structure 4 for reaction R3). The results for the APCI-CID results are similar; the E0 (in eV) values are 4.55 ± 0.10 (R1), 4.87 ± 0.30 (R4), and 4.18 ± 0.30 (average of reaction R3) whereas the Δ‡S 9876



Article

values (in J K−1 mol−1) are 13 ± 2 (R1), 23 ± 8 (R4), and either −1 ± 6 (using structure 5 for reaction R3) or 12 ± 3 (using structure 4 for reaction R3). In both cases, using either 4 or 5 gave excellent breakdown diagrams fits but the Δ‡S values for forming 5 are more consistent with a large reverse energy barrier, as suggested by the shape of the metastable fragment ion peak in the MIKES experiment. In addition, computational results on the potential energy surface for the production of 4 and 5 also indicate that 5 is the most likely product from ionized anthracene. One of the goals of this study was to determine if the APCICID method would be reliable to continue the dissociation study of larger PAHs. In the two reactions R1 and R3, observed by both techniques, the results were internally consistent between the two methods. Because the APCI-CID method has fewer size restrictions for the molecules it can study, it may be a viable alternative for larger and less volatile systems. Further experiments are underway in which the CID-based model will be tested against iPEPICO data to explore the postcollision internal energy distribution more fully.

■

Spectrometry and Imaging-Pepico Study of Dihydronaphthalene and Dihydrophenanthrene. J. Phys. Chem. A 2014, 118, 1807−1816. (4) Ling, Y.; Lifshitz, C. Time-Dependent Mass Spectra and Breakdown Graphs. 21. C14H10 Isomers. J. Phys. Chem. A 1998, 102, 708−716. (5) Ling, Y.; Martin, J. M. L.; Lifshitz, C. Energetics of Acetylene Loss from C14H10+ Cations: A Density Functional Calculation. J. Phys. Chem. A 1997, 101, 219−226. (6) Johansson, H. A. B.; Zettergren, H.; Holm, A. I. S.; Haag, N.; Nielsen, S. B.; Wyer, J. A.; Kirketerp, M.-B. S.; Støchkel, K.; Hvelplund, P.; Schmidt, H. T.; et al. Unimolecular Dissociation of Anthracene and Acridine Cations: The Importance of Isomerization Barriers for the C2H2 Loss and HCN Loss Channels. J. Chem. Phys. 2011, 135, 084304. (7) Zhen, J.; Paardekooper, D. M.; Candian, A.; Linnartz, H.; Tielens, A. G. G. M. Quadrupole Ion Trap/Time-of-Flight Photo-Fragmentation Spectrometry of the Hexa-Peri-Hexabenzocoronene (Hbc) Cation. Chem. Phys. Lett. 2014, 592, 211−216. (8) Comeau, A. N.; Renaud, J. B.; Mironov, G. G.; Berezovski, M. V.; Mayer, P. M. Investigating the Relationship between the Gas-Phase Conformations and Dissociation Energetics of Peptide−Saccharide Complexes. Int. J. Mass Spectrom. 2012, 316-318, 31−39. (9) Mayer, P. M.; Martineau, E. Gas-Phase Binding Energies for NonCovalent aβ-40 Peptide/Small Molecule Complexes from CID Mass Spectrometry and Rrkm Theory. Phys. Chem. Chem. Phys. 2011, 13, 5178−5186. (10) Renaud, J. B.; Martineau, E.; Mironov, G. G.; Berezovski, M. V.; Mayer, P. M. The Collaborative Role of Molecular Conformation and Energetics in the Binding of Gas-Phase Non-Covalent Polymer/Amine Complexes. Phys. Chem. Chem. Phys. 2012, 14, 165−172. (11) Renaud, J. B.; Overton, S.; Mayer, P. M. Energy and Entropy at Play in Competitive Dissociations: The Case of Uneven Positional Dissociation of Ionized Triacylglycerides. Int. J. Mass Spectrom. 2013, 352, 77−86. (12) Mayer, P. M.; Poon, C. The Mechanisms of Collisional Activation of Ions in Mass Spectrometry. Mass Spectrom. Rev. 2009, 28, 608−639. (13) Asano, K. G.; Goeringer, D. E.; Butcher, D. J.; McLuckey, S. A. Bath Gas Temperature and the Appearance of Ion Trap Tandem Mass Spectra of High-Mass Ions. Int. J. Mass Spectrom. 1999, 190/191, 281− 293. (14) Asano, K.; Goeringer, D.; McLuckey, S. Dissociation Kinetics in the Quadrupole Ion Trap. Proceeding of the 46th Conference for the American Society for Mass Spectrometry; ASMS: Santa Fe, NM, 1998. (15) Goeringer, D. E.; McLuckey, S. A. Evolution of Ion Internal Energy During Collisional Excitation in the Paul Ion Trap: A Stochastic Approach. J. Chem. Phys. 1996, 104, 2214−2221. (16) Hart, K. J.; McLuckey, S. A. Relative Dissociation Energy Measurements Using Ion Trap Collisional Activation. J. Am. Soc. Mass Spectrom. 1994, 5, 250−259. (17) Asano, K. G.; Butcher, D. J.; Goeringer, D. E.; McLuckey, S. A. Effective Ion Internal Temperatures Achieved Via Boundary Activation in the Quadrupole Ion Trap: Protonated Leucine Enkephalin. J. Mass Spectrom. 1999, 34, 691−698. (18) Goeringer, D. E.; Asano, K. G.; McLuckey, S. A. Ion Internal Temperature and Ion Trap Collisional Activation: Protonated Leucine Enkephalin. Int. J. Mass Spectrom. 1999, 182/183, 275−288. (19) Goeringer, D. E.; Duckworth, D. C.; McLuckey, S. A. CollisionInduced Dissociation in Quadrupole Ion Traps: Application of a Thermal Model to Diatomic Ions. J. Phys. Chem. A 2001, 105, 1882− 1889. (20) Goeringer, D. E.; McLuckey, S. A. Kinetics of Collisio-Induced Dissociation in the Paul Trap: A First-Order Model. Rapid Commun. Mass Spectrom. 1996, 10, 328−334. (21) Laskin, J.; Futrell, J. H. Collisional Activation of Peptide Ions in Ft-Icr Mass Spectrometry. Mass Spectrom. Rev. 2003, 22, 158−181. (22) Laskin, J.; Futrell, J. H. Activation of Large Ions in Ft-Icr Mass Spectrometry. Mass Spectrom. Rev. 2005, 24, 135−167.

ASSOCIATED CONTENT

S Supporting Information *

Full authorship for ref 43. Figure S1 shows the RRKM fitted TOF distributions for the dissociation channels of ionized anthracene. Figure S2 illustrates the CBS-QB3 calculated C2H2loss path from the anthracene and phenanthrene cations. Table S1 contains the vibrational frequencies for the processes modeled with eq 1 for both ions. This material is available free of charge via the Internet at http://pubs.acs.org

■

AUTHOR INFORMATION

Corresponding Author

*Paul M. Mayer. E-mail: [email protected]. Phone: 1-613562-5800, 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.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) 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. (3) 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 9877



Article

(43) 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. (44) Baer, T.; Hase, W. L. Unimolecular Reaction Dynamics, Theory and Experiments; Oxford University Press: New York, 1996. (45) Beyer, T.; Swinehart, D. R. Number of Multiply-Restricted Partitions [A1] (Algorithm 448). ACM Commun. 1973, 16, 379. (46) Fischer, K. H.; Schneider, M.; Fischer, I.; Pfaffinger, B.; Braunschweig, H.; Sztáray, B.; Bodi, A. Bonding in a Borylene Complex Investigated by Photoionization and Dissociative Photoionization. Chem. − Eur. J. 2012, 18, 4533−4540. (47) Hemberger, P.; Bodi, A.; Gerber, T.; Würtemberger, M.; Radius, U. Unimolecular Reaction Mechanism of an Imidazolin-2-Ylidene: An Ipepico Study on the Complex Dissociation of an Arduengo-Type Carbene. Chem. − Eur. J. 2013, 19, 7090−7099. (48) 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. (49) 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.

(23) Laskin, J.; Byrd, M.; Futrell, J. Internal Energy Distributions Resulting from Sustained Off-Resonance Excitation in Ftms. I. Fragmentation of the Bromobenzene Radical Cation. Int. J. Mass Spectrom. 2000, 195/196, 285−302. (24) Laskin, J.; Denisov, E.; Futrell, J. A Comparative Study of Collision-Induced and Surface-Induced Dissociation. 1. Fragmentation of Protonated Dialanine. J. Am. Chem. Soc. 2000, 122, 9703−9714. (25) Laskin, J.; Denisov, E.; Futrell, J. Fragmentation Energetics of Small Peptides from Multiple-Collision Activation and SurfaceInduced Dissociation in Ft-Icr Ms. Int. J. Mass Spectrom. 2002, 219, 189−201. (26) Nesatyy, V. J.; Laskin, J. Dissociation of Noncovalent Protein Complexes by Triple Quadrupole Tandem Mass Spectrometry: Comparison of Monte Carlo Simulation and Experiment. Int. J. Mass Spectrom. 2003, 221, 245−262. (27) 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. (28) Bastian, M. J.; Dressler, R. A.; Levandier, D. J.; Murad, E.; Muntean, F.; Armentrout, P. B. Low Energy Cid and Pd Studies of the (N2O,H2O)+ Cluster Ion. J. Chem. Phys. 1997, 106, 9570−9579. (29) Muntean, F.; Armentrout, P. B. Guided Ion Beam Study of Collision-Induced Dissociation Dynamics: Integral and Differential Cross Sections. J. Chem. Phys. 2001, 115, 1213−1228. (30) Ervin, K. M.; Armentrout, P. B. Translational Energy Dependence of Ar+ + XY → ArX+ + Y (XY=H2,D2,HD) from Thermal to 30 eV C.M. J. Chem. Phys. 1985, 83, 166−189. (31) Kenttamaa, H. I.; Cooks, R. G. Internal Energy Distributions Acquired through Collisional Activation at Low and High Energies. Int. J. Mass Spectrom. Ion Processes 1985, 64, 79−83. (32) Martínez-Núñez, E.; Fernández-Ramos, A.; Vázqueza, S. A.; Marques, J. M. C.; Xue, M.; Hase, W. L. Quasiclassical Dynamics Simulation of the Collision-Induced Dissociation of Cr(Co)6+ with Xe. J. Chem. Phys. 2005, 123, 154311. (33) Douglas, D. J. Applications of Collision Dynamics in Quadrupole Mass Spectrometry. J. Am. Soc. Mass Spectrom. 1998, 9, 101−113. (34) Wysocki, V. H.; Kenttamaa, H. I.; Cooks, R. G. Internal Energy Distributions of Isolated Ions after Activation by Various Methods. Int. J. Mass Spectrom. Ion Processes 1987, 75, 181−208. (35) Knyazev, V. D.; Stein, S. E. Monte Carlo/Rrkm/Classical Trajectories Modeling of Collisional Excitation and Dissociation of NButylbenzene Ion in Multipole Collision Cells of Tandem Mass Spectrometers. J. Phys. Chem. A 2010, 114, 6384−6393. (36) Knyazev, V. D.; Stein, S. E. Classical Trajectories and Rrkm Modeling of Collisional Excitation and Dissociation of Benzylammonium and Tert-Butyl Benzylammonium Ions in a QuadrupoleHexapole-Quadrupole Tandem Mass Spectrometer. J. Am. Soc. Mass. Spectrom. 2010, 21, 425−439. (37) 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. (38) Bodi, A.; Johnson, M.; Gerber, T.; Gengeliczki, Z.; Sztaray, B.; Baer, T. Imaging Photoelectron Photoion Coincidence Spectroscopy with Velocity Focusing Electron Optics. Rev. Sci. Instrum. 2009, 80, 7. (39) Johnson, M.; Bodi, A.; Schulz, L.; Gerber, T. Vacuum Ultraviolet Beamline at the Swiss Light Source for Chemical Dynamics Studies. Nucl. Instrum. Methods A 2009, 610, 597−603. (40) Bodi, A.; Hemberger, P. Imaging Breakdown Diagrams for Bromobutyne Isomers with Photoelectron-Photoion Coincidence. Phys. Chem. Chem. Phys. 2014, 16, 505−515. (41) Sztaray, B.; Baer, T. Suppression of Hot Electrons in Threshold Photoelectron Photoion Coincidence Spectroscopy Using Velocity Focusing Optics. Rev. Sci. Instrum. 2003, 74, 3763−3768. (42) Sztáray, B.; Bodi, A.; Baer, T. Modeling Unimolecular Reactions in Photoelectron Photoion Coincidence Experiments. J. Mass Spectrom. 2010, 45, 1233−1245. 9878


Dissociation of the Anthracene Radical Cation: A Comparative Look at

Recommend Documents