ARTICLE pubs.acs.org/JPCA
Thermal Decomposition of Pentacene Oxyradicals Xiaoqing You,† Dmitry Yu. Zubarev,‡ William A. Lester, Jr.,‡,§ and Michael Frenklach*,†,|| †
Department of Mechanical Engineering, University of California, Berkeley, California 94720-1740, United States Kenneth S. Pitzer Center for Theoretical Chemistry, Department of Chemistry, University of California, Berkeley, California 94720-1460, United States § Chemical Sciences Division and Environmental Energy Technologies Division, Lawrence Berkeley National Laboratory, Berkeley, California 94720, United States )
‡
bS Supporting Information ABSTRACT: The energetics and kinetics of the thermal decomposition of pentacene oxyradicals were studied using a combination of ab initio electronic structure theory and energy-transfer master equation modeling. The rate coefficients of pentacene oxyradical decomposition were computed for the range of 15002500 K and 0.0110 atm and found to be both temperature and pressure dependent. The computational results reveal that oxyradicals with oxygen attached to the inner rings are kinetically more stable than those with oxygen attached to the outer rings. The latter decompose to produce CO at rates comparable to those of phenoxy radical, while CO is unlikely to be produced from oxyradicals with oxygen bonded to the inner rings.
1. INTRODUCTION Chemical reactions underlying the growth and oxidation of polycyclic aromatic hydrocarbons (PAHs) have received increasing attention in recent years. Of intrinsic interest, as an important class of chemical species, PAHs have been observed in hydrocarbon pyrolysis and combustion,1,2 suggested to be the precursors to soot particles,2 formed in the interstellar medium,3,4 and associated with the formation of fullerenes,5 carbon nanotubes,6 and most recently with graphene sheets.7 Also, reactions occurring on surfaces of solid carbon materials, such as graphite and soot, are often modeled by similar reactions at PAH edges. Progress has been made in understanding reaction mechanisms of PAH formation and PAH-edge growth at high temperatures.811 Oxidation of PAH edges, while part of the overall growth processes in oxygen-containing environments, has received substantially less attention. The goal of the present study is to explore PAH-edge degradation mechanisms, focusing on the fate of PAH oxyradicals. A number of studies investigated the oxidation of small aromatic molecules1228 and oxygen chemisorptions at selective sites of two- and three-ring aromatics.2932 According to these studies,14,1619,27,28 at high temperatures, the reaction between a phenyl radical and molecular oxygen leads to the formation of a phenoxy radical, which subsequently decomposes to generate cyclopentadienyl radical, C5H5, and carbon monoxide, CO.
Similarly, a naphthoxy radical is produced from a naphthyl radical reacting with O2 and then decomposes to an indenyl radical and CO.28 Fewer studies have been carried out on edge oxidation of larger PAHs and those pertinent to the present work are all theoretical. Radovic33 examined the energetics and feasible pathways for the oxidation to CO2 of several five- to seven-ring mostly peri-condensed PAHs. Celnik et al.34 computed the rate of O2 reaction with a PAH radical site, generating an oxyradical and an oxygen atom for benzene, naphthalene, anthracene, phenanthrene, pyrene, and benzo[a]anthracene at the B3LYP/ 6-31G(d) level of theory. Their results, in accord with an earlier study of Kunioshi et al.,28 showed that the oxidation rates are similar for all radical sites studied except for armchair sites, which exhibited higher activation energy. However, the decomposition of the oxyradicals generated was not investigated in their study. All of these studies suggest that oxyradicals are not only the major intermediates during oxidation of small aromatics, but also of large PAHs and graphene edges. It is of interest, therefore, to investigate the fate of PAH oxyradicals, which should supplement present knowledge of the graphene-edge growth reaction mechanism11 and shed light on the evolution of graphene edges and soot surfaces during oxidation. Specifically, we want to determine whether graphene-edge oxyradicals decompose similar to phenoxy and naphthoxy radicals and, if so, with what rates. We answer these questions by performing a theoretical study, following the previous work35 on the thermodynamic stability of Received: September 16, 2011 Revised: November 3, 2011 Published: November 04, 2011
r 2011 American Chemical Society
14184
dx.doi.org/10.1021/jp208974b | J. Phys. Chem. A 2011, 115, 14184–14190
The Journal of Physical Chemistry A
ARTICLE
Table 1. Structures of Phenoxy and Pentacene Oxyradicals
Figure 1. Minimum potential energy paths for the thermal decomposition of phenoxy radical at the B3LYP/6-311G(d,p) level. Energies (in kcal/mol) at 0 K relative to reactants are shown for each transition state and local minimum.
graphene oxyradicals. We select the same level of theory and the same molecular substrate, pentacene, as in the prior study, to compute reaction energetics. We next examine the reaction kinetics of four pentacene oxyradicals, as well as that of the phenoxy radical, by solving energy-transfer master equations.
2. COMPUTATIONAL DETAILS 2.1. Potential Energy Surfaces. The potential energy surfaces of the unimolecular decomposition of phenoxy and pentacene oxyradicals were computed using density functional theory (DFT). Geometry optimization and vibrational frequency calculations were performed for all stationary points on the reaction pathways using the B3LYP hybrid functional and the 6-311G(d,p) basis set. Zero-point energies (ZPE) and vibrational frequencies were scaled by a factor of 0.967.36 By inspection of the normal modes of the corresponding imaginary frequencies, transition states were confirmed to connect the reactant and product species. All quantum-chemical calculations were carried out using the Gaussian 03 program package.37 2.2. Reaction Rate Coefficients. The rate coefficients of the thermally activated reaction systems were examined using the MultiWell suite of codes (MultiWell-2011.1).3840 MultiWell solves the time-dependent energy-transfer master equations for a multiwell, multichannel unimolecular reaction system using a Monte Carlo stochastic method. Microcanonical rate constants k(E) were computed using RiceRamspergerKasselMarcus (RRKM) theory. Input parameters, including reaction barriers, vibrational frequencies, and moments of inertia, were obtained from the DFT calculations at the B3LYP/6-311G(d,p) level. The level of theory employed is estimated to produce rate coefficients to within an order-of-magnitude accuracy.4144 Reaction rates were computed over a range of conditions pertinent to hydrocarbon combustion, namely, at temperatures ranging from 1500 to 2500 K and pressures from 0.01 to 10 atm. Lennard-Jones parameters were estimated from an empirical correlation.45 Argon was chosen as the bath gas collider. The exponential-down model with ÆΔEdownæ = 260 cm1 was used to
Figure 2. Minimum potential energy paths for the thermal decomposition of pentacene oxyradical I at the B3LYP/6-311G(d,p) level. Energies (in kcal/mol) at 0 K relative to reactants are shown for each transition state and local minimum.
describe the collisional energy transfer.46 An exact count, with an energy grain size of 10 cm1 for the first segment of the double array and a maximum energy of 500000 cm1, was employed to determine the density and sum of states. The translational and vibrational temperatures were set equal. For each set of initial conditions, the number of stochastic trials was varied from 1 104 to 1 108 to keep the statistical error below 5%. The rate coefficients of thermal decomposition were derived from the exponential decay of the reactant molecule after a period of initial relaxation, as outlined and tested recently by Golden.47 Following this procedure, a MultiWell simulation is started with a thermally activated energy distribution. After a period of initial relaxation, the average value of the internal energy relaxes to a constant value, and a fraction of the reactant molecule begins to decay exponentially. The slope of the decay on a semilog plot yields the rate coefficient of interest.
3. RESULTS AND DISCUSSION 3.1. Pentacene Oxyradicals. We examined four pentacene oxyradicals distinguished by O-atom bonding to carbon atoms, denoted I, II, III, and IV, respectively; these structures are displayed in Table 1. Although the four pentacene oxyradicals look similar in structure, they are quite different in potential energy and thermodynamic stability. Pentacene oxyradicals I, II, and III are 17.1, 14.1, and 2.9 kcal/mol higher in potential energy, respectively, than oxyradical IV at the B3LYP/6-311G(d,p) level. 14185
dx.doi.org/10.1021/jp208974b |J. Phys. Chem. A 2011, 115, 14184–14190
The Journal of Physical Chemistry A Zubarev et al.35 showed that below 1000 K the thermodynamic stability of the four oxyradicals follows the trend of their relative potential energies, that is, the order IV > III > II > I, while above 1000 K oxyradical III becomes thermodynamically more stable than IV because of the larger entropy contribution to the Gibbs free energy. The relative energies and stabilities can be explained by the different fragmentation of the delocalized π-electron system of the precursor pentacene molecule arising from the different locations of the chemisorbed oxygen atom.35 Examination of the decomposition path of the phenoxy radical indicates that the oxygen atom, O1, and the three carbon atoms, C2, C3, and C4 (labeled in Table 1) are involved in breaking existing and forming new CC bonds. It is worthwhile comparing bond lengths of these four atoms for pentacene oxyradicals with those for phenoxy radical. As listed in Table 1, the bond lengths or distances between O1 and C2, C2 and C3, C2 and C4, and C3 and C4 of oxyradical I are closest to those of the phenoxy radical among the four pentacene oxyradicals. For instance, the distance between C3 and C4 is in the order of
Figure 3. Minimum potential energy paths for the thermal decomposition of pentacene oxyradical II at the B3LYP/6-311G(d,p) level. Energies (in kcal/mol) at 0 K relative to reactants are shown for each transition state and local minimum.
ARTICLE
phenoxy < I < II < III < IV. Therefore, the similarity in structure of the pentacene oxyradicals to the phenoxy radical is in the order of I > II > III > IV. An oxyradical with a shorter distance between C3 and C4 is more likely to decompose in the same way as the phenoxy radical and to generate a cyclic intermediate. 3.2. Minimum Energy Path. We explored the potential energy surfaces at the B3LYP/6-311G(d,p) level for the thermal decomposition of the four pentacene oxyradicals. To facilitate analysis of the pentacene oxyradical systems, we also computed the phenoxy reaction system at the same level of theory. A detailed report of the computed molecular properties— electronic energies, geometries, and vibrational frequencies— for the species involved is given in the Supporting Information (Tables S1 and S2). Our results indicate that the decomposition pathways of pentacene oxyradicals I and II leading to CO generation are very similar to that of phenoxy radical, while CO is unlikely to be produced following the same pathways for oxyradicals III and IV. Figure 1 shows the minimum potential-energy paths for the decomposition of the phenoxy radical following different pathways. One proceeds through an electrocyclic mechanism, forming the cyclic intermediate 2, followed by CC bond cleavage to produce intermediate, 3. Elimination of CO from 3 leads to the main products, 4 and CO. The other pathway proceeds through the ring-opening, that is, by breaking either the C2C3 or C2C4 bond. As noted, the ring-opening requires much higher activation energy than the cyclization mechanism, in agreement with earlier studies.1619,27 As shown in Figures 2 and 3, pentacene oxyradicals I and II follow pathways similar to phenoxy radical. However, the reactions generating products 4 and CO from pentacene oxyradicals I and II are less endothermic, by 13.9 and 11.0 kcal/mol, respectively, than those of phenoxy radical. Compared with phenoxy radical, pentacene oxyradical I has a lower barrier (TS12) to generate intermediate 2, while oxyradical II has a higher barrier. The TS12 energy of the pentacene oxyradical I system is lower than that of II by10.4 kcal/mol, which contributes to a faster decomposition rate of oxyradical I.
Figure 4. Relaxed potential energy scan for pentacene oxyradicals III (left) and IV (right) at the B3LYP/6-311G(d,p) level of theory. 14186
dx.doi.org/10.1021/jp208974b |J. Phys. Chem. A 2011, 115, 14184–14190
The Journal of Physical Chemistry A
ARTICLE
Figure 5. Minimum potential energy paths for the thermal decomposition of pentacene oxyradical III (left) and IV (right) obtained by breaking the C2C3 and C2C4 bonds adjacent to the CdO bonds, respectively, at the B3LYP/6-311G(d,p) level. Energies (in kcal/mol) at 0 K relative to reactants are shown for each transition state and local minimum.
Similar to phenoxy radical, pentacene oxyradicals I and II each have distinct possible ring-opening pathways by breaking either of the two neighboring CC bonds. For phenoxy radical, these two pathways are the same. However, for pentacene oxyradicals I and II, the two possible ring-opening pathways are different. One of these pathways, via TS15, is comparable in energy to that of phenoxy radical. The other one, via TS16, is much higher in energy, which makes the fate of 6 essentially unimportant. In contrast to pentacene oxyradicals I and II, no cyclic intermediates were found for pentacene oxyradicals III and IV. A potential energy scan, in which the C3 to C4 distance was varied and other geometrical parameters allowed to relax, was performed to explore possible reaction pathways for III and IV along the reaction coordinate to form a cyclic intermediate, as shown in Figure 4. For oxyradical III, as the two interacting carbon atoms come close to form a bond, only a local inflection point was found, indicating that a local minimum either does not exist or the potential energy well is too shallow. In the case of oxyradical IV, the potential energy continues to increase as the
k¼
two carbon atoms get closer. Figure 5 depicts the decomposition of oxyradicals III and IV through their respective ring-opening pathways. In both of these cases, intermediates 5 and 6 have very high potential energies relative to reactants with potential-energy barriers over 100 kcal/mol. Therefore, it is unlikely for oxyradicals III and IV to produce CO through reaction pathways similar to those of oxyradicals I and II. Furthermore, the ring-opening pathways of oxyradicals III and IV involve internal torsional rotations and therefore form nonplanar molecular structures, which is unlikely to happen for multilayer PAHs. 3.3. Reaction Rate Coefficients. Based on the potential energy surfaces and the molecular properties calculated at the B3LYP/6-311G(d,p) level, we computed the rate coefficients of the thermal decomposition of phenoxy radical and the pentacene oxyradicals. For phenoxy and pentacene oxyradicals I and II, we found that the main decomposition products were carbon monoxide and species 4, and the production of other stable species was negligible for the conditions studied. Therefore, we may write the reactions as follows,
For pentacene oxyradicals I and II, the ring-opening pathway through TS16 can be neglected as its rate coefficient is orders of magnitude smaller than that through TS15; see Figures 2 and 3. Assuming intermediate species 2, 3, and 5 are in steady state, the overall rate coefficients of reactions R13 can be expressed as
k1 f 2 k2 f 3 k3 f 4 ðk5 f 1 þ k5 f 3 Þ þ k1 f 5 k5 f 3 k3 f 4 ðk2 f 1 þ k2 f 3 Þ k2 f 1 k3 f 2 ðk5 f 1 þ k5 f 3 Þ þ k5 f 1 k3 f 5 ðk2 f 1 þ k2 f 3 Þ þ k3 f 4 ðk2 f 1 þ k2 f 3 Þðk5 f 1 þ k5 f 3 Þ
where, for example, k1f2 is the rate coefficient of the elementaryreaction step 1 f 2. The high pressure limits of these rate coefficients, k∞, can be calculated from eq 1 using the respective high-pressure-limit rate coefficients for each of the respective elementary steps of the system (provided in Tables S3S5 of the Supporting Information). Such MultiWell-based results can be summarized by the modified Arrhenius expressions as shown in eqs 24 for phenoxy and pentacene oxyradical I and II, respectively, k1, ∞ ¼ 1:84 107 T 1:93 expð 25819=TÞs1
ð2Þ
k2, ∞ ¼ 9:23 107 T 1:61 expð 25078=TÞs1
ð3Þ
k3, ∞ ¼ 1:11 107 T 1:97 expð 29070=TÞs1
ð4Þ
ð1Þ
Analysis of eq 1 indicated that the cyclic pathway and the ringopening pathway through TS15 both contribute to k∞ and that the rate coefficients of steps 1 f 2 and 1 f 5 are smaller than those of the subsequent steps. The rate coefficients at finite pressures were computed by solving the master equations using the Multiwell suit of codes, and the results are displayed in Figures 68. Their values are also provided in Tables S6S8 of the Supporting Information. To compare our results with those in the literature on phenoxy decomposition, Figure 6 displays two sets of reported experimental results and our computed k1 as a function of temperature and pressure. One experimental study is that of Lin and Lin17 who investigated R1 at 10001580 K and 0.40.9 atm using anisole (C6H5OCH3) and allyl phenyl ether (C6H5OCHCH2) as precursors. These authors reported a rate coefficient of 1011.4(0.2 exp(22100 ( 450/T) s1. The other experimental result 14187
dx.doi.org/10.1021/jp208974b |J. Phys. Chem. A 2011, 115, 14184–14190
The Journal of Physical Chemistry A
Figure 6. Comparison of computed k1 with experimental rate coefficients.
comes from a shock-tube study of phenyl radicals reacting with molecular and atomic oxygen by Frank et al.,14 who determined the rate coefficient to be 7.4 1011 exp(22070/T) s1 in the range of 10201190 K and 1.32.5 bar. Compared with the experimental studies, our computed results are within the uncertainty ranges of the measured values. This agreement can serve as validation of the present method and procedures. We note that the computed k1 is both temperature and pressure dependent. As expected, k1 increases as temperature or pressure rises. At relatively low temperatures, k1 is close to the high-pressure limit, k1,∞, when the pressure is larger than 1 atm. At intermediate temperatures, it is in the falloff region. At high temperatures, it approaches the lowpressure limit due to high internal energy at these temperatures. The pressure dependence of k1 may explain the discrepancy between the experimental studies, because the measurements of Frank et al.14 were carried out at higher pressures than those of Lin and Lin.17 Figure 7 shows the computed rate coefficients k2 and k3 of CO formation from the decomposition of pentacene oxyradicals I and II, respectively. Similar to the phenoxy radical, the deviation of the rate coefficients from their respective high-pressure limits increases with temperature. We also observe that in the temperature range of 15002500 K and the pressure range of 0.0110 atm, the computed values of k2 are 1.6 to 8.2 times those of k3. Figure 8 displays the rate coefficients of CO formation from the decomposition of phenoxy radical and pentacene oxyradicals I and II computed at the high-pressure limit and at 1 atm. As can be seen, the rate coefficients are pressure dependent for all three species. At the high-pressure limit, k1,∞ is larger than both k2,∞ and k3,∞, while at 1 atm, k1 is smaller than k2 and close to k3. The pressure dependence is weaker for the pentacene oxyradicals than for the phenoxy radical, as expected for large molecules. The results obtained indicate that the decomposition rate is determined by the location of the oxygen atom in the pentacene oxyradicals. Oxyradicals with oxygen attached to inner rings are kinetically more stable than those with oxygen attached to outer rings. The latter can generate carbon monoxide at rates comparable
ARTICLE
Figure 7. Rate coefficients of CO formation from pentacene oxyradicals I and II as a function of temperature and pressure: solid lines, pentacene oxyradical I; dashed lines, pentacene oxyradical II.
Figure 8. Rate coefficients of CO formation from phenoxy and pentacene oxyradicals I and II: solid lines, the high-pressure limit; dashed lines, 1 atm.
to phenoxy radicals, while CO is unlikely to be produced from oxyradicals with oxygen bonded to inner rings.
4. CONCLUSIONS We studied the energetics and kinetics of the thermal decomposition of phenoxy and pentacene oxyradicals. The computed rate coefficients for the phenoxy radical agree well with available experimental measurements. Pentacene oxyradicals I and II were found to decompose similar to the phenoxy radical. The decomposition rates of phenoxy and pentacene oxyradicals I and II are 14188
dx.doi.org/10.1021/jp208974b |J. Phys. Chem. A 2011, 115, 14184–14190
The Journal of Physical Chemistry A found to be both temperature and pressure dependent. Pentacene oxyradical I, with oxygen bound to an outer ring and with “free edges” on both sides of the CO bond, has the most structural resemblance to a phenoxy radical. In the temperature range of 15002500 K, the unimolecular decomposition rate of pentacene oxyradical I at the high-pressure limit is slightly lower than that of the phenoxy radical. Pentacene oxyradical II, with oxygen bound to an outer ring and with a free edge on one side and a zigzag edge on the other side of the CO bond, has a lower decomposition rate than that of pentacene oxyradical I. Pentacene oxyradicals III and IV, with oxygen bound to inner rings and with zigzag edges on both sides of the CO bond, are unlikely to decompose and generate CO. The present findings may be generalized to oxyradicals formed on zigzag edges of graphene sheets. Such oxyradical sites can be characterized as follows. The first category has “free edges” on both sides of the CO bond, such as pentacene oxyradical I and phenoxy radical; the second category has a free edge on one side and a zigzag edge on the other side of the CO bond, such as pentacene oxyradical II; and the third category has zigzag edges on both sides of the CO bond, such as pentacene oxyradicals III and IV. Oxyradicals in the first category decompose relatively easily, with little influence from the adjacent rings. Oxyradicals in the second category decompose more slowly, constrained somewhat by the zigzag edge on one side. For oxyradicals in the third category, oxygen is bound to a carbon atom with zigzag edges on both sides, thereby preventing the escape of CO from the graphene edge. The high kinetic stability found for such sites, combined with high thermodynamic stability determined in the previous studies,35,48 imply that inner rings of graphene zigzag edges are highly resistant to oxidation, as are soot particle surfaces with similar edges.
’ ASSOCIATED CONTENT
bS
Supporting Information. Total, relative, and zero-point energies, expectation values of the S2 operator for the unrestricted open-shell calculations, vibrational frequencies, rotational constants, Cartesian coordinates of optimized structures at the B3LYP/6-311G(d,p) level, rate coefficients, and equilibrium constants. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*E-mail:
[email protected].
’ ACKNOWLEDGMENT X.Y. and M.F. were supported by the U.S. Army Corps of Engineers, Humphreys Engineering Center Support Activity, under Contract No. W912HQ-07-C-0044. D.Y.Z. was supported by the National Science Foundation under Grant NSF CHE0809969. W.A.L. and M.F. were supported by the Director, Office of Energy Research, Office of Basic Energy Sciences, Chemical Sciences, Geosciences and Biosciences Division of the U.S. Department of Energy, under Contract No. DE-AC0376F00098. The authors are indebted to Professors David Golden and John Barker for their help with the use of the MultiWell code. ’ REFERENCES (1) Glassman, I.; Yetter, R. Combustion; Academic: Burlington, MA, 2008.
ARTICLE
(2) Haynes, B. S.; Wagner, H. G. Prog. Energy Combust. Sci. 1981, 7, 229–273. (3) Allamandola, L. J.; Tielens, A. G. G. M.; Barker, J. R. Astrophys. J. 1985, 290, L25–L28. (4) Frenklach, M.; Feigelson, E. D. Astrophys. J. 1989, 341, 392. (5) Richter, H.; Grieco, W. J.; Howard, J. B. Combust. Flame 1999, 119, 1–22. (6) Lai, H. J.; Lin, M. C. C.; Yang, M. H.; Li, A. K. Mater. Sci. Eng., C 2001, 16, 23–26. (7) Dato, A.; Frenklach, M. New J. Phys. 2010, 12, 125013. (8) Frenklach, M. Phys. Chem. Chem. Phys. 2002, 4, 2028–2037. (9) Combustion Generated Fine Carbonaceous Particles; Bockhorn, H., D’Anna, A., Sarofim, A. F., Wang, H., Eds.; KIT Scientific: Karlsruhe, Germany, 2009. (10) Wang, H. Proc. Combust. Inst. 2011, 33, 41–67. (11) Whitesides, R.; Frenklach, M. J. Phys. Chem. A 2010, 114, 689–703. (12) Albert, D. R.; Davis, H. F. J. Phys. Chem. Lett. 2010, 1, 1107–1111. (13) Fadden, M. J.; Barckholtz, C.; Hadad, C. M. J. Phys. Chem. A 2000, 104, 3004–3011. (14) Frank, P.; Herzler, J.; Just, T.; Wahl, C. Symp. Int. Combust. 1994, 25, 833–840. (15) Frerichs, H.; Tappe, M.; Wagner, H. G. Symp. Int. Combust. 1991, 23, 21–27. (16) Lin, C.-Y.; Lin, M. C. Int. J. Chem. Kinet. 1985, 17, 1025–1028. (17) Lin, C.-Y.; Lin, M. C. J. Phys. Chem. 1986, 90, 425–431. (18) Liu, R.; Morokuma, K.; Mebel, A. M.; Lin, M. C. J. Phys. Chem. 1996, 100, 9314–9322. (19) Olivella, S.; Sole, A.; Garcia-Raso, A. J. Phys. Chem. 1995, 99, 10549–10556. (20) Park, J.; Xu, Z. F.; Lin, M. C. J. Phys. Chem. A 2009, 113, 5348–5354. (21) Shaddix, C. R.; Brezinsky, K.; Glassman, I. Combust. Flame 1997, 108, 139–157. (22) Sivaramakrishnan, R.; Tranter, R. S.; Brezinsky, K. Proc. Combust. Inst. 2005, 30, 1165–1173. (23) Tokmakov, I. V.; Kim, G. S.; Kislov, V. V.; Mebel, A. M.; Lin, M. C. J. Phys. Chem. A 2005, 109, 6114–6127. (24) Xu, Z. F.; Lin, M. C. J. Phys. Chem. A 2006, 110, 1672–1677. (25) Yu, T.; Lin, M. C. J. Am. Chem. Soc. 1994, 116, 9571–9576. (26) Zhang, H. Y.; McKinnon, J. T. Combust. Sci. Technol. 1995, 107, 261–300. (27) Colussi, A. J.; Zabel, F.; Benson, S. W. Int. J. Chem. Kinet. 1977, 9, 161–178. (28) Kunioshi, N.; Touda, M.; Fukutani, S. Combust. Flame 2002, 128, 292–300. (29) Montoya, A.; Mondragon, F.; Truong, T. N. Fuel Proc. Technol. 2002, 7778, 125–130. (30) Sendt, K.; Haynes, B. S. Proc. Combust. Inst. 2005, 30, 2141–2149. (31) Sendt, K.; Haynes, B. S. Combust. Flame 2005, 143, 629–643. (32) Sendt, K.; Haynes, B. S. J. Phys. Chem. A 2005, 109, 3438–3447. (33) Radovic, L. R. J. Am. Chem. Soc. 2009, 131, 17166–17175. (34) Celnik, M. S.; Sander, M.; Raj, A.; West, R. H.; Kraft, M. Proc. Combust. Inst. 2009, 32, 639–646. (35) Zubarev, D. Y.; Robertson, N.; Domin, D.; McClean, J.; Wang, J.; Lester, W. A., Jr.; Whitesides, R.; You, X.; Frenklach, M. J. Phys. Chem. C 2010, 114, 5429–5437. (36) Johnson, R. D., III, Ed. NIST Computational Chemistry Comparison and Benchmark Database; National Institute of Standards and Technology: Gaithersburg, MD, Aug 2005; http://srdata.nist.gov/cccbdb. (37) Frisch, M. J.; Trucks, G. M.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C. et al. Gaussian 03, Revision A.1; Gaussian, Inc.: Pittsburgh, PA, 2003. (38) Barker, J. R. Int. J. Chem. Kinet. 2001, 33, 232–245. (39) Barker, J. R.; Ortiz, N. F.; Preses, J. M.; Lohr, L. L.; Maranzana, A.; Stimac, P. J.; Nguyen, T. L.; Kumar, T. J. D. MultiWell-2011.1; 14189
dx.doi.org/10.1021/jp208974b |J. Phys. Chem. A 2011, 115, 14184–14190
The Journal of Physical Chemistry A
ARTICLE
University of Michigan: Ann Arbor, MI, 2011. http://aoss.engin.umich. edu/multiwell/. (40) Barker, J. R. Int. J. Chem. Kinet. 2009, 41, 748–763. (41) Whitesides, R.; Kollias, A. C.; Domin, D.; Lester, W. A., Jr.; Frenklach, M. Proc. Combust. Inst. 2007, 31, 539–546. (42) Whitesides, R.; Domin, D.; Salomon-Ferrer, R.; Lester, W. A.; Frenklach, M. J. Phys. Chem. A 2008, 112, 2125–2130. (43) Whitesides, R.; Domin, D.; Salomon-Ferrer, R.; Lester, W. A., Jr; Frenklach, M. Proc. Combust. Inst. 2009, 32, 577–583. (44) You, X. Q.; Whitesides, R.; Zubarev, D.; Lester, W. A., Jr; Frenklach, M. Proc. Combust. Inst. 2011, 33, 685–692. (45) Wang, H.; Frenklach, M. Combust. Flame 1994, 96, 163–170. (46) Hippler, H.; Troe, J.; Wendelken, H. J. J. Chem. Phys. 1983, 78, 6709–6717. (47) Golden, D. M. Z. Phys. Chem. 2011, 225, 1–14. (48) Zubarev, D. Y.; You, X.; McClean, J.; Lester, W. A., Jr.; Frenklach, M. J. Mater. Chem. 2011, 21, 3404–3409.
14190
dx.doi.org/10.1021/jp208974b |J. Phys. Chem. A 2011, 115, 14184–14190
