Article pubs.acs.org/JPCA
Theoretical Study on the Reaction Mechanisms of CH3O− with O2(X3Σg−) and O2(a1Δg) Hai-xia Lin,† Hai-long Liang,† Guang-hui Chen,*,† Feng-long Gu,*,‡ Wen-guang Liu,† and Shao-fei Ni† †
Department of Chemistry, Shantou University, Guangdong 515063, People’s Republic of China Key Laboratory of Theoretical Chemistry of Environment, Ministry of Education; School of Chemistry and Environment, South China Normal University, Guangzhou 510006, People’s Republic of China
‡
ABSTRACT: The detailed potential energy surfaces (PESs) of poorly understood ion−molecule reactions of CH3O− with O2(X3Σg−) and O2(a1Δg) are accounted for by the density functional theory and ab initio of QCISD and CCSD(T) (single-point) theoretical levels with 6-311++G(d,p) and 6311++G(3df,2pd) basis sets for the first time. For the reaction of CH3O− with O2(X3Σg−) (3R), it is shown that a hydrogenbonded complex 31 is initially formed on the triplet PES, which is 1.8 kcal/mol above reactants 3R at the CCSD(T)// QCISD level, from which all the products P1−P8 can be generated. As to the reaction of CH3O− with O2(a1Δg) (1R), it is found that the two energetically low-lying complexes of 1 1(−31.5 kcal/mol) and 12(−24.1 kcal/mol) are initiated on the singlet PES. Starting from them, a total of seven products may be possible, that is, besides P1, P2, P3, P4, and P8, which are the same as on the triplet PES, there exist also another two products, P9 and P10. For both reactions, taking the thermodynamics and kinetics into consideration, the hydride-transfer species P1(CH2O + HO2−) should be the most favorable product followed by P8(e + CH2O + HO2), which is a secondary product of electron-detachment from P1, and the generation of endothermic P7(17.7 kcal/mol) for the reaction of CH3O− with O2(X3Σg−) is also possible at high temperature, whereas the remaining products are negligible. The measured branching ratio of products for CH3O− with O2(X3Σg−) by Midey et al. is 0.85:0.15 for P1 and P8, and that of CH3O− with O2(a1Δg) is 0.52:0.48 with more P8, which can be rationalized by our theoretical results that P8 on the triplet PES is 4.9 kcal/mol above 3R, whereas both P1 and P8 on the singlet PES are very low-lying at 45.6 and 25.2 kcal/mol below 1R energetically. The measured total reaction rate constant of CH3O− with O2(a1Δg) is k = 6.9 × 10−10 cm3 s−1 at 300 K, which is larger than that of k = 1.1 × 10−12 cm3 s−1 for the reaction of CH3O− with O2(X3Σg−). This is understandable because both P1 and P8 on the singlet PES can be generated barrierlessly, whereas to give all the products on the triplet PES has to pass the barrier of 31(1.8 kcal/mol) at the CCSD(T)//QCISD level. It is expected that the present theoretical study may be helpful for understanding the reaction mechanisms related to CH3O− and even CH3S−. On the other hand, organic methoxide anions (CH3O−) is a persistent organic pollutant (POP)12 that can be generated from the methoxide anion donors, such as the anions CH3O−−CO and CH3OCO−−CO, etc.13 In the recent years, considerable experimental and theoretical efforts have been contributed to the study of methoxy anion (CH3O−).14−19 For example, in 1990, Williams et al.17 studied the reaction surface topographies for hydrogen transfer of isoelectronic systems CH3O− + CH2O using the ab initio molecular orbital (MO) methods; in 2006, Ikeda et al.18 studied the mechanism of the deprotonation of 2-butanone with methoxide anion by the same methods. It should be noted that in 2008, Midey et al.19 measured the rate constants and product branching ratios for the reactions of negative ions CH3O− with O2(X3Σg−) and O2(a1Δg) by the selected ion flow tube (SIFT) technology, and it was deduced that CH3O− could react with both
1. INTRODUCTION Molecular oxygen promises to be a powerful reagent for gasphase chemical degradation and structural elucidation of ions because it appears to induce specific cleavage reactions that reflect the various sites of negative charge in the anion.1,2 The ground state O 2(X 3Σ g −), well-known as an important component of atmosphere, not only is essential for human, animals, plants, and combustion but also is being used in medical and industrial production as well as life support.3−6 Generally, energy transfer between O2(X3Σg−) with organic molecule under the ultraviolet irradiation can generate its excited singlet molecular oxygen O2(a1Δg), which is a reactive oxygen species important in the fields ranging from health and medicine to material sciences.7,8 It is prevalent in both the quiescent and aurorally excited lower ionosphere, which is formed through the combination of photodissociation of O3 and direct electron excitation.9 Although appreciable concentration of O2(a1Δg) has been observed in the ionosphere, little is known about its reactivity in the gas phase.10,11 © XXXX American Chemical Society
Received: July 3, 2012 Revised: November 5, 2012
A
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 1. Optimized structures of reactants and products of reactions CH3O− with O2(X3Σg−) and O2(a1Δg) in the singlet state and triplet states at the B3LYP/6-311++G(d,p) level. Bond lengths are in angstroms and angles in degrees. The values in parentheses are at the QCISD/6-311++G(d,p) level; the values in square bracket are at the BHandHLYP/6-311++G(d,p) level. B
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 2. continued
C
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 2. continued
D
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 2. (a) Optimized structures of intermediates and transition states of reaction CH3O− with O2(X3Σg−) in the triplet state at the B3LYP/6-311++G(d,p) level. Bond lengths are in angstroms and angles in degrees. The values in parentheses are at the QCISD/6-311++G(d,p) level; the values in square brackets are at the BHandHLYP/6-311++G(d,p) level. (b) Optimized structures of intermediates and transition states of reaction CH3O− with O2(a1Δg) in the singlet state at the B3LYP/6-311++G(d,p) level. Bond lengths are in angstroms and angles in degrees. The values in parentheses are at the QCISD/6-311++G(d,p) level.
rate constants at 300 K are k = 1.1 × 10−12 cm3 s−1 and 6.9 × 10−10 cm3 s−1 for reactions of CH3O− with O2(X3Σg−) and O2(a1Δg), respectively, with the branching ratios listed as follows:
O2(X3Σg−) and O2(a1Δg) to produce the most favorable hydrogentransfer product CH2O + HO2− followed by the electron detachment product of it, i.e., e + CH2O + HO2. The measured CH3O− + O2 (X3Σg −) → CH 2O + HO2−
0.85 ΔHrxn = −18.9 kcal/mol
→ e + CH 2O + HO2 0.15 ΔHrxn = 5.0 kcal/mol −
CH3O + O2 (a Δg ) → CH 2O + HO2− 1
0.52 ΔHrxn = −41.6 kcal/mol
→ e + CH 2O + HO2 16.2 ΔHrxn = −16.2 kcal/mol
E
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Figure 3. (a) Schematic of the triplet potential energy surface of the reaction CH3O− with O2(X3Σg−) at the CCSD(T)/6-311++G(3df,2pd)// B3LYP/6-311++G(d, p) level. The values in parentheses are at the CCSD(T)/6-311++G(3df,2pd)//QCISD/6-311++G(d,p) level; the values in square brackets are at the CCSD(T)/6-311++G(3df,2pd)//BHandHLYP/6-311++G(d,p) level. (b) Schematic of singlet the potential energy surface of the reaction CH3O− with O2(a1Δg) at the CCSD(T)/6-311++G(3df,2pd)//B3LYP/6-311++G(d, p) level. The values in parentheses are at the CCSD(T)/6-311++G(3df,2pd)//QCISD/6-311++G(d,p) level. F
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Table 1. Relative Energies (kcal/mol) of the Reactants, Products, Intermediates, and Transition States for the (a) CH3O− with O2(X3Σg−) and (b) CH3O− with O2(a1Δg) Reactions at the B3LYP, BHandHLYP, QCISD, and Single-Point CCSD(T)//B3LYP and CCSD(T)//QCISD Levels species
B3LYPa
CCSD(T)b//B3LYPa
BHandHLYPa −
R(CH3O− + O2(X3Σg−))d P1(CH2O + HO2−) P2(CH3O + O2−) P3(CH(O)OH− + OH) P4(trans-CH(O)O2−(1A′) + H2) P5(CH2(O)OH− + O) P6(trans-CH(O)O2−(3A) + H2) P7(CH2OH + O2−) P8(e + CH2O + HO2) 3 1 3 2 3 3 3 4 3 5 3 6 3 7 3 TS1/2 3 TS1/3 3 TS1/7 3 TS1/P1 3 TS1/P5 3 TS3/4 3 TS3/5 3 TS5/6 3 TS5/P6 3 TS6/P3 3 TSP6/P4 3 TSP7/P2
0.0 −12.6 20.9 −28.4 −40.0 0.2 9.2 14.5 15.7 −14.4 7.2 5.7 −14.1 4.2 −25.3 −14.1 36.4 37.8 0.2 9.8 16.1 6.3 13.3 49.2 −22.6 17.3 55.1
21.6 18.0 3.5 1.6 53.5 −17.4 24.2 79.6
R(CH3O− + O2(a1Δg))e P1(CH2O + HO2−) P2(CH3O + O2−) P3(CH(O)OH− + OH) P4(trans-CH(O)O2−(1A′) + H2) P8(e + CH2O + HO2) P9(cis-CH(O)O2− + H2) P10(CH3 + O3−) 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 TS1/3 1 TS1/4 1 TS1/P1 1 TS4/5 1 TS4/6 1 TS4/P1 1 TS4/P3 1 TS4/P9 1 TS5/P9 1 TS6/7 1 TS6/P3 1 TS7/P4 1 TSP1/P4
0.0 −51.2 −17.7 −67.0 −78.6 −22.9 −83.2 10.1 −39.3 −33.1 1.0 −85.0 −76.1 −88.9 −80.8 42.4 11.5 −38.7 −75.6 −83.3 −48.1 −44.0 −22.3 −26.1 −79.3 −49.6 −62.9 −24.5
0.0 −45.7 −3.4 −62.6 −72.8 −25.2 −77.9 25.6 −31.6 −24.1 4.7 −83.7 −73.8 −86.9 −77.4 43.8 17.2 −30.8 −73.0 −82.4 −15.4 −42.4 −16.0 −19.8 −74.8 −43.9 −57.8 −18.2
3
0.0 −15.5 26.8 −32.4 −42.6 −5.5 14.2 17.7 5.0 1.8 11.5 2.7 −11.4 1.0 −21.2 −11.4 47.0 71.1 2.3
CCSD(T)b//BHandHLYPa
QCISDa
CCSD(T)b//QCISDa
0.0 −9.2
0.0 −13.4
0.0 −15.5
0.8
17.2 0.2 4.7
17.7 4.9 1.8
−15.4
−11.6
Imagc
−
(a) CH3O with O2(X Σg ) 0.0 −7.1 3
−2.8
−1.3
3.1
2.4
0.0 −46.6
0.0 −45.6
−33.0
−25.2
−23.2
−31.5
1.1
1880.6i 1899.6i 275.8i 626.3i 816.4i 762.2i 285.1i 1282.4i 200.6i 816.4i 700.6i 2024.1i
(b) CH3O− with O2(a1Δg) 1
−21.1
G
−30.7
997.2i 1551.8i 365.9i 117.8i 886.9i 1584.4i 1624.5i 745.0i 790.1i 135.2i 1236.8i 802.5i 1106.3i
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
Table 1. continued species
B3LYPa
CCSD(T)b//B3LYPa
BHandHLYPa −
CCSD(T)b//BHandHLYPa
QCISDa
CCSD(T)b//QCISDa
Imagc
(b) CH3O with O2(a Δg) −64.1
1
TSP9/P4
1
−59.0
423.7i
a
The basis set is 6-311++G(d,p) for B3LYP and QCISD. bThe basis set is 6-311++G(3df,2pd) for CCSD(T). cAt the B3LYP/6-311++G(d,p) level. d The total energy of reference reactants 3R at the B3LYP/6-311++G(d,p) level is −265.5172837 au, at the CCSD(T)/6-311++G(3df,2pd)// B3LYP/6-311++G(d,p) level it is −265.0566916 au, at the BHandHLYP/6-311++G(d,p) level it is −265.359884 au, at the CCSD(T)/6-311+ +G(3df,2pd)//BHandHLYP/6-311++G(d,p) level it is −265.0553913 au, at the QCISD/6-311++G(d,p) level it is −264.8696337 au, and at the CCSD(T)/6-311++G(3df,2pd)//QCISD/6-311++G(d,p) level it is −265.0567154 au, respectively. eThe total energy of reference reactants 1R at the B3LYP/6-311++G(d,p) level is −265.4557707 au, at the CCSD(T)/6-311++G(3df,2pd)//B3LYP/6-311++G(d,p) level it is −265.0085649 au, at the QCISD/6-311++G(d,p) level it is −264.8167262 au, and at the CCSD(T)/6-311++G(3df,2pd)//QCISD/6-311++G(d,p) level is −265.008748 au, respectively.
Table 2. Energies (au) of Products for the (a) CH3O− with O2(X3Σg−) and (b) CH3O− with O2(a1Δg) Reactions with Possible Components at the B3LYP and Single-Point CCSD(T)//B3LYP Levels B3LYPa
species −
P1(CH2O(1A1) + HO2−(1A′)) P1′(CH2O(3A2) + HO2−(3A″)) P2(CH3O(2A1) + O2−(2Πg)) P3(CH(O)OH− (2A′) + OH(2Πg)) P4(trans-CH(O)O2−(singlet) + H2(1∑g)) P5(CH2(O)OH−(2A′) + O(3P)) P6(trans-CH(O)O2−(1A′) + H2(1∑g)) P7(CH2OH(2A″) + O2−(2Πg)) P8(e + CH2O(1A1) + HO2(2A″)) P1(CH2O(1A1) + HO2−(1A′)) P1′(CH2O(3A2) + HO2−(3A″)) P2(CH3O(2A1) + O2−(2Πg)) P3(CH(O)OH− (2A′) + OH(2Πg)) P4(trans-CH(O)O2−(singlet)+ H2(1∑g)) P8(e + CH2O(1A1) + HO2(2A″)) P9(cis-CH(O)O2− (1A′) + H2(1∑g)) P10(CH3(2A2) + O3−(2B1)) a
CCSD(T)b//B3LYPa
Σg−)
3
(a) CH3O with O2(X −265.5373124 −265.3970316 −265.4839529 −265.5625518 −265.5883788 −265.4159627 −265.5883663 −265.4935761 −265.4999151 (b) CH3O− with O2(a1Δg) −265.5373124 −265.3970316 −265.4839529 −265.5625518 −265.5883788 −265.4999151 −265.5811114 −265.4366103
−265.0813716 −264.9214328 −265.0140564 −265.1083885 −265.0340639 −264.9838792 −265.1326291 −265.0285754 −265.0318937 −265.0813716 −264.9214328 −265.0140564 −265.1083885 −265.0340639 −265.0318937 −265.0684931 −265.0493611
The basis set is 6-311++G(d,p) for B3LYP. bThe basis set is 6-311++G(3df,2pd) for CCSD(T).
However, without establishing the reaction PESs of CH3O− with O2(X3Σg−) and O2(a1Δg), it is impossible to interpret the reaction mechanisms. In this article, the ion−molecule reaction of CH3O− with O2(X3Σg−) and O2(a1Δg) are investigated theoretically for the first time with the quantum chemical calculation methods.
simplified as CCSD(T)//B3LYP. For the species on the favorable pathways, to obtain more reliable geometrical structures and relative energies, further QCISD30 optimization which includes high-level electron correlation was also employed and correspondingly the single-point energy calculations were denoted as CCSD(T)//QCISD. In addition, it is well-known that the inclusion of zero-point vibrational energy (ZPVE) may lead to the relative energies of transition states below the reactants, which is not accurate to analyze the PES. Herein the relative energies employed are just at the single-point energy calculations of the CCSD(T) level considering the geometries of the B3LYP or QCISD without the ZPVE.
2. COMPUTATIONAL METHODS All the calculations in this work were performed with the Gaussian 03 suite of programs.20 The species found on both the triplet and singlet PESs were characterized by harmonic frequency calculations at the density functional theory (DFT) level, employing the hybrid Becke three-parameter Lee−Yang−Parr exchange correlation functional (B3LYP)21,22 with the standard 6-311++G(d,p) basis set.23−25 If the structures have positive definite Hessian matrices, then the species were identified as stationary points, and the transition states (TSs) showed only one negative eigenvalue in their diagonalized force constant matrices. Connectivity of the optimized transition states with the corresponding isomers was verified by following the reaction coordinate backward and forward with the intrinsic reaction coordinate method (IRC).26 To obtain more reliable energies, the single-point energy calculations at the coupled cluster singles and doubles perturbated estimate of triples [CCSD(T)]27,28 were carried out at the B3LYP/6-311++G(d,p) geometries using the 6-311++G(3df,2pd)29 basis set, which is
3. RESULTS AND DISCUSSION For conciseness, the optimized structures of reactants and products at the B3LYP/6-311++G(d,p) and QCISD/6-311++G(d,p) levels are shown in Figure 1, whereas that of intermediates and transition states obtained at the same levels are depicted in Figure 2a,b, where the eigenvectors of the imaginary frequencies of transition states are denoted by arrows; by means of connection of reactants, intermediates, transition states, and products, the schematic triplet and singlet PESs of the title reactions at the CCSD(T)//B3LYP and CCSD(T)//QCISD levels are plotted in Figure 3a,b, respectively; the energetic data for various species involved on H
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
overcome. Therefore, the pathways to P5 and P6 are kinetically unfavorable. It is clear that three products P1, P7, and P8 can be generated from 31 and are possibly detected experimentally. Among these, P1 is the lowest one with 15.5 kcal/mol below 3R followed by the electron detachment product P8, which is 4.9 kcal/mol above 3 R. Considering that P 7 (17.7 kcal/mol) is an endothermic product of 12.8 kcal/mol above P8, generation of it is thermodynamically unfavorable at room temperature. However, it may be produced at high temperature. Therefore, the hydride-transfer species P1 should be the most competitive product with predominant abundance followed by the second feasible product P8 with neglectable P7 with pathways listed as follows:
the PESs at the B3LYP/6-311++G(d,p) and QCISD/6-311+ +G(d,p) levels are listed in Table 1, where the imaginary frequencies for all the transition states calculated at the B3LYP/6311++G(d,p) level are also listed; the relative energies of the products with possible correlated components for reactions CH3O− with O2(X3Σg−), O2(a1Δg) are listed in Table 2a,b, respectively. For convenient discussion, the total energies of the reactants CH3O− + O2(X3Σg−) (3R) and CH3O− + O2(a1Δg) (1R) are set as zero for references on the triplet and singlet PESs, respectively, and the symbol sTSm/n (s = 1, 3) is used to denote the transition state with spin multiplicity of singlet or triplet connecting the isomers sm and sn. 3.1. Reaction of CH3O− + O2(X3Σg−) on the Triplet PES. 3.1.1. Initial Association. As the component of reactants, CH3O− was optimized at the B3LYP/6-311++G(d,p) level initially at the C3v symmetry with electron configuration of [(1A1)2(2A1)2(3A1)2(4A1)2(1E)2(2E)2(5A1)2(3E)2(4E)2] and the negative charge mainly concentrated on O(−0.727e) atom, as shown in Figure 1. For the reaction of CH3O− with O2(X3Σg−) on the triplet PES, the association of O2(X3Σg−) with CH3O− may have only one pattern after numerous attempts, that is, the H of nucleophile CH3O− should attack the O of O2(X3Σg−) to form a triplet state complex 31(OCH3O2−) of Cs symmetry with a weak hydrogen-bond at 2.033 Å which is 1.8 (1.8) kcal/mol above reactants 3R as shown in Figures 2a and 3a, respectively. Note that the relative energy in parentheses is at the CCSD(T)//QCISD level. But we failed to find the pattern which corresponds to the attack of O of CH3O− on O of O2(X3Σg−). 3.1.2. Reaction Mechanism. After numerous attempts, we failed to locate the TS connecting 31(OCH3O2−) and P1(CH2O + HO2−) at the B3LYP, MP2, and QCISD levels. At the BHandHLYP/6-311++G(d,p) level, the 3TS1/P1 can be obtained, which is just 0.3 kcal/mol above 31(OCH3O2−) at the CCSD(T)//BHandHLYP level, indicating that the hydride can transfer to O2(X3Σg−) almost barrierlessly, as shown in Figure 3a. This is in reasonably consistent with the calculated result at the CCSD(T)//B3LYP level. Note that P1 is energetically 15.5 (15.5) kcal/mol below 3R[0.0 (0.0) kcal/mol], indicating an exothermic product. After P1, a secondary product of electron detachment from HO2−, P8(e + CH2O + HO2) [5.0 (4.9) kcal/mol), can be generated endothermically. Besides the pathway to P1, there exists another low-barrier pathway, that is, 31[1.8 (1.8) kcal/mol] can undergo a 1,2-H-shift (from C to O) via 3TS1/7[2.3 (2.4) kcal/mol] to form an intermediate 37(H2COH···O2), and then dissociate directly to P7(CH2OH + O2−) [17.7 (17.7) kcal/mol]. Although there is another path to P7, which has to pass a quite high barrier of 3TS1/2(47.0 kcal/mol) with a 1,2-H-shift to give 3 2(HOCH2···O2) and then dissociate to P7, it is kinetically unfavorable. To generate the secondary product of P2(26.8 kcal/mol), a very high barrier of TSP7/P2 at 79.6 kcal/mol has to be surmounted, indicating kinetically and thermodynamically unfeasible. Although P3(CH(O)OH− + OH) and P4(trans-CH(O)O2−(1A′)+ H2) on the triplet PES are very low-lying at −32.4 and −42.6 kcal/mol, the largest barrier to them is very high at 71.1 kcal/mol (3TS1/3), so generation of them is far more impossible and the pathways to them will not be discussed further. In addition to P1, P2, P3, P4, P7, and P8, the remaining two products P5(CH2(O)OH− + O) and P6(trans-CH(O)O2−(3A) + H2) may be produced, as shown in Figure 3a. However, to generate them, two high-lying transition states of 3 TS1/P5(21.6 kcal/mol) or 3TS1/3(71.1 kcal/mol) have to be
P1: 3R → 31 → P1 P8 : 3R → 31 → P1 → P8
P7 : 3R → 31 → 3TS1/7 → 37 → P7
3.2. Reaction of CH3O− + O2(a1Δg) on the Singlet PES. 3.2.1. Initial Associations. For the reaction of CH3O− with O2(a1Δg) on the singlet PES, it is shown that the reaction may be initiated by two patterns. The first one corresponds to the attack of oxygen and hydrogen atoms of nucleophile CH3O− on two oxygen atoms of O2(a1Δg) simultaneously in planar to form a Cs symmetry complex 11(H3CO3−) [−31.6 (−31.5) kcal/mol], comprising a OOOCH five-membered ring with a weak O···O bonding and a hydrogen bond at 2.017 and 1.757 Å, respectively; the other pattern corresponds to the association of the O of CH3O− with the O of O2(a1Δg) to give the complex 12(H3CO···O2−) (−24.1 kcal/mol) with a weak O···O bond at 1.925 Å, as shown in Figure 2b. Although the O···O bond (2.017 Å) of complex 11 is slightly longer than that of (1.925 Å) complex 12, there is another C−H···O hydrogen-bond with H···O at 1.757 Å in 11, which may rationalize that 11 is 7.5 kcal/mol more stable than 12 energetically. Note that both 11 and 12 are energetically very low-lying as 31.5 and 24.1 kcal/mol below 1R, respectively. 3.2.2. Reaction Mechanism. At the B3LYP/6-311++G(d,p) level, totally seven dissociation products can be found on the s in g l et PES via t he com pl exes 1 1 (H 3 CO 3 − ) o r 1 2(H3CO···O2−). In addition to P1, P2, P3, P4, and P8, which are the same as on the triplet PES, two other products P9(cisCH(O)O2− + H2) and P10(CH3+O3−) are also found on the plotted singlet PES in Figure 3b. Among these, only P10(CH3+O3−) (25.6 kcal/mol) (1R → 11 → 1TS1/3 → 13 → P10) is energetically higher than the 1R and thus it is difficult to produce, whereas the remaining products P1, P2, P3, P4, P8, and P9 are energetically under 1R. To evaluate the accuracy of the singlereference method of coupled-cluster calculations in describing the species of 1TS1/3 and 13 on the singlet PES, the diagnostic factors (T1) are calculated. It is shown that the T1 values of 1TS1/3 and 1 3 at the CCSD(T)/6-311++G(3df,2dp) level are very small, 0.013 and 0.025, respectively. As we know that the multireference wave function is significant only if the T1 value is greater than 0.045,31,32 this indicates that the single-reference based methods used in this work should be reliable. Therein we will just discuss the more feasible isomerization and dissociation channels from 11 or 12 to the above six products in the followings. Starting from 11, a transition states of 1TS1/4(17.2 kcal/mol) is involved to give P3(−62.6 kcal/mol), P4(−72.8 kcal/mol), and P9(−77.9 kcal/mol) as shown in Figure 3b. Note that the essential barriers to overcome on the minimum energy pathways to P3, P4 I
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
and P9 are very high as 48.8(11→1TS1/4) and 67.7(14→1TS4/ P9) kcal/mol, respectively, indicating kinetically unfavorable. Once starting from 11, P1(CH2O + HO2−) can be formed through 1TS1/P1(−30.7 kcal/mol) with a low barrier of [0.8 (0.8)] kcal/mol, which is 45.6 kcal/mol below the reactants 1R at the CCSD(T)//QCISD level. Taking both the kinetics and thermodynamics into consideration, it is favorable to give P1 via the complex 1 1, and the channel to P1(CH2O + HO2−) can be described as follows:
To identify the favorable products with possible bimolecular components of different electronic state for reactions CH3O− with O2(X3Σg−), O2(a1Δg) on the singlet and triplet PESs, the relative energies of the possible electronic sate are calculated as listed in Table 2a,b, respectively. It is shown that the relative energy of P1[CH2O(1A1) + HO2−(1A′)] in singlet is 0.159 938 8 au lower than that of P1′[CH2O(3A2) + HO2−(3A″)] in triplet at the CCSD(T)//B3LYP level, taking the thermodynamics into consideration, the products in the singlet state are more favorable than that of in the triplet state. To check the reliability of B3LYP calculations, for the involved ten species in the most favorable pathways on the both triplet and singlet PESs to P1, P7 and P8, the geometrical parameters and single-point energies are refined at the QCISD/6-311++G(d,p) and CCSD(T)/6-311++G(3df,2pd)//QCISD/6-311++G(d,p) levels, respectively. As shown in Table 1 as well as Figures 1−3, the calculated geometrical parameters and single-point energies at the B3LYP and CCSD(T)//B3LYP levels are in great agreement with that of at the QCISD and CCSD(T)//QCISD levels. So the present B3LYP and CCSD(T)//B3LYP calculations are reliable for the title reaction. 3.3. Comparisons with Experiment. For the reactions of CH3O− with O2(X3Σg−) and O2(a1Δg), the only experimental study was carried out by Midey et al.19 using selective ion flow tube (SIFT) technology in 2008. As for the reaction of CH3O− with O2(X3Σg−), two products, the hydride-transfer species P1(CH2O + HO2−) and the electron detachment species P8(e + CH2O + HO2), were detected with the respective branching ratios at about 0.85 and 0.15 experimentally, which can be rationalized by the present theoretical calculations that P8 is a secondary product and 20.4 kcal/mol above P1, with the vertical detachment energy (VDE) of HO2− at 35.6 kcal/mol at the CCSD(T)//B3LYP level. In addition, we predict the endothermic product of P7(CH2OH+O2−) (17.7 kcal/mol) to be detected at high temperature. As to the reaction of CH3O− with O2(a1Δg), just the same as that of CH3O− with O2(X3Σg−), the hydride-transfer species of P1(CH2O + HO2−) and the electron detachment product of P8(e + CH2O + HO2) were detected with the branching ratio at 0.52:0.48 experimentally.19 It is also rationalized that P1 and P8 are the first and second most competitive products theoretically, respectively. Note that the branching ratio (0.52:0.48) of P1 and P8 for reaction of CH3O− with O2(a1Δg) is very different from that (0.85:0.15) of the reaction of CH3O− with O2(X3Σg−). This is understandable that although there exists same 20.4 kcal/mol energetic difference between P1 and P8 on the two reaction PESs, note that P1 and P8 on the singlet PES are 45.6 and 25.2 kcal/mol below 1R, respectively, whereas P8 on the triplet PES is 4.9 kcal/ mol above 3R and P1 is 15.5 kcal/mol below 3R at the CCSD(T)//QCISD level, and thus the yield of P8 on the triplet PES should be less than that on the singlet PES. In addition, the measured rate constant for CH3O− with O2(X3Σg−) reaction is 1.1 × 10−12 cm3 s−1 at 300 K, which is just within the detection limit of the SIFT, indicating a very slow reaction, whereas that of CH3O− with O2(a1Δg) at 300 K is 6.9 × 10−10 cm3 s−1, indicating a fast reaction. Our calculations support the above results that all the channels to products on the triplet PES have to undergo the complex 31 which is 1.8(1.8) kcal/mol above 3R, whereas the generations of P1 and P8 from the reaction of CH3O− with O2(a1Δg) are exothermic processes without any barrier. Of course, for some competitive processes, it is desirable to perform kinetic calculations, but such studies are beyond the scope of the present article.
P1: 1R → 11 → 1TS1/P1 → P1
Just like on the triplet PES, the electron detachment product of P8(e + CH2O + HO2) may be produced, which is 25.2 kcal/ mol below the 1R. On the contrary, the P8 on the triplet PES is 4.9 kcal/mol above the 3R at the CCSD(T)//QCISD level. The generation pathway of P8 is listed as follows: P8 : 1R → 11 → 1TS1/P1 → P1 → P8
Starting from 12(H3CO···O2−), only the dissociation product P 2 (CH 3 O + O 2 − ) (−3.4 kcal/mol) can be obtained barrierlessly, which is a charge-transfer species with respect to 1 R(CH3O− + O2). However, the calculated electron affinity (EA) of 37.9 kcal/mol for O2(a1Δg) is slightly larger than 37.3 kcal/mol of CH3O at the G2 level,33 indicating that the charge transfer is not ready to occur. The pathway to P2 can be described as follows: P2 : 1R → 12 → P2
It is shown that the P1, P2, and P8 involved channels are with the lowest barrier processes. Therefore, we believe that the lowest-lying P1(−45.6 kcal/mol) will be produced with the largest branching ratio experimentally. From P1, just the same as the situation on the triplet PES, we can also get the electron detachment product P8, which is 20.4 kcal/mol above P1 and 25.2 kcal/mol below 1R at the CCSD(T)//QCISD level. Considering that P2 is energetically very high-lying as 42.3 and 21.8 kcal/mol above P1 and P8 at the CCSD(T)//B3LYP level, observation of it experimentally is far more impossible, and the hydride-transfer species P1 should be the most competitive product with predominant abundance followed by P8 as the second competitive product, and generation of the remaining products are unfeasible due to the fact that they are either highlying or kinetically unfavorable. The most feasible reaction pathways are listed as follows: P1: 1R → 11 → 1TS1/P1 → P1 P8 : 1R → 11 → 1TS1/P1 → P1 → P8
In summary, for the reactions of CH3O− with O2(X3Σg−) and O2(a1Δg), the most possible products are P1(CH2O + HO2−), P8(e + CH2O + HO2), and P7(CH2OH + O2−). For the hydride-transfer product P1, the electron configurations of CH2O(1A1) and HO2−(1A′) are (1A1)2(2A1)2(3A1)2(4A1)2(1B2)2(5A1)2(1B1)2(2B2)2 and (1A′)2(2A′)2(3A′)2(4A′)2(5A′) 2 (6A′) 2 (1A″) 2 (7A′) 2 (2A″) 2 , respectively; that of HO2(2A″), as the component of electron detachment product P8, is (1A′)2(2A′)2(3A′)2(4A′)2(5A′)2(1A″)2(6A′)2(2A″)2(7A′)1, whereas those of CH2OH(2A) and O2− in P7(CH2OH + O2−) are (1A)2(2A)2(3A)2(4A)2(5A)2(6A)2(7A)2(8A)2(9A)1 and (1σg)2(1σu)2(2σg)2(2σu)2(3σg)2(1πu)4(1πg)3, respectively. J
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
innovative academic team of Shantou University. The financial support from the Project Supported by Guangdong Province Universities and Colleges Pearl River Scholar Funded Scheme (2011) is also acknowledged. We thank Miss Min-min Ma for her help in polishing the language points. The authors are greatly thankful to the reviewers’ invaluable comments.
4. CONCLUSIONS Both triplet and singlet potential energy surfaces (PESs) of ion−molecule reactions of CH3O− with O2(X3Σg−) and O2(a1Δg) are studied theoretically for the first time at the B3LYP/6-311++G(d,p), BHandHLYP/6-311++G(d,p), QCISD/6-311++G(d,p), and single-point CCSD(T)/6-311+ +G(3df,2pd) levels, respectively. On the triplet PES, it is shown that the association of CH3O− with O2(X3Σg−) will generate a hydrogen-bond complex 3 1(OCH3O2−) at 1.8(1.8) kcal/mol above 3R, which will undergo a variety of isomerization and dissociation pathways to give eight dissociation products, P1(CH2O + HO2−), P2(CH3O + O2−), P3(CH(O)OH− + OH), P4(trans-CH(O)O2−(1A′) + H2), P5(CH2(O)OH− + O), P6(trans-CH(O)O2−(3A) + H2), P7(CH2OH + O2−), and P8(e + CH2O + HO2). Among these, hydride-transfer species P1(−15.5 kcal/mol) can be produced via 31 as the primary product followed by P8, which is the electron-detachment product of P1 as the secondary feasible product. In addition, the endothermic product of P7(CH2OH + O2−) (17.7 kcal/mol) is predicted to be detected at the high temperature. The observation of the remaining products is far more impossible due to either high barriers or the products being energetically high-lying. It should be noted that P1 and P8 were detected with branching ratio of 0.85 and 0.15 in the experiment by Midey et al.19 in 2008; the much lower branch ratio of P8 may be due to the fact that P8 is 4.9 kcal/mol above 3R whereas P1 is 15.5 kcal/mol below 3R. The measured rate constant is 1.1 × 10−12 cm3 s−1 at 300 K for the reaction with O2(X3Σg−), indicating a slow reaction, which is in reasonable agreement with our calculations that there exists a barrier of 1.8 kcal/mol between reactants and products. As for the reaction of CH3O− with O2(a1Δg), it is shown that the barrierless association of reactants will generate two low-lying complexes, that is, 11(H3CO3−) and 12(H3CO···O2−), from which seven dissociation products can be found on the singlet PES. Among these, the low-lying P1(−45.6 kcal/mol), which is the proton-transfer species of 11, can be generated almost barrierlessly as the primary product, followed by the electron detachment product of P8(−25.2 kcal/mol) which is just the same as that of the reaction of CH3O− with O2(X3Σg−). Compared with the P1 and P8, the charge-transfer species P2(CH3O + O2−) (−3.4 kcal/ mol) is thermodynamically unfavorable, whereas the generation of P3, P4, P9(cis-CH(O)O2− + H2), and P10(CH3+O3−) has to undergo a variety of isomerization and dissociation pathways with quite high barriers. It should be noted that both P1(CH2O + HO2−) and P8(e + CH2O + HO2) were detected with branching ratios of 0.52 and 0.48 experimentally with a larger rate constant at 6.9 × 10−10 cm3 s−1 at 300 K, which is in reasonable agreement with our calculations that the generations of P1 and P8 are barrierless and exothermic processes from 1R. We hope that the present theoretical research may be helpful for the understanding of the reaction mechanisms related to CH3O− or CH3S−.
■
■
REFERENCES
(1) Schmitt, R. J.; Bierbaum, V. M.; DePuy, C. H. J. Am. Chem. Soc. 1979, 101, 6443−6445. (2) Bierbaum, V. M.; Schmitt, R. J.; DePuy, C. H. Environ. Health Perspect. 1980, 36, 119−124. (3) Sim, M. A.; Dean, P.; Kinsella, J.; Black, R.; Carter, R.; Hughes, M. Anaesthesia 2008, 63, 938−940. (4) Piantadosi, C. A. Undersea Hyperb Med. 2004, 31, 167−177. (5) Morgenthaler, G. W.; Fester, D. A.; Cooley, C. G. Acta Astronaut. 1994, 32, 39−49. (6) Webb, J. T.; Olson, R. M.; Krutz, R. W.; Dixon, G.; Barnicott, P. T. Aviat. Space Environ. Med. 1989, 60, 415−421. (7) Fang, Y. G.; Liu, F. W.; Bennett, A.; Ara, S.; Liu, J. B. J. Phys. Chem. B 2011, 115, 2671−2682. (8) Eyet, N.; Midey, A.; Bierbaum, V. M.; Viggiano, A. A. J. Phys. Chem. A 2010, 114, 1270−1276. (9) Upschulte, B. L.; Marinelli, W. J.; Green, B. D. J. Phys. Chem. 1994, 98, 837−842. (10) Evans, W. F. J.; Hunten, D. M.; Llewellyn, E. J.; Jones, A. V. J. Geophys. Res. 1968, 73, 2885−2896. (11) Lin, H. X.; Chen, G. H.; Liu, H. L.; Li, D.; Huang, X. C.; Liu, W. G.; Jiao, Y. Q. J. Phys. Chem. A 2011, 115, 13581−13588. (12) Haranczyk, M.; Puzyn, T.; Ng, E. G. Environ. Pollut. 2010, 158, 2786−2789. (13) Wang, T. F.; Eichinger, P. C. H.; Dua, S.; Bowie, J. H. Mass Spectrom. 2010, 24, 1895−1910. (14) Nee, M. J.; Osterwalder, A.; Zhou, J.; Neumark, D. M. J. Chem. Phys. 2006, 125, 014306. (15) Chiu, S. W.; Lau, J. K. C.; Li, W. K. J. Phys. Chem. A 2001, 105, 432−441. (16) Laszlo, B.; Alfassi, Z. B.; Neta, P.; Huie, R. E. J. Phys. Chem. A 1998, 102, 8498−8504. (17) Williams, I. H.; Miller, A. B.; Maggiora, G. M. J. Am. Chem. Soc. 1990, 112, 530−537. (18) Ikeda, H.; Yukawa, M.; Niiya, T. J. Chem. Pharm. Bull. 2006, 54, 731−734. (19) Midey, A.; Dotan, I.; Seeley, J. V.; Viggiano, A. A. Int. J. Mass Spectrom. 2009, 280, 6−11. (20) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A. Jr.; Stratmann, R. E.; Burant, J. C.; et al. Gaussian03, revision C.02; Gaussian, Inc.: Pittsburgh, PA, 2003. (21) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (22) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785− 789. (23) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257−2261. (24) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys. 1980, 72, 650−654. (25) Li, Y.; Liu, H. L.; Sun, Y. B.; Li, Z.; Huang, X. R.; Sun, C. C. J. Phys. Chem. A 2010, 114, 2874−2884. (26) Fukui, K. Acc. Chem. Res. 1981, 14, 363−368. (27) Raghavachari, K.; Trucks, G. W.; Pople, J. A.; Head-Gordon, M. Chem. Phys. Lett. 1989, 157, 479−483. (28) Scuseria, G. E.; Schaefer, H. F., III. J. Chem. Phys. 1989, 90, 3700−3704. (29) Li, Y.; Liu, H. L.; Huang, X. R.; Wang, D. Q.; Sun, C. C.; Tang, A. C. J. Phys. Chem. A 2008, 112, 12252−12262. (30) Pople, J. A.; Head-Gordon, M.; Raghavachari, K. J. Chem. Phys. 1987, 87, 5968−5975.
AUTHOR INFORMATION
Corresponding Author
*E-mail: G.H.C.,
[email protected]; F.L.G.,
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by 211 project foundation of Guangdong Province and the construct project about K
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX
The Journal of Physical Chemistry A
Article
(31) Li, J. C.; Song, X. L.; Peng, Z.; Hou, H.; Wang, B. S. J. Phys. Chem. A 2008, 112, 12492−12497. (32) Rienstra-Kiracofe, J. C.; Allen, W. D.; Schaefer, H. F., III. J. Phys.Chem. A 2000, 104, 9823−9840. (33) Curtiss, L. A.; Raghavachari, K.; Redfern, P. C.; Rassolov, V.; Pople, J. A. J. Chem. Phys. 1998, 109, 7764−7776.
L
dx.doi.org/10.1021/jp3065618 | J. Phys. Chem. A XXXX, XXX, XXX−XXX