A Computational Study on the Chemical Fixation of Carbon Dioxide

ARTICLE pubs.acs.org/JPCA

A Computational Study on the Chemical Fixation of Carbon Dioxide with Epoxide Catalyzed by LiBr Salt Ying Ren, Cai-Hong Guo, Jian-Feng Jia, and Hai-Shun Wu* School of Chemistry and Materials Science, Shanxi Normal University, Linfen 041004, China

bS Supporting Information ABSTRACT: The chemical fixation of carbon dioxide with 2,3epoxypropyl phenyl ether catalyzed by LiBr salt to produce a fivemembered cyclic carbonate, 4-(phenoxymethyl)-1,3-dioxolan-2-one, has been extensively investigated at the B3LYP density functional level of theory. The solvent effects have been studied by means of a PCM model. All possible pathways are examined, and their corresponding energetics are demonstrated. Our results reveal that the overall reaction comprises three main steps: epoxide ring-opening, carbon dioxide insertion, and ring-closure of cyclic carbonate, none of which contains significantly large barriers. On the basis of the computed free energies of activation, the rate-determining step can be the ringopening of epoxide or the ring-closure of cyclic carbonate with variation in the reaction conditions in N-methylpyrrolidinone (NMP) solvent. Our calculations indicate that path 2 is more favorable than path 1 in the gas phase, while both of them exist possibly in NMP solvent. The overall reaction is exothermic. Furthermore, the free energy profiles of all reaction pathways along the minima energy path in the gas phase and in NMP solvent were obtained and compared. It is shown that NMP solvent does not change the general trends for the reaction potential energy surfaces.

1. INTRODUCTION Recently, carbon dioxide as an attractive C1 building block in organic synthesis has attracted much attention. It is an abundant, inexpensive, and nontoxic biorenewable resource.1-3 One of the most promising methodologies in this area has been the synthesis of five-membered cyclic carbonates via the reaction of carbon dioxide with epoxides.4,5 These cyclic carbonates have very important practical applications in aprotic polar solvents, fine chemical intermediates, and sources for polymer synthesis.2,6 As early as 1969, Inoue et al. demonstrated that it was possible to copolymerize carbon dioxide and propylene oxide in the presence of a catalyst derived from a 1:1 mixture of (CH3CH2)2Zn and water.7 In recent decades, numerous catalyst systems have been reported for this transformation,5,8-11 such as quaternary ammonium salts,12 alkali metal salts,12-15 halostannanes,16,17 antimony compounds,18 porphyrin,19,20 and transition-metal complexes.21-24 However, most of these catalysts have the following drawback: high pressures/temperatures needed or the low catalytic activity, which limits their applications. In the coupling reaction of carbon dioxide with epoxides, high pressures of CO2 have been thought to be necessary. Recently, Kihara and co-workers reported that the LiBr salt showed high catalytic activity in the reaction of 2,3-epoxypropyl phenyl ether and carbon dioxide under atmospheric pressure (Scheme 1).4 The general mechanism of the title reaction proposed by Kihara et al. includes three elementary steps involving some possible intermediates (Scheme 2). The proposed intermediate 5 was not captured, but it was indirectly proved by a reaction of 2,3-epoxypropyl phenyl ether with LiBr in r 2011 American Chemical Society

Scheme 1

the absence of CO2 in NMP, which leads to 1-phenoxy-2propanone as the rearrangement product of 5. The kinetic analysis indicated that the ring-opening of epoxide (1 f 2) was the rate-determining step of the catalytic cycle (Scheme 3). In the past few years, there were several theoretical studies on the reactions of CO2 with epoxides. For example, heterobimetallic Ru-Mn complexes were first synthesized to mediate the coupling reaction of CO2 with epoxide, and the corresponding reaction mechanism was studied by means of the B3LYP level of density functional theory.25 Sun and Zhang reported the mechanism of the cycloaddition reaction of carbon dioxide with propylene oxide catalyzed by alkylmethylimidazolium chlorine ionic liquids at the B3PW91/6-31G(d,p) level.26 Guo et al. studied the transition-metal-mediated coupling reaction of carbon dioxide with epoxides by the B3LYP methold.27,28 To our best knowledge, no theoretical study has been conduced to understand the mechanism of LiBr-catalyzed reaction of carbon dioxide with 2,3-epoxypropyl phenyl ether. Many mechanistic Received: May 7, 2010 Revised: November 23, 2010 Published: March 01, 2011 2258

dx.doi.org/10.1021/jp104184v | J. Phys. Chem. A 2011, 115, 2258–2267

The Journal of Physical Chemistry A

ARTICLE

Scheme 2

details of the reaction process remain ambiguous. The structural and energetic details about how these intermediates transform to each other are still unclear. Moreover, how many possible pathways does the reaction have? Which pathway is more favorable? These arouse our great interest to study the reaction potential energy surfaces of carbon dioxide with 2,3-epoxypropyl phenyl ether catalyzed by LiBr salt. In this Article, a thorough theoretical study using density functional theory (DFT) on the reaction of carbon dioxide with 2,3-epoxypropyl phenyl ether catalyzed by LiBr salt is reported. Through the study, the detailed structural and energetic information about each step of catalytic cycle is obtained. The Gibbs free energy profiles for the three possible pathways are presented, and the comparison of the three mechanisms is addressed. In addition, the natural bond orbital (NBO) analysis is carried out to illustrate the bond order changes in the process of the reaction, and the solvent effects of NMP are examined by using the self-consistent reaction field theory.

(PCM)44,45 were performed on the optimized gas-phase geometries for all intermediates and transition states at the B3LYP/ 6-31G(d,p) level. The dielectric constant ε was assumed to be 32.2 for the NMP. The solvation free energy was calculated at the B3LYP/6-31G(d,p) level and added to the gas-phase free energy to obtain the Gibbs free energy in solution. To support our choice for the functional and basis set, further single-point B3LYP/6-311G(d,p), B3PW91/6-311G(d,p), and MP2/6-311G(d,p) calculations were performed on some key species optimized at the B3LYP/6-31G(d,p) level. The results are presented in Table S1 (Supporting Information). It is found that the single-point energy differences are small between B3LYP/6-31G(d,p) and B3LYP/6-311G(d,p) and B3PW91/6311G(d,p) levels. In addition, the MP2/6-311G(d,p) level cannot change the reaction mechanism. Thus, these data confirm the reliability of the calculation method (B3LYP/6-31G(d,p)) used for describing the present system.

3. RESULTS AND DISCUSSION 2. COMPUTATIONAL DETAILS All calculations were performed at the density functional theory (DFT)29-31 using the B3LYP32 hybrid functional with the Gaussian 03 program.33 This B3LYP method has been confirmed to be appropriate for the LiBr-catalyzed complex system in a number of recent studies.34-37 For all atoms, the all-electron split valence basis set 6-31G(d,p) is used, which contains polarization functions on heavy atoms and on hydrogen atoms.38-40 The geometries of each species involved in the catalytic cycle were fully optimized without any symmetry constraints. The frequency calculations were carried out at the same level to confirm that the optimized structures were ground states without imaginary frequency (NImag = 0) or transition states with only one imaginary frequency (NImag = 1), and the single imaginary frequency of each transition state displayed the desired displacement orientation. The intrinsic reaction coordinate (IRC)41 was followed at the same theoretical level. All thermodynamic data reported in this Article were estimated at the experimental temperature of 100 °C and pressure of 1.0 atm as presented in ref 4. Wiberg bond indices and natural charges were analyzed using the NBO method.42,43 To clarify the entropy effects, the following discussion is based on the free energies (ΔG) of activation and reaction. For evaluating the solvent effects, single-point self-consistent reaction field (SCRF) calculations based on the polarized continuum model

3.1. Reaction Mechanism in the Gas Phase. In the absence of the NMP solvent, all of the relevant stationary points have been located at the B3LYP/6-31G(d,p) level. The proposed reaction mechanisms for the fixation of carbon dioxide with 2,3epoxypropyl phenyl ether catalyzed by LiBr salt are drawn in Scheme 3. On the basis of Kihara’s proposed mechanism, we postulate three possible pathways of the title reaction. Both path 1 and path 2 involve three elementary steps: epoxide ring-opening, carbon dioxide insertion, and ring-closure of cyclic carbonate. For path 3, by contrast, the activation of CO2 is the first step, and the second step is the ring-opening of epoxide as well as ring-closure of cyclic carbonate. According to these reaction paths, the chemical fixation of carbon dioxide with 2,3-epoxypropyl phenyl ether catalyzed by LiBr salt is thoroughly investigated. The structures of the stable points located on each path, with the main geometrical parameters, are shown in Figures 1, 2, 4, and 6. The free energy profiles along the reaction coordinate are depicted in Figures 3, 5, and 7, where the sum of the free energies of the CO2 þ LiBr þ epoxide is taken as zero energy. After that, the solvent effects are discussed. The 2,3-epoxypropyl phenyl ether exists as three possible isomers a, b, and c. The lowest free energy structure is isomer a, which is more stable than b and c by 0.24 and 0.83 kcal/mol, respectively. Therefore, the following discussion focuses on a as the reactant. The optimized structures 2259

dx.doi.org/10.1021/jp104184v |J. Phys. Chem. A 2011, 115, 2258–2267


ARTICLE

Scheme 3

and relative energies are given in Figure S1 (Supporting Information). 3.1.1. Path 1. Let us first discuss the path 1 mechanism proposed by Kihara and co-workers. As depicted in Figure 1, when the 2,3-epoxypropyl phenyl ether and the lithium bromide are close to each other, complex 1 is formed. Complex 1 is more stable than the separated reactants by 13.91 kcal/mol. As shown by the natural bond order (NBO) analysis, this energy stabilization is caused by strong electrostatic interaction between positively charged Li atom and the negatively charged O1 atom of epoxide. From 1, the attack of Br atom on C2 atom easily takes place, which can be ascribed to the greater steric hindrance of the substitute on C1 atom. In this step, the transition state TS(1/2) has been located, where the C2-O1 bond is breaking and the C2-Br bond is forming. The imaginary frequency is 331.01i cm-1, which is associated with the C2-O1 bond stretching motion. Overcoming the activation barrier, intermediate 2 is generated. The lengths of the C2-Br and C2-O1 bonds

in 2 are 2.044 and 2.431 Å, respectively. Wiberg bond indices of C2-Br and C2-O1 bonds are 0.9176 and 0.0616, which implies that the C2-Br bond has been completely formed and the C2O1 bond has been completely broken. Figure 3 clearly shows that this process is endergonic by 11.54 kcal/mol and needs to overcome the free energy barrier of 34.84 kcal/mol. In the following step, phenoxymethyl group rotation results in the formation of 3. Interestingly, intermediates 2 and 3 are in a fast equilibrium, indicated by the low free energy barrier of 6.32 kcal/mol and the slightly endergonic value of 3.32 kcal/mol. Next, intermediate 3 converts into species 4 via the transition state TS(3/4). The transition vector corresponding to the imaginary frequency (87.78i cm-1) of TS(3/4) indicates that Li atom is migrating from Br atom to O4 atom while Li-Br bond is breaking. The migration of Li atom can be attributed to the larger electronegativity of O4 atom than Br atom. On the basis of the conformation of species 4, the more stable intermediate 5 is produced via the rotation of the bromomethyl group. 2260



ARTICLE

Figure 1. Optimized structures with selected structural parameters (bond distances in angstroms) for the species involved in the ring-opening of epoxide along path 1.

Starting with intermediate 5, the incoming CO2 molecule approaches the Li-O1 bond in 5, forming intermediate 6 that consists of a five-membered ring and a four-membered ring. The O2-C3-O3 bond angle is 141.29°, which implies that CO2 is activated via the transition state TS(5/6). As illustrated in Figure 2, the length of the forming C3-O1 bond is 1.668 Å in TS(5/6). The imaginary frequency mode is associated with the C3-O1 bond stretching motion. In 6, the Li-O1 and Li-O4 distances are 1.913 and 1.979 Å, which are both longer than that in 5. It is indicated that the interaction between Li atom and the O1 and O4 atoms has weakened. These interesting geometrical features provide conditions for the later interaction between Li atom and O2 atom. The profile in Figure 3 shows that the process of CO2 addition is predicted to be exergonic by 0.91 kcal/mol and has a moderate free energy barrier of 16.25 kcal/mol. Subsequently, a new complex 7 is generated through the breaking of Li-O4 bond and the forming of Li-O2 bond, as indicated by the lengths of the Li-O2 and Li-O4 as 1.852 and 5.039 Å, respectively. Next, 7 rotates its phenyl ether group around the C1-C4 bond to lead to intermediate 8. Finally, the nucleophilic attack of O3 atom on C2 atom leads to the formation of the cyclic carbonate and regeneration of the catalyst. Furthermore, the transition state TS(8/9) has been located. Its imaginary frequency is 283.51i cm-1, and the corresponding transition vector is associated with C2-O3 bond stretching motion. As depicted in Figure 2, after the transition

state TS(8/9), complex 9 with the cyclic carbonate coordinated to Li atom through O2 atom is given. From the total energy aspect, complex 9 is more stable than the initial reactants by 19.31 kcal/mol, and its direct dissociation results in the formation of cyclic carbonate and the release of LiBr. From Figure 3, it is clearly seen that the coupling reaction in path 1 involves three major elementary steps. In addition, the ring-closure of cyclic carbonate is concluded to be the rate-determining step, due to the highest energy barrier in the overall cyclic mechanism by 45.14 kcal/mol (1 f TS(8/9)). 3.1.2. Path 2. In contrast with the path 1 mechanism discussed above, there exists another possible mechanism, with the attack of Br atom on C1 atom as the first step and the second step is the CO2 insertion and ring-closure of cyclic carbonate. According to this idea, the path 2 mechanism has been examined for the title coupling reaction. As illustrated in Scheme 3, taking the most stable 1 as the starting point, the five-membered-ring intermediate 10 is generated via transition state TS(1/10). The imaginary vibration mode indicates that the attack of Br atom on C1 atom is accompanied by a simultaneous cleavage of C1-O1 bond. In 10, the cleavage of C1-O1 bond is 2.432 Å, and the formation of C1-Br bond is 2.059 Å. As compared to path 1, TS(1/10) and 10 are predicted to be higher in free energy than TS(1/2) and 2 due to the larger steric hindrance of the substitute group on C1 atom. The free energy profile in Figure 5 clearly shows that from 1 to TS(1/10) is endergonic by 15.98 kcal/mol and needs to overcome 2261



ARTICLE

Figure 2. Optimized structures with selected structural parameters (bond distances in angstroms) for the species involved in carbon dioxide insertion and formation of cyclic carbonate along path 1.

Figure 3. Gibbs free energy profiles for the production of cyclic carbonate along path 1 at the B3LYP/6-31G(d,p) level. Gas-phase Gibbs free energies (-) and solvent-corrected Gibbs free energies (- - -) are given in kcal/mol.

the free energy barrier of 36.46 kcal/mol. These data indicate that this step is less favored than the process (1 f 2). The above results suggest that the ring-opening of the epoxide process is a crucial factor for determining the observed regioselectivity.

The subsequent step is CO2 insertion. As shown in Figure 4, the CO2 molecule attacks on Li and C1 atoms in intermediate 10, forming a complex 11 that consists of a five-membered ring and a four-membered ring. This step goes via the transition state 2262



ARTICLE

Figure 4. Optimized structures with selected structural parameters (bond distances in angstroms) for the species involved in path 2.

TS(10/11), in which the O2-C3-O3 bond angle is 138°. It is indicated that the C-O double bond of CO2 is activated through the transition state TS(10/11). The CO2 insertion process is exergonic by 4.26 kcal/mol. The free energy barrier from 10 to TS(10/11) is calculated to be 17.41 kcal/mol. In 11, the Li-O1 bond distance is elongated relative to that in 10 (1.876 vs 1.698 Å), which implies that the interaction between Li atom and O1 atom begins to weaken. Thus, we have considered the possibility of the tautomerization of 11 through the broken of the Li-O1 bond to form a seven-membered ring. Taking this conformation as the starting point, the optimization leads to 12. As shown in Figure 4, intermediate 12 is a seven-membered ring complex, which is considered the result of the cleavage of Li-O1 bond. The calculated free energy barrier from 11 to the transition state TS(11/12) is 19.38 kcal/mol. The next step is the further isomerization of 12 to form 13. Following the formation of intermediate 13, the ring-closure of cyclic carbonate to yield the product like complex 14 occurs via the transition state TS(13/14). Complex 14 is almost as stable as 9. In 14, the cyclic carbonate is coordinated to Li atom through O1 atom. Next, the product is generated by LiBr decoordination. The activation energy from 13 to TS(13/14) is 20.00 kcal/mol.

The total reaction is calculated to be exergonic by 17.11 kcal/mol as compared to the separate reactants. Similar to path 1, the ringclosure of cyclic carbonate is also concluded to be the ratedetermining step in path 2, which is attributed to the highest energy barrier in the overall cyclic mechanism by 42.55 kcal/mol (1 f TS(13/14)). In addition, the free energy barrier of the ratedetermining step in path 2 is lower than in path 1 by 2.59 kcal/mol. Therefore, path 2 is favored over path 1 in the gas phase. 3.1.3. Path 3. To account for the coupling of carbon dioxide with epoxide, Kihara and co-workers assumed that carbon dioxide may be first activated by the bromine anion (path B in Scheme 2). Herein, we postulate the path 3 mechanism. Figure 6 displays the detailed structures of the intermediates, transition states, and products. First, the bromine anion, instead of ring-opening the epoxide, activates the carbon dioxide molecule by forming the bromo-formate. Meanwhile, the lithium ion activates the 2,3epoxypropyl phenyl ether molecule, leading to a new complex. From Figure 7, it can be seen that the total free energy of the two fragments is largely higher than the initial reactants by 122.72 kcal/mol. It is concluded that path 3 is not competitive with path 1 and path 2 in the gas phase. To explore whether the reaction may proceed along path 3 in solvent, the solvent effects will be discussed in the 2263



ARTICLE


Figure 6. Optimized structures with selected structural parameters (bond distances in angstroms) for the species involved in path 3.

subsequent section. To continue with the next step of the catalytic cycle, intermediate 15 is formed through electrostatic interaction between the Li atom and the O1 and O2 atoms. From 15, the nucleophilic attack of O3 atom to C2 atom induced by the nucleophilic attack of Br anion to C3 atom of CO2

results in the ring-opening of 2,3-epoxypropyl phenyl ether. This step is very clear by observing the vivid transition vectors corresponding to the imaginary frequency of TS(15/16) (295.18i cm-1). IRC calculations reveal that TS(15/16) is indeed associated with intermediate 16. In TS(15/16), the 2264



ARTICLE


O1-C1-C2 bond angle is 97.26° and the C2-O1 distance is 2.142 Å, which indicates that the C2-O1 bond has been partly broken and the C2-O3 bond has been partly formed. In 16, the O1-C1-C2 bond angle is 113.09° and the length of C2-O3 bond is 1.483 Å, which implies that C2-O3 bond is formed. The free energy profile in Figure 7 clearly shows that the formation of 16 is endergonic by 14.36 kcal/mol and has a relatively higher free energy barrier of 43.29 kcal/mol. These data indicate that this step does not easily occur. Next, intermediate 16 converts into product like intermediate 17, as demonstrated by transition state TS(16/17), where Br anion is getting away from C3 atom due to the attack of O1 atom from the front. The frequency calculation on TS(16/17) gives one imaginary frequency of 155.17i cm-1, and the corresponding imaginary vibration mode indicates the formation of C3-O1 bond and the breaking of C3-Br bond. The activation energy barrier from 16 to TS(16/17) is 4.60 kcal/mol. This step is thermodynamically downhill because 17 is predicted to be more stable than 16 by 42.65 kcal/mol. From Figure 7, it is shown that the energy decreases from the reactants to the products, which implies that the process is exothermic. Also, the ring-opening is the rate-determining step in this path. 3.2. Solvent Effects. The above calculations were conducted in the gas phase. To estimate the solvent effects of NMP solution, the single-point energies were calculated for the optimized geometries at the B3LYP/6-31G(d,p) level by using the SCRF method based on PCM in NMP solvent at 100 °C. As a result, three free energy profiles for the reaction in the solution were also obtained and shown in Figures 3, 5, and 7, respectively. Comparison of energies between gas phase and solvent phase reveals some interesting insights into the whole catalytic cycle. From Figures 3, 5, and 7, it can be seen that most of the reaction species are more stable in NMP solvent than in the gas phase. It is therefore summarized that a polar solvent like NMP can stabilize

species involved in catalytic cycle to some extent, but cannot change the general trends for the reaction free energy profiles. The Gibbs free energy of reaction is -5.92 kcal/mol, which is lower than that in the gas by 3.87 kcal/mol. It is indicated that the reaction proceeds more easily in NMP solvent than in the gas phase. Similar to the results in the gas phase, we have investigated three pathways in NMP solvent. In path 1, from Figure 3, it can be found that the free energy is downhill, and this reaction is exothermic and stepwise in NMP solvent. It is very interesting to determine the rate-determining step of the entire reaction in path 1. The ringopening of epoxide and the ring-closure of cyclic carbonate have close activation free energies (47.24 vs 50.55 kcal/mol). Therefore, each of them can be the rate-determining step. For path 2, from Figure 5, it is also revealed that the free energy is downhill and the reaction is exothermic and stepwise. Similarly, in path 2, the ring-opening of epoxide and the ring-closure of cyclic carbonate also can be the rate-determining steps, indicated by the close activation free energies (48.09 vs 49.32 kcal/mol). By comparing the energetics of paths 1 and 2, it can then be observed that the barrier of the ring-opening of epoxide in path 1 is lower than that in path 2 by 0.85 kcal/mol and the barrier of the ring-closure of cyclic carbonate in path 1 is higher than that in path 2 by 1.23 kcal/mol, which implies that paths 1 and 2 both possibly exist in NMP solvent. For path 3, from Figure 7, it is noted that the solvent has a remarkable effect on the relative Gibbs free energy of the species involved. As stated above, path 3 is impossible to continue in the gas phase, because the free energy of bromo-formate and the epoxide with a lithium cation in the gas phase is very high, but they are stabilized in NMP solvent. It is supposed that there is a strong interaction between the two fragments. The free energy of activation in the rate-determining step (reactants f TS(15/ 16)) is 82.90 kcal/mol, and it is higher as compared to that value in the gas phase (43.29 kcal/mol). Thus, path 3 is unfavorable as viewed, and it is ruled out. 2265


The Journal of Physical Chemistry A On the basis of the above studies, we have located three possible pathways for the coupling reaction in the presence of LiBr. The path 1 mechanism is mainly in accordance with the experimental findings by Kihara et al.4 Meanwhile, we have proposed another possible path 2 mechanism, which is an analogue of path 1, and found that paths 1 and 2 both possibly exist in the reaction in NMP solvent. Furthermore, Kihara et al. suggested that the ring-opening of epoxide is the rate-determining step. Our calculations indicate that both the ring-opening of epoxide and the ring-closure of cyclic carbonate may be the ratedetermining step with variation in the reaction conditions (temperature, pressure, and solvent). With respect to path 3, it cannot be the main reaction path, which is in agreement with Kihara’s proposal. To sum the above discussions, although there are no quantitative kinetic and thermodynamic data available for direct comparison, our computations agree qualitatively with these experimental results. In addition, we have also considered the possibility of Nmethylpyrrolidinone as ligand for changing the reaction mechanism. Two competing steps have been examined: (i) epoxide ring-opening in path 1 and (ii) epoxide ring-opening in path 2. It is shown that the free energy of N-methylpyrrolidinone coordination to 1, TS(1/2), and TS(1/10) is exergonic by 11.12, 11.32, and 11.32 kcal/mol (the corresponding values in solvent are 28.20, 25.49, and 25.85 kcal/mol), respectively. Comparing these energy data with the free energy barrier of the two competing steps, it can be found that the NMP as ligand does not change the reaction mechanism.

4. CONCLUSIONS We have investigated the reaction mechanisms for the chemical fixation of carbon dioxide with 2,3-epoxypropyl phenyl ether was catalyst by LiBr to produce 4-(phenoxymethyl)-1,3-dioxolan-2-one in the gas phase and in NMP solvent using DFT at the B3LYP/6-31G(d,p) level and the PCM model. All species involved in the catalytic cycle have been fully characterized to be energy minimum structures for the intermediates or saddle point structures for the transition states. Our results show that the reaction proceeds via three possible pathways. The overall reaction is stepwise and exothermic. Path 2 is more favorable than path 1 in the gas phase, while they both possibly exist in NMP solvent; in contrast, path 3 may be available only under high energetics. Both path 1 and path 2 involve three major elementary steps: epoxide ring-opening, carbon dioxide insertion, and ring-closure of cyclic carbonate. On the basis of the constructed catalytic cycle in Scheme 3 for paths 1 and 2, it is very interesting to determine the rate-determining step of the entire reaction. The ring-opening of epoxide and the ring-closure of cyclic carbonate have close activation free energies. Therefore, each of them can be the rate-determining step with variation in the reaction conditions. It is found that the NMP as solvent does not change the tendency of free energy surface as compared to that found in the gas phase. The overall reaction free energy is calculated to be -5.92 kcal/mol in NMP solvent. The present theoretical results have provided insight into the detailed elementary-step mechanism that helps us to understand the intrinsic properties of the important reaction sequence. ’ ASSOCIATED CONTENT

bS Supporting Information. Figure S1 showing the geometric structures and relative energies for the optimized isomers

ARTICLE

of 2,3-epoxypropyl phenyl ether, Table S1 listing the single-point calculations by using B3LYP/6-31G(d,p), B3LYP/6-311G(d,p), B3PW91/6-311G(d,p), and MP2/6-311G(d,p) on some key species optimized at the B3LYP/6-31G(d,p) level, Figures S2-S4 presenting the vibrational analysis results of 4-(phenoxymethyl)1,3-dioxolan-2-one, Tables S2 and S3 listing calculated excited states of 4-(phenoxymethyl)-1,3-dioxolan-2-one and 2,3-epoxypropyl phenyl ether, and the optimized Cartesian coordinates for all species. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*Tel./fax: þ86 0357 2052468. E-mail: [email protected].

’ ACKNOWLEDGMENT This work was supported by the Natural Science Foundations of China (20871077). ’ REFERENCES (1) Leitner, W. Coord. Chem. Rev. 1996, 153, 257. (2) Behr, A. Carbon Dioxide Activation by Metal Complexes; VCH Publishers: New York, 1988. (3) Marks, T. J.; et al. Chem. Rev. 2001, 101, 953. (4) Kihara, N.; Hara, N.; Endo, T. J. Org. Chem. 1993, 58, 6198. (5) Darensbourg, D. J.; Holtcamp, M. W. Coord. Chem. Rev. 1996, 153, 155. (6) Shaikh, A. A. G.; Sivaram, S. Chem. Rev. 1996, 96, 951. (7) Inoue, S.; Koinuma, H.; Tsuruta, T. J. Polym. Sci., Part B: Polym. Lett. 1969, 7, 287. (8) Kihara, N.; Endo, T. Macromolecules 1994, 27, 6239. (9) Kawanami, H.; Ikushima, Y. Chem. Commun. 2000, 21, 2089. (10) Aida, T.; lnoue, S. J. Am. Chem. Soc. 1983, 105, 1304. (11) Kim, H. S.; Kim, J. J.; Lee, B. G.; Jung, O. S.; Jang, H. G.; Kang, S. O. Angew. Chem., Int. Ed. 2000, 39, 4096. (12) Peppel, W. J. Ind. Eng. Chem. 1958, 50, 767. (13) Sandescu, F.; Scurtu, M. Mater. Plast. 1985, 22, 226. (14) Rokicki, G.; Kuran, W.; Marciniak, B. P. Monatsh. Chem. 1984, 115, 205. (15) Iwasaki, T.; Kihara, N.; Endo, T. Bull. Chem. Soc. Jpn. 2000, 73, 713. (16) Matsuda, H.; Ninagawa, A.; Nomura, R.; Tsuchida, T. Chem. Lett. 1979, 573. (17) Baba, A.; Nozaki, T.; Matauda, H. Bull. Chem. Soc. Jpn. 1987, 60, 1552. (18) Matsuda, H.; Ninagawa, A.; Nomura, R. Chem. Lett. 1979, 1261. (19) Takeda, N.; Inoue, S. Bull. Chem. Soc. Jpn. 1978, 51, 3564. (20) Aida, T.; Inoue, S. J. Am. Chem. Soc. 1983, 105, 1304. (21) Pasquale, R. J. D. J. Chem. Soc., Chem. Commun. 1973, 157. (22) Fujinami, T.; Suzuki, T.; Kamiya, M.; Fukuzawa, S.; Sakai, S. Chem. Lett. 1985, 199. (23) Aresta, M.; Quaranta, E.; Ciccarese, A. J. Mol. Catal. 1987, 41, 355. (24) Jiang, J. L.; Gao, F. X.; Hua, R. M.; Qiu, X. Q. J. Org. Chem. 2005, 70, 381. (25) Man, M. L.; Lam, K. C.; Sit, W. N.; Ng, S. M.; Zhou, Z.; Lin, Z.; Lau, C. P. Chem.-Eur. J. 2006, 12, 1004. (26) Sun, H.; Zhang, D. J. Phys. Chem. A 2007, 111, 8036. (27) Guo, C. H.; Wu, H. S.; Zhang, X. M.; Song, J. Y.; Zhang, X. J. Phys. Chem. A 2009, 113, 6710. (28) Guo, C. H.; Jia, J. F. J. Mol. Struct. (THEOCHEM) 2009, 916, 125. (29) Becke, A. D. Phys. Rev. A 1988, 38, 3098. (30) Lee, C.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785. 2266



ARTICLE

(31) Perdew, J. P.; Wang, Y. Phys. Rev. B 1992, 45, 13244. (32) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (33) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, revision D.01; Gaussian, Inc.: Wallingford, CT, 2004. (34) Wiberg, K.; Sklenak, S. Organometallics 2001, 20, 771. (35) Johansson, P.; Tegenfeldt, J.; lindgren, J. Polymer 2001, 42, 6573. (36) Han, Y. K.; Lee, S. U.; Ok, J. H.; Cho, J. J.; Kim, H. J. Chem. Phys. Lett. 2002, 360, 359. (37) Qi, Y. H.; Feng, D. C.; Li, R.; Feng, S. Y. J. Organomet. Chem. 2009, 694, 771. (38) Ditchfield, R.; Hehre, W. J.; Pople, J. A. J. Chem. Phys. 1971, 54, 724. (39) Hehre, W. J.; Ditchfield, R.; Pople, J. A. J. Chem. Phys. 1972, 56, 2257. (40) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213. (41) Fukui, K. J. Phys. Chem. 1970, 74, 4161. (42) Reed, A. E.; Weinstock, R. B.; Weinhold, F. J. Chem. Phys. 1985, 83, 735. (43) Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899. (44) Miertus, S.; Scrocco, E.; Tomasi, J. Chem. Phys. 1981, 55, 117. (45) Miertus, S.; Tomasi, J. Chem. Phys. 1982, 65, 239.

2267


A Computational Study on the Chemical Fixation of Carbon Dioxide

Recommend Documents