Article pubs.acs.org/JPCA
Formation of Complex Organics in the Gas Phase by Sequential Reactions of Acetylene with the Phenylium Ion Abdel-Rahman Soliman, Ahmed M. Hamid, Paul O. Momoh,† and M. Samy El-Shall* Department of Chemistry, Virginia Commonwealth University, Richmond, Virginia 23284-2006, United States
Danielle Taylor, Lauren Gallagher, and Samuel A. Abrash Department of Chemistry, University of Richmond, Richmond, Virginia 23173, United States S Supporting Information *
ABSTRACT: In this paper, we report a study on the reactivity of the phenylium ion with acetylene, by measuring product yield as a function of pressure and temperature using mass-selected ion mobility mass spectrometry. The reactivity is dominated by a rapid sequential addition of acetylene to form covalently bonded C8H7+ and C10H9+ ions with an overall rate coefficient of 7−5 × 10−10 cm3 s−1, indicating a reaction efficiency of nearly 50% at room temperature. The covalent bonding nature of the product ions is confirmed by high temperature studies where enhanced production of these ions is observed at temperatures as high as 660 K. DFT calculations at the UPBEPBE/6-31+ +G** level identify the C8H7+ adduct as 2-phenyl-ethenylium ion, the most stable C8H7+ isomer that maintains the phenylium ion structure. A small barrier of 1.6 kcal/mol is measured and attributed to the formation of the second adduct C10H9+ containing a four-membered ring connected to the phenylium ion. Evidence for rearrangement of the C10H9+ adduct to the protonated naphthalene structure at temperatures higher than 600 K is provided and suggests further reactions with acetylene with the elimination of an H atom and an H2 molecule to generate 1-naphthylacetylene or acenaphthylene cations. The high reactivity of the phenylium ion toward acetylene is in sharp contrast to the low reactivity of the benzene radical cation with a reaction efficiency of 10−4−10−5, confirming that the first step in the cation ring growth mechanism is the loss of an aromatic H atom. The observed reactions can explain the formation of complex organics by gas phase ion−molecule reactions involving the phenylium ion and acetylene under a wide range of temperatures and pressures in astrochemical environments.
■
INTRODUCTION Many complex organics including polycyclic aromatic hydrocarbons (PAHs) and species with 30-or-more carbon atoms are present in flames and combustion processes as well as in outer space.1−9 The formation and growth mechanisms of these molecules involve organic chemistry dealing with radical and ion−molecule reactions as well as surface catalysis on dust particles.3,5,6,8 Gas phase organics usually participate in the process of gas-to-particle conversion, which is often quite complex and can include photochemical gas phase reactions, heterogeneous surface chemistry, nucleation, and eventually, liquid-phase reactions.10−12 The resulting particles play important roles in urban smog formation, soot formation, global climate, organic aerosols, and human health.10−12 Because molecules in outer space are subject to ionizing radiation and reaction rates of ion−molecule reactions may exceed by orders of magnitude those of gas phase neutral reactions at the low interstellar temperature near 10 K (with the exception of fast radical-molecule reactions),13 ion chemistry becomes increasingly competitive to gas phase neutral chemistry in cold ionizing environments.14,15 In ion−molecule reactions, the processes of particular interest in the field of astrochemistry are those that lead to larger molecules, which may lead to the formation of PAHs found in soot, meteorites and interstellar clouds.9,16−20 © 2012 American Chemical Society
Identifying the chains of ion−molecule reactions that can lead to the formation of complex organics and understanding the formation and growth mechanisms of key molecular ions and oligomers require detailed information on their structures, bonding, reactivity, isomerization, and dissociation kinetics. The discovery of benzene in the space surrounding carbonrich stars in the direction of the proto-planetary nebula CRL 61821 and on Titan4,22 has raised the possibility that benzene could contribute to the formation of PAH cations and other complex organics made of hundreds of carbon atoms that could be responsible for the unidentified infrared bands (UIBs)9,19,23 and the diffuse interstellar bands.24 In addition to benzene, large mass positive and negative ions were discovered on Titan,4,25−29 which were attributed to fused-ring polycyclic aromatic hydrocarbon compounds such as naphthalene and anthracene and suggested that these are the precursors to the haze particles that form the optically thick haze layer lower in Titan’s atmosphere.4,28,29 The wide range of conditions in solar nebulae, from cold dilute gases (10−50 K) to >1000 K with a variable pressure range, allows diverse reactions of these ions with interstellar/ Received: June 19, 2012 Revised: August 10, 2012 Published: August 13, 2012 8925
dx.doi.org/10.1021/jp306046w | J. Phys. Chem. A 2012, 116, 8925−8933
The Journal of Physical Chemistry A
Article
nebula molecules.2,30−32 Energy barriers may prevent reactions at low temperatures in interstellar medium, but these reactions may be permitted at high temperatures in the solar nebulae.30−32 For example, we recently discovered new sequential reactions of acetylene with the benzene cation at higher temperatures, leading to the formation of naphthalenetype ions.33 Interestingly, no reaction leading to covalent addition is observed at room temperature, and even the associated C6D6•+(C2H2)n products are not observed due to weak binding. However, at high temperatures (600 K), sequential covalent addition of two acetylene molecules onto the benzene cation leading to the formation of naphthalenetype ions has been observed.33 These reactions may explain the formation of complex organics by the sequential reactions of acetylene with the benzene cation at high temperatures.19,20 The barrier to the covalent addition of acetylene onto the benzene radical cation (C6H6•+) originates from the presence of six C−H bonds in the benzene cation and the absence of an available addition site. Theoretical calculations indicate that the first step in the cation ring growth reaction is the loss of an aromatic H atom.34−36 This could occur in interstellar and solar nebulae environments by ionizing radiation, and in combustion processes this would take place by the abstraction of H by a reactive radical species such as H and C2H in nebulae environments or O and OH radicals in combustion processes. Therefore, one may expect the reaction of acetylene with the phenylium ion (C6H5+) to be much faster than the reaction with the benzene radical cation. In fact, theoretical calculations have predicted that acetylene adds covalently to the phenylium ion in a barrierless exothermic reaction to form the C8H7+ adduct ion.36 The calculations also predict that addition of the second acetylene to form a four-membered ring has a very small barrier (1 kcal/mol) and is also an exothermic reaction.36 These theoretical predictions have not been confirmed by experimental observation since there is no report in the literature on the sequential reactions of acetylene with the phenylium ion at different temperatures in the high pressure regime. Surprisingly, very little experimental work has been reported on the sequential reactions of acetylene with ionized aromatics at different temperatures. Early work carried out using ICR and SIFT experiments reported that the phenylium ion is uncreative toward acetylene.37−39 This has been contrasted by other reports that documented the high reactivity of an electrophile such as the phenyl cation toward alkynes.40,41 In fact, the phenylium ions are known to be strong electrophiles and are readily attacked by species such as water, alcohols, and amines.42−45 Two stable isomeric structures are known to exist for the C6H5+ ions, the low energy cyclic (phenylium), and the higher energy acyclic isomers that exhibit distinctly different reactivities.40,46−49 The more reactive isomer of C6H5+ was originally attributed to the acyclic form,37−39 but later work has shown that the lower energy phenylium ion is more reactive than the acyclic isomer.40,46−49 The C6H5+ ions produced by sequential ion reactions of acetylene (C2H2•+ + C2H2 →) were found to contain about 50% of the reactive phenylium ion structure.48 The reaction rate coefficient of the cyclic C6H5+ ion with acetylene was determined by the SIFT technique48 to be 6.0 × 10−10 cm3 s−1, significantly larger than the rate measured by ICR (1.3 to 1.7 × 10−10 cm3 s−1).40,49 However, in all previous ICR and SIFT experiments, no temperature studies, product identifications, or proposed reaction mechanisms have been reported. On the other hand, under the high pressure conditions involved in the decay radiolytic experiments (100
Torr), the product analysis of the phenylium ion reaction with acetylene in the presence of methanol was explained based on the formation of the C8H7+ adduct proposed to have the structure of the α-phenylvinylium ion.50 This ion was found to react with methanol to give acetophenone and other products.50 The reactivity of the phenylium ion has also been demonstrated in the reaction with benzene to yield protonated biphenyl (C12H11+), protonated naphthalene (C10H9+), and C9H7+.51,52 As part of our program to study the reactions of acetylene with cyclic ions that could lead to the formation of PAH-type cations, we investigated the reactions of the phenylium ion (C6H5+) with acetylene over a wide temperature range using the mass-selected ion mobility technique.33,53 Here, we report experimental evidence for the efficient formation of cyclic organics in the gas phase by the sequential addition of acetylene on the phenylium ion. The observed reactions could contribute to the formation of complex organics such as PAHs found in a wide variety of locations ranging from flames and combustion processes to interstellar space.2,4,6,16,17,28,54,55
■
EXPERIMENTAL SECTION We studied the reactions of the phenylium ion with acetylene using the mass-selected ion mobility (drift cell) technique.33,53 Two different sets of experiments were performed involving (1) a mixture of the phenylium (C6H5+) and benzene cations (C6H6+•) generated by electron impact (EI) ionization (EI, 50 eV) of the supersonic beam expansion of a benzene (2%)/ helium mixture and (2) pure phenylium cations produced by EI dissociative ionization of bromobenzene (Fluke 99.5% GC grade). In the experiments, the mass-selected ions (C6H5+, m/z 77 and C6H6+•, m/z 78 or C6H5+, m/z 77 only) were injected into a drift cell containing a purified acetylene−helium gas mixture or pure purified acetylene gas at well-defined pressures and temperatures. In all experiments, no fragmentation was observed from the mass-selected ions as a result of the ion injection process into the drift cell. The injection energies used in the experiments (11−13 eV, laboratory frame) are slightly above the minimum energies required to introduce the ions into the cell against the C2H2/He outflow from the entrance orifice. Ion thermalization occurs outside the cell entrance by collisions with the C2H2/He gas escaping from the cell entrance orifice. At a cell pressure of 0.6 Torr, the number of collisions that the C6H5+ ion encounters with the neutral molecules within one millisecond residence time inside the cell is about 104 collisions, which is sufficient to ensure efficient thermalization of the C8H5+ ions. Residence times of the various ions are measured by monitoring the signals corresponding to each ion as a function of time after injection into the cell. Residence time can be varied by changing the voltage gradient in the cell. The reactant and product ions exiting the cell are analyzed and detected using a second quadrupole mass spectrometer. Timeresolved studies allow the identification of primary and secondary reaction products and the measurement of the rate coefficient.33
■
THEORETICAL SECTION Geometries and energies were calculated using the Gaussian 09 suite of programs.56 All calculations were completed using density functional theory using the density functionals of Perdew, Burke, and Ernzerhof and the Pople 6-31++G** basis set (UPBEPBE/6-31++G** level).56 Frequency calculations 8926
dx.doi.org/10.1021/jp306046w | J. Phys. Chem. A 2012, 116, 8925−8933
The Journal of Physical Chemistry A
Article
conditions of our experiments (0.2−1.0 Torr) ensure complete thermalization of the injected ions by collisions with the C2H2/ He outflow from the entrance orifice of the drift cell. This suggests that the phenylium structure is the most likely structure in our experiments in view of the high energy barrier (2.0 ± 0.3 eV)40 required for ring-opening. To confirm the structure of the C6H5 ions in our experiments, we measured the mobility of the mass-selected C6H5+ fragment produced from the ionization of bromobenzene. Mobility can provide structural information on ions on the basis of their collision integrals (Ω), which depend on the geometric shapes of the ions.33,53,58 Figure 1 displays the arrival time distributions (ATDs) for the mass selected C6H5 ions measured at various cell voltages. The inset in Figure 1 shows a plot of td (the drift time) versus P/V (V is the drift voltage and P is the He buffer gas pressure (Torr)), with the solid line representing the least-squares fitting to the data points. The resulting reduced mobility K0 of the C6H5+ fragment produced from the ionization of bromobenzene is 11.5 ± 0.3 cm2 V−1 s−1, in excellent agreement with the value measured for the benzene cation (11.4 ± 0.4 cm2 V−1 s−1), which has the same structure as the phenylium ion.58 The corresponding Ωs at 300 K for the C6H5+ fragment ion and the benzene ion are 47.2 ± 1.4 and 47.9 ± 1.4 Å2, respectively. To compare with the linear C6H5+ isomers, collision integrals for the five lowest energy structures of the C6H5+ ions were calculated using the trajectory method.59 The optimized geometries of the C6H5+ isomers were determined from the DFT calculations using the unrestricted PBEPBE functional with the 6-31++G** basis set.56 Table 1 shows the collision integrals (Ω) calculated for the C6H5+ isomers and their relative total energies (with the energy of the phenylium ion, the most stable C6H5+ isomer, taken as zero). It is clear that mobility nicely distinguishes the cyclic phenylium ion from the acyclic isomers. Therefore, among the C6H5+ lowest energy isomers, only the phenylium cation matches the measured collision cross section of the C6H5+ fragment produced from the ionization of bromobenzene. This indicates that the C6H5+ ions generated in our experiments have the cyclic phenylium structure. If the EI ionization of bromobenzene in our experiments produces both linear and cyclic isomers as reported by other groups,40,47 then we must assume that the linear isomers rearrange to the more stable phenylium ions during the injection of the ions into the drift cell where the injection energy is higher than the barrier to isomerization of the C6H5+ ions.60 Formation of Covalent Adducts at Room Temperature. Figure 2 displays the mass spectra obtained following the injection of both the C6H5+ and C6H6+• ions (generated by the EI ionization of benzene) into the drift cell containing He carrier gas with variable amounts of acetylene vapor at room temperature. As shown in Figure 2a, only the C6H5+ and C6H6+• ions are present upon the injection into the drift cell,
Figure 1. Arrival time distributions (ATDs) of the mass-selected C6H5+ fragment resulting from the EI ionization of bromobenzene at different drift voltages (decreasing from left to right). The inset shows a plot of mean arrival time (td) as a function of P/V (Torr V−1).
were completed using Gaussian’s analytic algorithm for all structures to confirm that the structures were minima.56 All relative energies were zero-point corrected.
■
RESULTS AND DISCUSSION Structure of the Phenylium Ion. The C6H5+ ions are generated in our experiments by EI fragmentation of benzene or bromobenzene. It is generally accepted that the C6H5+ fragment formed in the EI ionization of benzene has about 90% of the phenylium cyclic structure depending on the energy of the ionizing electrons and the pressure regime of the experiments.40 The energy barrier required for ring-opening of the phenylium ion was estimated from ICR kinetic measurements to be 2.0 ± 0.3 eV40 in good agreement with ab initio calculations.57 The C6H5 cations can also be generated in a very high yield by EI dissociative ionization of bromobenzene. However, in this case a mixture of the phenylium and acyclic structures could be produced depending on the energy of the ionizing electrons. For example, the percentage of the phenylium structure produced by the EI ionization of bromobenzene (50−70 eV) was estimated to be about 70% based on the different reactivities of the phenylium and acyclic isomers with N2 and C2H2.40,47 In our experiments, EI ionization of benzene or bromobenzene under the high pressure conditions of the supersonic beam expansion of benzene/He or bromobenzene/He mixtures leads to the deposition of a small amount of energy on the ions as evident by the absence of high energy fragments upon the injection of the ions into the drift cell. Furthermore, the high pressure
Table 1. Relative Energies of the Lowest Energy Structures of the C6H5+ Cations Calculated at UPBEPBE/6-31++G** level,56 and the Corresponding Collision Cross Sections in Helium at 300 K Calculated Using the Trajectory Method59
8927
dx.doi.org/10.1021/jp306046w | J. Phys. Chem. A 2012, 116, 8925−8933
The Journal of Physical Chemistry A
Article
as a result of the weak charge-induced dipole interaction. For example, the association complexes of acetylene with the benzene cation [C6D6•+(C2H2)n, n = 1−4] were observed in 0.5 Torr acetylene only at temperatures below 160 K due to the small binding energies of 3−4 kcal/mol.62 The formation kinetics and the covalent nature of the observed adducts are discussed in the following sections. Time-Dependent Product Distribution and Rate Coefficient. The time-dependent product distribution at 304 K is determined by changing the voltage gradient in the drift cell, as shown in Figure 3. Figure 3a shows the mass spectrum
Figure 2. Mass spectra obtained upon the injection of the massselected phenylium and benzene cations (C6H5+ and C6H6+•, respectively) generated by the EI ionization of benzene into the drift cell containing helium and/or purified acetylene gas at different pressures at 300 K. The cell field was 3.8 V.cm−1 and the injection energy (IE) was 11.9 eV (lab).
indicating the absence of other products arising from fragmentation or impurity gases in the drift cell. In the presence of a low concentration of acetylene (0.02 Torr), the C6H5+ ions react very rapidly, with the C2H2 molecules to generate the adduct ions C8H7+, while the C6H6+• ions show no reactivity, as shown in Figure 2b. This is consistent with previous work that showed that the benzene cation does not react with acetylene under ordinary conditions at 300 K.33,61 On the other hand, the sequential addition of acetylene onto the phenylium ion is evident from the appearance of the first adduct C8H7+ at a low concentration of acetylene (Figure 2b) followed by the appearance of the second adduct C10H9+ as the concentration of acetylene increases, as shown in Figure 2c. At higher pressure of acetylene (0.79 Torr) at 300 K, the third adduct C12H11+ starts to appear, while at the same conditions, no addition is observed to the benzene cation (Figure 2d). These results clearly verify the much higher reactivity of the phenylium ion toward the sequential addition of acetylene as compared to the benzene cation. At room temperature, the observed reactions are C6H5+ + HCCH → C8H 7+ +
C8H 7 + HCCH → C10H 9
+
C10H 9+ + HCCH → C12H11+
(m/z 103) (m/z 129) (m/z 155)
Figure 3. Mass spectra obtained upon the injection of the massselected phenylium cation (C6H5+) generated from the EI ionization of bromobenzene into the drift cell containing 0.5 Torr helium and 0.01 Torr purified acetylene gas at different cell voltages at 300 K.
of the C6H5+ ions generated from the EI ionization of bromobenzene and injected into the drift cell containing 0.6 Torr He at 300 K. Figure 3b shows the mass spectrum obtained following the injection of the C6H5+ ions into the drift cell containing 0.01 Torr acetylene + 0.50 Torr He at 300 K. At such a low concentration of acetylene only the first adduct C8H7+ is observed, and by decreasing the voltage gradient in the drift cell, the residence time of the injected C6H5+ ions inside the drift cell increases, which results in increasing the product yield of the C8H7+ ions, as shown in Figure 3b−d. The fast nature of the reaction of the phenylium ion with acetylene is confirmed by measuring the rate of disappearance of C6H5+ ions and the formation of the first adduct C8H7+, as shown in Figure 4. The rate coefficient measured for the C6H5+ ions generated from bromobenzene is 7.1 (±1.4) × 10−10 cm3 s−1 (at 305 K), which is very close (could be considered similar within the experimental uncertainties) to the rate measured for the C6H5+ ions generated from benzene (5.2 (±1.5) × 10−10 cm3 s−1 at 299 K), indicating that the C6H5+ ions have the cyclic phenylium structure in both cases. Also, no difference in the product ions was observed when using the C6H5+ ions
(1) (2) (3)
These product ions formed at room temperature are most likely covalent adducts because the formation of ion−molecule association complexes would require much lower temperatures 8928
dx.doi.org/10.1021/jp306046w | J. Phys. Chem. A 2012, 116, 8925−8933
The Journal of Physical Chemistry A
Article
low pressure regime where three body collisional stabilization is not effective. The high reactivity of the phenylium ion toward acetylene is in sharp contrast to the very low reaction efficiency (10−4) observed for the addition of acetylene on the benzene cation.33 This confirms that the barrier to the aromatic cation ring growth can be overcome by the loss of an aromatic H atom.36 Therefore, the first addition of acetylene onto the phenylium ion to generate the adduct C8H7+ does not involve a barrier, in agreement with the theoretical calculations, which predict that acetylene adds covalently to the phenylium ion in a barrierless exothermic reaction to form the C8H7+ ion.36 High Temperature Measurements. To measure the product distribution at higher temperatures and to investigate the possibility of barriers for the formation of higher products, we carried out a temperature study of the sequential reactions of acetylene with the phenylium ion (generated from the EI ionization of bromobenzene over a wide range of temperature from 300−660 K, as shown in Figure 5). A small increase in the
Figure 4. Integrated arrival time distributions of the reactant (C6H5+) and product (C8H7+) ions as a function of reaction time. (a) C6H5+ ions generated from the EI ionization of bromobenzene were injected into the drift cell containing 0.82 Torr He and 0.008 Torr acetylene at 305 K. (b) C6H5+ ions generated from the EI ionization of benzene were injected into the drift cell containing 0.75 Torr He and 0.008 Torr acetylene at 299 K. Figure 5. Mass spectra obtained upon the injection of the massselected phenylium ion (C6H5+) into the drift cell containing 0.54 Torr purified acetylene at different temperatures. The cell field was 5.9 V·cm−1 and the injection energy (IE) was 11.9 eV (lab).
generated from benzene or bromobenzene, confirming that the ions have the cyclic phenylium structure in both cases. The measured rate coefficient indicates a reaction efficiency (defined here as the ratio of the measured rate coefficient to the Langevin capture rate coefficient taken as 1.5 × 10−9 cm3 s−1)46 of nearly 35−47% at room temperature, consistent with the rate measured by SIFT (6.0 × 10−10 cm3 s−1) for the reaction of cyclic C6H5+ with acetylene in 0.5 Torr He.48 However, the rate coefficient measured by SIFT is larger than that measured by ICR as 1.3 ± 0.3 × 10−10 and 1.7 × 10−10 cm3 s−1, as reported in refs 40 and 49, respectively. The difference could be attributed to contributions from the less reactive acyclic C6H5+ isomers where a rate coefficient of ≤8.0 × 10−12 cm3 s−1 has been reported by SIFT measurements at 0.5 Torr He at room temperature.48 The rate coefficients measured by SIFT and the current ion mobility technique reflect contributions from three body processes, unlike the ICR measurements that reflect the
intensity of the second adduct (C10H9+) is observed by increasing the temperature within the range of 300−400 K, as shown in Figure 5a,b. However, a significant enhancement of the intensity of the second adduct, and to a less extent the third adduct, is observed when the temperature of the drift cell increases within the range of 570−660 K, as shown in Figure 5c−e. This observation confirms the covalent nature of the phenylium ion-acetylene adducts because ion−molecule clusters involving weak ion-induced dipole interactions would not survive at such higher temperatures. Interestingly, above 600 K, the intensity of the third adduct C12H11+ decreases and a new peak corresponding to the ion C12H8+ appears, as shown in 8929
dx.doi.org/10.1021/jp306046w | J. Phys. Chem. A 2012, 116, 8925−8933
The Journal of Physical Chemistry A
Article
Figure 5e. This may suggest an endothermic rearrangement of the adduct C12H11+ by the elimination of an H atom and H2 molecule to generate a more stable C12H8+ ion which could correspond to the radical cation of 1-naphthylacetylene or the polycyclic acenaphthalene. These possibilities will be further considered in the discussion of the ring growth mechanism section. The fact that the formation of the second and third adducts are enhanced at high temperatures suggests that the formation of these products could involve energy barriers. Such barriers may prevent the formation of the adducts at low temperatures such as in interstellar medium, but these adducts may be formed at high temperatures in flames or high temperature regions of solar nebulae.30−32 The presence of a small barrier is confirmed by measuring the overall rate of disappearance of the C6H5+ ions and the generation of the first and second adducts (C8H7+ and C10H9+). Arrhenius plot obtained over a narrow temperature range (300−375 K) yields an activation energy of 1.6 ± 0.6 kcal/mol, as shown in Figure 6. In light of the weak temperature dependence within a narrow range, the measured activation energy should be considered an approximate value.
Figure 7. Mass spectra obtained upon the injection of the massselected phenylium and benzene cations (C6H5+ and C6H6+•, respectively) into the drift cell containing purified acetylene gas at different temperatures. The cell field was 7.4 V·cm−1, the injection energy (IE) was 11.6 eV (lab) except of panel (a) 13.9 eV (lab), and the cell pressure was 0.36, 0.35, 0.34, and 0.32 Torr for panels (a−d), respectively. The apparent increase in the intensity of the m/z 104 peak in panel (a) is due to a minor product resulting from a small fraction of energetic C6H6+• ions formed at early time by higher injection energy. The dots indicate the products of the C6H5+ reactions with acetylene.
temperatures
