Theoretical investigations toward TMEDA-catalyzed [2 + 4] annulation of allenoate with 1-aza-1,3-diene: mechanism, regioselectivity, and role of the catalyst

Abstract

A theoretical investigation on the mechanisms as well as regioselectivity of N,N,N′,N′-tetramethylethane-1,2-diamine (TMEDA)-catalyzed [2 + 4] annulation of allenoate with 1-aza-1,3-diene leading to functionalized pyridines has been performed using density functional theory (DFT). Multiple possible reaction pathways (A–C) have been characterized, and the most favorable pathway C is remarkably different from the mechanism (i.e. pathway A) proposed in Angew. Chem., Int. Ed., 2013, 52, 8584. Generally, there are several steps in the entire catalytic cycle, including activation of allenoate by TMEDA, nucleophilic attack to 1-aza-1,3-diene, intramolecular cyclization, 1,3-hydrogen shift, hydrogen elimination by TMEDA and desulfonation. In pathway A, the 1,3-hydrogen shift is rate-limiting and takes place before the intramolecular cyclization. In the alternative pathway C, cyclization takes place before the 1,3-hydrogen shift, and it is found that TMEDA can function as a proton shuttle to mediate the 1,3-hydrogen shift and lower the energy barrier significantly. The results presented here demonstrate that the catalyst TMEDA can not only serve as a Lewis base to activate allenoate, but also as a Brønsted acid/base to mediate the 1,3-hydrogen shift process, thus accelerating the reaction. Furthermore, the observed regioselectivity is attributed to the more developed negative charge on the α carbon atom of activated allenoate, the stronger C–H⋯π interaction, as well as hydrogen bond interaction between the two fragments. We believe that the present work is helpful to understand the multiple competing pathways for amine-catalyzed annulation reactions of allenoates with electrophiles, and provides valuable insights for predicting the regioselectivity for this kind of reaction.

---

Introduction

Pyridine rings occur in many important compounds, including natural products, pharmaceuticals, functional materials, supramolecular structures, and also in organocatalysts.^1,2 Due to the wide applications in diverse chemical domains, the pyridine core is one of the most studied aromatics. As a consequence, chemists have developed diverse synthetic methodologies to elaborate this structure. For example, the traditional thermal condensation of carbonyl compounds for synthesis of pyridine³ and some other modern synthetic strategies like Hantzsch approach,^4,5 Chichibabin reaction,^6,7 Mannich reaction,^8,9 Vilsmeier–Haack reaction,^10–12 Bohlmann–Rahtz reaction^13–15 and so on. The various synthetic methods have greatly enriched the pyridine family. However, many of these methodologies require either high temperature or toxic metals as catalysts. The shortcomings thus limit their applications.

During the past decades, non-metal Lewis-base (LB) catalysts, such as N-heterocyclic carbenes (NHCs), phosphines, and amines, have been demonstrated to be able to catalyze the reactions of allene-containing compounds with electrophiles to afford heterocyclic products. Such reactions are initiated by the conjugation of a LB to the allene group producing a zwitterionic intermediate, which can then go through various reaction pathways depending on the nature of LB and the reaction conditions. For example, phosphine-catalyzed [3 + 2] cycloadditions of allenoates^16–19 with electron-deficient electrophiles or [4 + n] annulations of α/γ-substituted allenoates^20–24 with electrophiles have been realized. NHC-catalyzed [2 + 2 + 2] cycloaddition of allenoates to trifluoromethyl ketones have also been reported.²⁵

Noteworthy, nitrogen-based amine LB catalysts, like 1,4-diazabicyclo[2.2.2]octane (DABCO) and 4-(dimethylamino) pyridine (DMAP), have also been applied to catalyze the [2 + 4] cycloaddition of allenoates with enones^26–28 or acyldiazenes²⁹ and the [2 + 2] cycloaddition of allenoates with ketones^30,31 or imines.^32–34 Recently, Shi and Loh developed the amine-catalyzed annulations of allenoates with 1-aza-1,3-dienes for synthesis of highly functionalized pyridines (Scheme 1).³⁵ This method is a metal-free multicomponent reaction (MCR), which is very attractive because of its efficiency, simplicity, and environmental friendly characteristic.

Scheme 1 N,N,N′,N′-Tetramethylethane-1,2-diamine (TMEDA)-catalyzed synthesize of tetrasubstituted pyridines.

In the experimental report, Shi et al. speculated that this is a novel aza-Rauhut–Currier/cyclization/desulfonation cascade reaction (depicted in black in Scheme 2). Specifically, the reaction is initiated by the activation of 2,3-butadienoate 1a by TMEDA giving rise to the zwitterionic intermediate A, which subsequently nucleophilic attack to the 1-aza-1,3-diene 2a forming intermediate B. Then, two consecutive intramolecular proton transfer within B provides intermediate D, which would then undergo intramolecular cyclization to afford tetrahydropyridine adduct E. The following extrusion of TMEDA gives the dihydropyridine F, which can finally deliver the pyridine adduct 3a via desulfonation. Although the proposed reaction mechanism seems reasonable, there are still some ambiguous problems upon closer inspection on the reaction mechanism shown in Scheme 2. For example, (1) the transformation from F to 3a is a desulfonation process accompanied by hydrogen elimination from C4. However, how the hydrogen is eliminated from C4 remains elusive. (2) The negative charge on the allenic moiety is delocalized between C1 and C3 atoms, i.e. the zwitterionic intermediate A is believed to exist in equilibrium with the intermediate A′. Therefore, it is possible that C1 atom can also nucleophilic attack to 2a and afford intermediate 3a′, which may finally produce some other products.^26,29 Nevertheless, no other product was observed in experiment, which means that this alternative pathway is impossible to occur under the condition. Therefore, what factors determine the observed regioselectivity needs to be answered. (3) The previous experimental^29,30 and theoretical study²⁶ on amine-catalyzed [2 + 4]/[2 + 2] cycloadditions of allenoates with electrophiles demonstrated that the nitrogen-based LB catalytic reactions proceed following the reaction mechanism shown in Scheme 3: activation of allene ester, nucleophilic attack to the electrophile, intramolecular cycloaddition accompanied by elimination of catalyst. Therefore, based on these observations, it is possible that intermediate B can first undergo intramolecular cyclocyclization to produce intermediate C′, which can then go through intramolecular 1,3-hydrogen shift to afford F. Since no intermediate has been detected in experiment, it is still very difficult to understand the detailed reaction mechanisms as well as the regioselectivity thoroughly.

Scheme 2 The possible reaction pathways for reaction between 1a and 2a catalyzed by TMEDA.

Scheme 3 Amine-catalyzed [2 + 4] cycloadditions of allenoates with enones and [2 + 2] cycloadditions of allenoates with ketones or imines.

Over the past two decades, density functional theory (DFT) has been demonstrated to be a powerful method to clarify the detailed reaction mechanisms, and predict the regioselectivities and stereoselectivities as well as chemoselectivities with high accuracy in organocatalytic and biological reactions.^36–48 It is worth mentioning that phosphine-catalyzed annulations of allenoate with electrophiles^49–51 and the DABCO-catalyzed [2 + 4] cycloaddition of ethyl allenoate with arylidenoxindoles has been theoretically investigated,²⁶ and the theoretical study has greatly enhanced people's understanding for the mechanistic insights into the catalytic cycle. However, to the best of our knowledge, computational investigations of the reaction between allenoate and 1-aza-1,3-diene catalyzed by TMEDA have not been reported, and obviously the mechanism of this annulation reaction is more complicated than the previously reported amine-catalyzed [2 + 4]/[2 + 2] cycloadditions. Our interests in organocatalytic reactions and the above questions concerning this complex organocatalytic reaction prompt us to investigate this domino process in detail. In the present study, a DFT theoretical investigation towards the title reaction was pursued in order to shed light on details of each elementary step at the molecular level and to reach more comprehensive understanding to the regioselectivity of this interesting catalytic annulation. Furthermore, we would also explore the role of organocatalyst TMEDA in this kind of reactions, which may provide valuable insights for rational design of more efficient organocatalyst to replace TMEDA. This computational work was carried out by using the model reaction shown in Scheme 1, in which R₃ = Ph and R₄ = H.

Computational details

All theoretical calculations were performed using the Gaussian 09 (ref. 52) suite of programs. The geometrical structures of all the stationary points in the free energy profiles were optimized by using M06-2X method^53–55 with 6-31G(d, p) basis set in gas phase. The Berny algorithm was employed for both minimizations and optimizations to transition states.⁵⁶ The corresponding vibrational frequencies were calculated at the same level to take account of the thermal correction to Gibbs free energy and to ascertain whether the structure is a transition state or a minimum. We confirmed that all reactants and intermediates had no imaginary frequencies, and each transition state had only one imaginary frequency. Intrinsic reaction coordinate (IRC) calculations,^57,58 at the same level of theory, were performed to ensure that the transition states led to the expected reactants and products. We then refined the energy by performing single-point energy calculations at the M06-2X/6-311++G(d, p) level based on the M06-2X/6-31G(d, p) optimized structures in solvent toluene using IEFPCM solvent model.^59,60 In the following discussion, the energies obtained by addition of thermal correction at M06-2X/6-31G(d, p) level to the corresponding single-energy at the M06-2X/6-311++G(d, p) level in the solvent are used.

Natural bond orbital (NBO) analyses were performed to assign the atomic charges. Moreover, the non-covalent interaction (NCI) analysis was performed by using the Multiwfn program.⁶¹ Furthermore, the molecular global electrophilicity character is measured by electrophilicity index ω, which is given by the expression of ω = (μ²/2η) (eV),⁶² with respect to the electronic chemical potential μ and the chemical hardness η. Both quantities may be approached in terms of the frontier molecular orbital HOMO and LUMO, ε_H and ε_L, as μ ≈ (ε_H + ε_L)/2 and η ≈ (ε_L − ε_H) respectively. Based on the HOMO energies obtained within the Kohn–Sham scheme, Domingo et al. introduced an empirical (relative) nucleophilicity index, N, defined as N = ε_H(Nu) − ε_H(TCE) (eV).^63–65 This nucleophilicity scale takes tetracyanoethylene (TCE) as a reference.

Results and discussion

Reaction mechanisms

In the present study, multiple possible reaction pathways (shown in Schemes 4 and 5) for the title reaction were thoroughly investigated through DFT calculations. Detailed insights on the different reaction pathways are discussed in the following sections.

Scheme 4 Free energy profile for reaction pathway A.

Scheme 5 The free energy profile of reaction pathway B and C for TMEDA-catalyzed annulation of allenoate with 1-aza-1,3-diene.

Pathway A

In pathway A, TMEDA initiates the reaction by nucleophilic attack on the positively-charged center carbon atom of allene group, i.e. C2 atom. This reaction process is accomplished via transition state TS1A and generates intermediate M1A, in which C1 and C3 atoms are rendered more negative. As shown in Scheme 4, the free energy barrier for this step amounts to 21.7 kcal mol⁻¹ and the generated intermediate M1A is 13.2 kcal mol⁻¹ less stable than the reactant, indicating that this reaction process is highly endergonic. This observation is in agreement with the computational results reported by Huang et al.⁶⁶ Next, M1A can nucleophilic attack to R2. In principle, both C1 and C3 atoms of M1A can initiate the nucleophilic attack to C4 atom of R2. Herein, both the two possibilities were taken into consideration, and the two corresponding transition states TS2A and TS2A′ were located. In TS2A, it is C3 atom that initiates the nucleophilic attack, while in TS2A′, it is C1 atom. Our calculated results reveal that the free energy of TS2A′ is 6.5 kcal mol⁻¹ higher than that of TS2A. Moreover, the resultant intermediate M2A′ is 3.5 kcal mol⁻¹ less stable than M2A. Due to these reasons, we think that C1 atom in M1A can not initiate the nucleophilic attack to R2. Therefore, this possible reaction pathway is excluded. The optimized transition state structures are shown in Fig. 1.

Fig. 1 The optimized transition state structures associated with pathway A (the key distances are shown in Å, and the hydrogens not involved in the reaction are omitted).

In the following, M2A would isomerize to M4A via intramolecular 1,3-hydrogen shift. Direct isomerization should go through a four-membered ring (C1–C2–C3–H) transition state, which is impossible due to the strong strain. Herein, Shi et al. proposed that M2A isomerize to M4A via two consecutive proton transfer processes. Specifically, the negative nitrogen atom N7 can first abstract the hydrogen attached with C3 via the six-membered ring transition state TS3A to afford intermediate M3A. As shown in Scheme 4, the Gibbs free energy barrier for this step amounts to 10.1 kcal mol⁻¹, which is not a high barrier as a consequence of the developed negative charge on N7 atom. Accompanied by the hydrogen transfer from C3 to N7, the negative charge gradually accumulates on C1 atom. This facilitates the subsequent hydrogen transfer from N7 to C1, which is accomplished via an eight-membered ring transition state TS4A with free energy barrier of 24.8 kcal mol⁻¹.

Then, the addition of N7 to C2 atom and extrusion of TMEDA affords intermediate M5A. During this process, the bond distance of N7–C2 changes from 3.71 Å in M4A to 1.93 Å in TS5A and then to 1.42 Å in M5A, indicating that N7–C2 bond is gradually formed (shown in Fig. 2). Simultaneously, the C2–N8 bond changes from 1.52 Å in M4A to 1.58 Å in TS5A and then to 3.06 Å in M5A, indicating that C2–N8 bond is gradually broken. These distance variations demonstrate that N7–C2 bond formation and C2–N8 bond breaking are concerted. The free energy barrier for this single step amounts to 8.0 kcal mol⁻¹, indicating that it is easily to occur. Noteworthy, the free energy barrier for the reversed reaction amounts to 47.4 kcal mol⁻¹, indicating that this reaction process is irreversible.

Fig. 2 The optimized transition state structures and product associated with pathway A (the key distances are shown in Å, and the hydrogens not involved in the reaction are omitted).

Followed by the intramolecular cyclization is the hydrogen elimination from atom C4 instead of desulfonation, which is unanticipated and different from the proposed mechanism in experiment. According to our calculated results, TMEDA can serve as the base to abstract the hydrogen from C4 via transition state TS6A. The energy barrier for this step was calculated to be 11.3 kcal mol⁻¹, indicating that the revised mechanism is reasonable. In the generated intermediate M6A, TMEDA is protonated as TMEDA·H⁺ ammonium ion, which is hydrogen bonded with CO₂Bn group. The last step is desulfonation via transition state TS7A with free energy barrier of only 0.1 kcal mol⁻¹. In the finally formed product P, N7–S bond is totally broken and the proton on TMEDA·H⁺ ammonium cation is transferred to SO₂PMP group. As shown in Scheme 4, the highest free energy barrier for the entire pathway A should be the energy difference between the reactants and the highest stationary point TS4A, i.e. ΔG = 26.5 kcal mol⁻¹, thus the fourth step should be the rate-determining step for the entire reaction pathway.

Water-mediated pathway B

In pathway A, 1,3-hydrogen shift takes place before cyclization. To investigate whether or not the two reaction processes can be reversed, we also investigated other two possible pathways, i.e. pathways B and C. As shown in Scheme 5, pathways A and B diverge from intermediate M2A. In pathway B, intermediate M2A does not isomerize to M4A but first undergo an intramolecular cyclization to afford M3B via transition state TS3B (as shown in Fig. 3). The calculated results reveal that accompanied by the formation of N7–C2 bond, the catalyst TMEDA is simultaneously liberated. The corresponding free energy barrier for TS3B was calculated to be 19.0 kcal mol⁻¹, which is not a high barrier at room temperature. In the following, M3B would isomerize to M4B via 1,3-hydrogen shift. Similarly, direct hydrogen transfer from C3 to C1 is not feasible due to the strong strain associated with the four-membered ring transition state. Since a small amount of water may exist in the reaction system, and protic solvent has been demonstrated to be able to assist the proton transfer process in many reactions.^{50,51,67–72} Herein, we proposed that water can assist the hydrogen transfer from C3 atom to C1 atom. According to our calculated results, water-mediated hydrogen transfer is a one-step process, for which the free energy barrier was calculated to be 28.1 kcal mol⁻¹. Comparing Scheme 5 with 4, we found that the complexation of M4B with TMEDA can actually form M5A, which can then produce product P1 following pathway A. Thus, we will not further discuss the following steps. As shown from Scheme 5, the free energy barrier for pathway B should be the energy difference between M3B and TS4B, i.e. 28.1 kcal mol⁻¹. Apparently, reaction pathway A (ΔG = 26.5 kcal mol⁻¹) is 1.6 kcal mol⁻¹ more energetically favorable than pathway B (ΔG = 28.1 kcal mol⁻¹).

Fig. 3 The optimized transition state structures associated with pathway B and C (the distances are shown in Å, the hydrogens not involved in the reaction are omitted).

TMEDA·H⁺-mediated pathway C

Besides pathway B, we also considered pathway C, in which 1,3-hydrogen shift is mediated by TMEDA·H⁺ ammonium cation. It has been demonstrated that TMEDA·H⁺ can assist the proton transfer process in the organocatalytic reactions.^73,74 In pathway C, we proposed that M3B can eliminate TMEDA and then adsorb TMEDA·H⁺ giving rise to M3C. According to our calculated results, M3C is 8.8 kcal mol⁻¹ more stable than M3B, indicating that this process is thermodynamically favorable. Subsequently, the TMEDA·H⁺ in M3C can first deliver the hydrogen to C1 atom via TS4C with free energy barrier of 20.2 kcal mol⁻¹. In the following, TMEDA in the generated intermediate M4C can abstract the hydrogen from C3 atom via TS5C and afford intermediate M5C with free energy barrier of only 1.3 kcal mol⁻¹. Then, the elimination of TMEDA·H⁺ from M5C generate M4B, which can further produce M5A by complexation with TMEDA. According to our calculated results, the 1,3-hydrogen shift in pathway C requires an energy barrier of 20.2 kcal mol⁻¹, which is 6.3 and 7.9 kcal mol⁻¹ lower than that of pathways A and B, respectively.

As shown in Scheme 5, due to the lower free energy barrier for the 1,3-hydrogen shift step, the first reaction step associated with TS1A become rate-determining. Therefore, the highest free energy barrier for pathway C amounts to 21.7 kcal mol⁻¹. By comparison, pathway C is most energetically favorable in all the above three possible reaction pathways. Therefore, it is possible that the intramolecular cyclization take place before 1,3-hydrogen shift.

Further analyses on the regioselectivity

To explain why C1 atom in M1A is more likely to initiate the nucleophilic attack to R2, we performed NBO charge analysis on M1A as well as non-covalent interaction (NCI) analyses on TS2A and TS2A′. As can be seen from Table 1, NBO charge analysis reveals that the nucleophilic addition of TMEDA to R1 does not make the charge on C2 atom change significantly. However, C1 and C3 atoms become more negative, especially C3, for which the charge changes from −0.429e to −0.592e. In intermediate M1A, C3 is more negative than C1 atom. Therefore, C3 atom is more likely to initiate the nucleophilic attack to R2.

Table 1 The NBO charges of some atoms in R1, TMEDA and M1A (units of e)

	C1	C2	C3	N
R1	−0.509	0.140	−0.429	—
TMEDA	—	—	—	−0.516
M1A	−0.584	0.145	−0.592	−0.309

---

The NCI analysis can distinguish the strong interactions, van der Waals interactions and repulsive steric interactions. As seen from Fig. 4, the green clouds (C–H⋯π interaction) in transition state TS2A are more conspicuous than that in TS2A′, indicating that the interaction between the two fragments of TS2A is stronger. Moreover, as shown in Fig. 3, the hydrogen bond distances between methyl group on TMEDA and –NSO₂– group in TS2A are shorter than that in TS2A′, indicating that the intramolecular hydrogen bond interactions in TS2A are also stronger. On the whole, the more negative charge on C3 atom, the stronger C–H⋯π interaction and hydrogen bond interaction would be responsible for the lower energy of TS2A.

Fig. 4 The non-covalent interaction (NCI) analyses on TS2A and TS2A′.

The nature of the TMEDA catalyst

To better understand the role of TMEDA catalyst in this kind of reaction, we performed global reactivity indexes calculations, which have been frequently used for explanation of the role of catalyst.^75–78 As can be seen from Table 2, the electronic chemical potentials of M1A, μ = −2.08 eV, is higher than that for R1, μ = −4.05 eV, which is in agreement with the NBO charge analysis that there is charge transfer from TMEDA catalyst to R1. Moreover, the coordination of TMEDA catalyst to R1 enhances its nucleophilicity but weakens its electrophilicity, which facilitates the following nucleophilic attack to R2.

Table 2 Electronic chemical potential, (μ, in eV), chemical hardness, (η, in eV), global electrophilicity, (ω, in eV), and global nucleophilicity (N, in eV) of R1, M1A, and R2

	μ	η	ω	N
R1	−4.05	8.37	0.98	2.36
M1A	−2.08	6.52	0.33	5.25
R2	−4.82	6.12	1.90	2.72

---

Conclusion

In the present study, the detailed reaction mechanisms as well as the regioselectivity for TMEDA-catalyzed annulation of allenoate with 1-aza-1,3-diene were investigated. On the whole, the domino process consists of several reaction steps: activation of allenoate by TMEDA giving rise to zwitterionic intermediate M1A, nucleophilic attack of which to 1-aza-1,3-diene affording M2A, intramolecular cyclization accompanied by catalyst elimination, 1,3-hydrogen shift, hydrogen elimination by TMEDA and desulfonation. Three possible reaction pathways (A–C) were taken into consideration. In the experimentally proposed pathway A, 1,3-hydrogen shift, which has the highest free energy barrier of 26.5 kcal mol⁻¹, takes place before the intramolecular cyclization. While in pathway B and C, 1,3-hydrogen shift is followed by the intramolecular cyclization, which is different from the experimentally proposed pathway A. In pathway B, the 1,3-hydrogen shift is mediated by water with free energy barrier of 28.1 kcal mol⁻¹. In pathway C, it is TMEDA·H⁺ that mediate the 1,3-hydrogen shift which requires an energy barrier of only 20.2 kcal mol⁻¹. Due to the lower energy barrier for 1,3-hydrogen shift, the first reaction step with activation free energy barrier of 21.7 kcal mol⁻¹, i.e. activation of allenoate by TMEDA, becomes the rate-determining step in pathway C. On the whole, the free energy barriers for pathways A, B and C are 26.5, 28.1 and 21.7 kcal mol⁻¹, respectively, demonstrating that pathway C is most energetically favorable. Therefore, it can be concluded that the intramolecular cyclization might as well take place before 1,3-hydrogen shift.

NBO charge analysis and NCI analyses on transition states TS2A and TS2A′ reveal that the more negative charge on the α carbon atom in the allene group, the stronger intramolecular C–H⋯π interaction, and hydrogen bond interaction would be responsible for the observed regioselectivity.

Furthermore, we investigated the role of TMEDA by performing global reactivity indexes calculations. The calculated results show that the coordination of TMEDA catalyst to R1 enhances its nucleophilicity but weakens its electrophilicity, which facilitates the following nucleophilic attack to R2. Along with the function for mediating the 1,3-hydrogen shift in the most favorable pathway C, the role of TMEDA in the reaction system has been identified not only as the activator of allenoate towards its reaction with 1-aza-1,3-diene, but also as a proton shuttle to accelerate the reaction. We believe that the multiple roles of TMEDA discovered in this work would be useful for the rational designs for this kind of reactions with high efficiency and selectivity.

Acknowledgements

This work was supported financially by the National Natural Science Foundation of China (No. 21403199 and 21303167) and China Postdoctoral Science Foundation (No. 2013M530340, 2014M552010, and 2015T80776).

References

1. C. Allais, J. M. Grassot, J. Rodriguez and T. Constantieux, Chem. Rev., 2014, 114, 10829.
2. N. B. Patel and S. N. Agravat, Chem. Heterocycl. Compd., 2009, 45, 1343.
3. D. L. Boger, J. S. Panek and M. M. Meier, J. Org. Chem., 1982, 47, 895.
4. D. M. Stout and A. I. Meyers, Chem. Rev., 1982, 82, 223.
5. I. C. Cotterill, A. Y. Usyatinsky, J. M. Arnold, D. S. Clark, J. S. Dordick, P. C. Michels and Y. L. Khmelnitsky, Tetrahedron Lett., 1998, 39, 1117.
6. A. E. Chichibabin and O. A. Zeide, J. Russ. Phys.-Chem. Soc., 1906, 37, 1229.
7. R. L. Frank and R. P. Seven, J. Am. Chem. Soc., 1949, 71, 2629.
8. T. R. Kelly and R. L. Lebedev, J. Org. Chem., 2002, 67, 2197.
9. A. Winter and N. Risch, Synthesis, 2004, 35, 2667.
10. O. Meth-Cohn, B. Narine and B. Tarnowski, J. Chem. Soc., Perkin Trans. 1, 1981, 1531.
11. O. Meth-Cohn and K. T. Westwood, J. Chem. Soc., Perkin Trans. 1, 1984, 1173.
12. B. Gangadasu, P. Narender, S. Bharath, K. M. Ravinder, B. A. Rao, C. Ramesh, B. C. Raju and V. J. Rao, Tetrahedron, 2006, 62, 8398.
13. M. C. Bagley, K. Chapaneri, J. W. Dale, X. Xiong and J. Bower, J. Org. Chem., 2005, 70, 1389.
14. V. S. Aulakh and M. A. Ciufolini, J. Org. Chem., 2009, 74, 5750.
15. V. S. Aulakh and M. A. Ciufolini, J. Am. Chem. Soc., 2011, 133, 5900.
16. C. Zhang and X. Lu, J. Org. Chem., 1995, 60, 2906.
17. C. E. Henry and O. Kwon, Org. Lett., 2007, 9, 3069.
18. Z. Xu and X. Lu, Tetrahedron Lett., 1997, 38, 3461.
19. T. Wang and S. Ye, Org. Biomol. Chem., 2011, 9, 5260.
20. Y. S. Tran and O. Kwon, J. Am. Chem. Soc., 2007, 129, 12632.
21. X.-F. Zhu, J. Lan and O. Kwon, J. Am. Chem. Soc., 2003, 125, 4716.
22. T. Wang and S. Ye, Org. Lett., 2010, 12, 4168.
23. E. Li, Y. Huang, L. Liang and P. Xie, Org. Lett., 2013, 15, 3138.
24. Q. Zhang, L. Yang and X. Tong, J. Am. Chem. Soc., 2010, 132, 2550.
25. L. Sun, T. Wang and S. Ye, Chin. J. Chem., 2012, 30, 190.
26. X.-Y. Chen, M.-W. Wen, S. Ye and Z.-X. Wang, Org. Lett., 2011, 13, 1138.
27. X.-C. Zhang, S.-H. Cao, Y. Wei and M. Shi, Org. Lett., 2011, 13, 1142.
28. K. D. Ashtekar, R. J. Staples and B. Borhan, Org. Lett., 2011, 13, 5732.
29. Q. Zhang, L. G. Meng, J. F. Zhang and L. Wang, Org. Lett., 2015, 17, 3272.
30. T. Wang, X.-Y. Chen and S. Ye, Tetrahedron Lett., 2011, 52, 5488.
31. P. Selig, A. Turockin and W. Raven, Chem. Commun., 2013, 49, 2930.
32. G.-L. Zhao, J.-W. Huang and M. Shi, Org. Lett., 2003, 5, 4737.
33. L.-J. Yang, S. Li, S. Wang, J. Nie and J.-A. Ma, J. Org. Chem., 2014, 79, 3547.
34. C. K. Pei and M. Shi, Chem.–Eur. J., 2012, 18, 6712.
35. Z. G. Shi and T. P. Loh, Angew. Chem., Int. Ed., 2013, 52, 8584.
36. P. H.-Y. Cheong, C. Y. Legault, J. M. Um, N. Çelebi-Ölçüm and K. N. Houk, Chem. Rev., 2011, 111, 5042.
37. L. Xu, M. J. Hilton, X. Zhang, P.-O. Norrby, Y.-D. Wu, M. S. Sigman and O. Wiest, J. Am. Chem. Soc., 2014, 136, 1960.
38. G.-J. Cheng, X. Zhang, L. W. Chung, L. Xu and Y.-D. Wu, J. Am. Chem. Soc., 2015, 137, 1706.
39. X.-K. Guo, L.-B. Zhang, D. Wei and J.-L. Niu, Chem. Sci., 2015, 6, 7059.
40. D. H. Wei, B. L. Lei, M. S. Tang and C. G. Zhan, J. Am. Chem. Soc., 2012, 134, 10436.
41. D. H. Wei, L. Fang, M. S. Tang and C. G. Zhan, J. Phys. Chem. B, 2013, 117, 13418.
42. Y. Qiao, D. H. Wei and J. B. Chang, J. Org. Chem., 2015, 80, 8619.
43. R.-Z. Liao, M. D. Karkas, T. M. Laine, B. Akermark and P. E. M. Siegbahn, Catal. Sci. Technol., 2016, 6, 5031–5041.
44. L. Zhang, Y. Wang, Z.-J. Yao, S. Wang and Z.-X. Yu, J. Am. Chem. Soc., 2015, 137, 13290.
45. A. Dermenci, R. E. Whittaker, Y. Gao, F. A. Cruz, Z. X. Yu and G. B. Dong, Chem. Sci., 2015, 6, 3201.
46. W.-H. Mu, S.-Y. Xia, J.-X. Li, D.-C. Fang, G. Wei and G. A. Chass, J. Org. Chem., 2015, 80, 9108.
47. B. Fang, W. Ren, G. Hou, G. Zi, D.-C. Fang, L. Maron and M. D. Walter, J. Am. Chem. Soc., 2014, 136, 17249.
48. W. Cao, X.-J. Zheng, D.-C. Fang and L.-P. Jin, Dalton Trans., 2015, 44, 5191.
49. Y. Qiao and K. L. Han, Org. Biomol. Chem., 2012, 10, 7689.
50. P. Z. Xie, W. Q. Lai, Z. S. Geng, Y. Huang and R. Y. Chen, Chem.–Asian J., 2012, 7, 1533.
51. E. Mercier, B. Fonovic, C. Henry, O. Kwon and T. Dudding, Tetrahedron Lett., 2007, 48, 3617.
52. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr, T. Vreven, K. N. Udin, J. C. Burant, J. M. Millam, S. S. Yengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Shida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonzalez and J. A. Pople, Gaussian 09, Revision A.02, Gaussian, Inc., Wallingford CT, USA, 2009.
53. Y. Zhao and D. G. Truhlar, Theor. Chem. Acc., 2008, 120, 215.
54. Y. Zhao and D. G. Truhlar, J. Chem. Theory Comput., 2008, 4, 1849.
55. Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41, 157.
56. X. S. Li and M. J. Frisch, J. Chem. Theory Comput., 2006, 2, 835.
57. C. Gonzalez and H. B. Schlegel, J. Chem. Phys., 1989, 90, 2154.
58. C. Gonzalez and H. B. Schlegel, J. Phys. Chem., 1990, 94, 5523.
59. W. Sang-Aroon and V. Ruangpornvisuti, Int. J. Quantum Chem., 2008, 108, 1181.
60. J. Tomasi, B. Mennucci and E. Cancès, J. Mol. Struct.: THEOCHEM, 1999, 464, 211.
61. T. Lu and F. W. Chen, J. Comput. Chem., 2012, 33, 580.
62. R. G. Parr, L. v. Szentpály and S. Liu, J. Am. Chem. Soc., 1999, 121, 1922.
63. L. R. Domingo, E. Chamorro and P. Pérez, J. Org. Chem., 2008, 73, 4615.
64. L. R. Domingo and P. Perez, Phys. Chem. Chem. Phys., 2014, 16, 14108.
65. L. R. Domingo, J. A. Saez and S. R. Emamian, Org. Biomol. Chem., 2015, 13, 2034.
66. G. T. Huang, T. Lankau and C. H. Yu, Org. Biomol. Chem., 2014, 12, 7297.
67. B. Cheng, G. Huang, L. Xu and Y. Xia, Org. Biomol. Chem., 2012, 10, 4417.
68. G. Huang, B. Cheng, L. Xu, Y. Li and Y. Xia, Chem.–Eur. J., 2012, 18, 5401.
69. Y. Z. Xia, Y. Liang, Y. Y. Chen, M. Wang, L. Jiao, F. Huang, S. Liu, Y. H. Li and Z. X. Yu, J. Am. Chem. Soc., 2007, 129, 3470.
70. F. Q. Shi, X. Li, Y. Xia, L. Zhang and Z. X. Yu, J. Am. Chem. Soc., 2007, 129, 15503.
71. L. Zheng, Y. Qiao, J. Chang and M. Lu, Org. Biomol. Chem., 2015, 13, 7558.
72. G. Yang and L. Zhou, Catal. Sci. Technol., 2016, 6, 3378.
73. L. Wang, Q. Ni, M. Blümel, T. Shu, G. Raabe and D. Enders, Chem.–Eur. J., 2015, 21, 8033.
74. Y. Wang, B. H. Wu, L. J. Zheng, D. H. Wei and M. S. Tang, Org. Chem. Front., 2016, 3, 190.
75. Y. Wang, D. Wei, Y. Wang, W. Zhang and M. Tang, ACS Catal., 2016, 6, 279.
76. Y. Wang, X. Guo, M. Tang and D. Wei, J. Phys. Chem. A, 2015, 119, 8422.
77. Z. Y. Li, D. H. Wei, Y. Wang, Y. Y. Zhu and M. S. Tang, J. Org. Chem., 2014, 79, 3069.
78. Y. Wang, X. K. Guo, B. H. Wu, D. H. Wei and M. S. Tang, RSC Adv., 2015, 5, 100147.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra09507k

---