Cyclization mechanisms of the cyclic dimer of aziridine aldehyde with vinyl aldehyde

Cyclization mechanisms of the cyclic dimer of aziridine aldehyde with vinyl aldehyde

Computational and Theoretical Chemistry 1113 (2017) 105–109 Contents lists available at ScienceDirect Computational and Theoretical Chemistry journa...

971KB Sizes 2 Downloads 47 Views

Computational and Theoretical Chemistry 1113 (2017) 105–109

Contents lists available at ScienceDirect

Computational and Theoretical Chemistry journal homepage: www.elsevier.com/locate/comptc

Cyclization mechanisms of the cyclic dimer of aziridine aldehyde with vinyl aldehyde Xueli Cheng School of Chemistry and Chemical Engineering, Taishan University, Tai’an, Shandong 271000, China

a r t i c l e

i n f o

Article history: Received 27 December 2016 Received in revised form 9 March 2017 Accepted 13 May 2017 Available online 17 May 2017 Keywords: Aziridine aldehyde CH2@CHCHO M06-2X Polycyclic compounds 8-exo-trig aldol

a b s t r a c t Aziridine aldehydes are the notable representative of a limited number of reported amphoteric molecules containing aziridine and aldehyde groups, which can react both as an acid and as a base. In the present work, the intriguing dimerization mechanism of aziridine aldehyde free from any substituents and solvents, as well as the addition and cyclization mechanisms of its dimer with vinyl aldehyde in acetonitrile were investigated systematically. The NAH bond of aziridine aldehyde couples with the C@O bond in another aziridine aldehyde molecule, and then the OAH bond couples to the adjacent C@O bond to form a stable polycyclic dimer. In acetonitrile, aziridine aldehyde and CH2@CHCHO are inclined to cyclize to polycyclic compounds. The addition reaction of the dimer and CH2@CHCHO will form various enol and keto isomers, and the enol isomers cyclizing to 8-exo-trig aldol is energetically favorable. Ó 2017 Elsevier B.V. All rights reserved.

1. Introduction Amphoteric molecules comprise nucleophilic and electrophilic functional groups, so they can react both as an acid and as a base. In reversible and irreversible polar reactions, they can offer a versatile platform for the formation of bifunctional structures with high bond-forming efficiency and atom economy [1–3]. Importantly, amphoteric molecules with amine/imine and aldehyde groups can intramolecularly couple to various cyclic compounds and chiral structures in synthetic systems and biochemistry [4–7]. Although there are a limited number of reported amphoteric molecules, their reactivity has been attracted much attention due to their participation in multicomponent and domino processes [8–10]. A notable representative of amphoteric molecules is aziridine aldehydes, which contains NH aziridine and CHO aldehyde groups. So far, a variety of aziridines with varied functional groups have been synthesized [11,12], and the reaction mechanisms of aziridines with aldehydes were investigated experimentally and theoretically [13–17]. As to the dimerization of aziridine aldehydes, in 2012, Assem et al. [18] carried out an experimental investigation on their reversible dimerization mechanisms; in 2014, Belding and coworkers [19] reported a density functional theory (DFT) study on the multicomponent reaction between an aziridine aldehyde dimer, isocyanide and l-proline. However, up to now, there are no systematical theoretical investigations on the dimerization and the reaction mechanisms of the dimer of the simplest E-mail address: [email protected] http://dx.doi.org/10.1016/j.comptc.2017.05.013 2210-271X/Ó 2017 Elsevier B.V. All rights reserved.

aziridine aldehyde with vinyl aldehyde to form polycyclic structures, which will be discussed in the present work. This study will help us understand the fundamental chemistry of amphoteric aziridine aldehydes. 2. Computational details All molecular structures were fully optimized without any constraints at M062X/6-31G(d, p) level with G09 program package [20]. The M06-2X method was selected to cope with the weak interactions in intermediates and transition states because it has good performance in the main-group thermochemistry and in describing noncovalent interactions [21–24]. Frequency calculations were also performed to identify whether the stationary point was a transition state or a local minimum, and to obtain energies after various corrections. A transition state has only one imaginary frequency, and its vibrational mode will correspond to a stretch mode of a breaking bond, or an asymmetrical stretch mode or bend-in-plane mode of a group/atom-shifting process. Then key transition states were further validated by intrinsic reaction coordinate (IRC) calculations [25,26] to confirm their connections. Initially, the dimerization of aziridine aldehyde devoid of any solvents and substituents was investigated. Then the addition reaction mechanism of the dimer of aziridine aldehyde and vinyl aldehyde CH2@CHCHO was explored in acetonitrile as solvent, and the cyclization mechanisms of their intermediates formed from aziridine aldehyde and CH2@CHCHO were further discussed to determine the feasibility of electrophile-initiated domino processes.

106

X. Cheng / Computational and Theoretical Chemistry 1113 (2017) 105–109

The solvation effect was considered using polarizable continuum model (PCM) [27–29] to describe acetonitrile.

3. Results and discussion 3.1. Dimerization of aziridine aldehyde The lowest frequencies and their vibrational mode assignments, electronic and zero-point energies E, thermal enthalpies H, thermal free energies G and entropy involved in this work are listed in Table S1, and the coordinates for all species are shown in Table S2 of Supporting Information. First of all, various gas-phase configurations of aziridine aldehyde (R) were fully optimized devoid of any solvents. The structural parameters of its most stable conformation R and two isomers R1 and R2 are depicted in Fig. 1. The isomer R is sterically the most stable one. In R1, although C@O and the neighboring CAH bond are coplanar, its relative energy is only higher than R by 0.4 kJ/mol. I deem that there is hydrogenbond-like n ? r⁄ bonding between the C@O and CAH bonds, viz. the lone pair electrons on O atom filling into the anti-bonding r⁄ of CAH bond because its trans-isomer R2 is much less stable by R (11.2 kJ/mol). The frontier orbitals of R and R1 (Fig. S1 of Supporting Information) show that their HOMOs are mainly the nonbonding orbitals involved n electrons, and the LUMOs are the anti-bonding p orbitals. In R1, the n electrons almost delocalize to the whole molecule. The electrostatic potential (ESP) surfaces and Mulliken charges (Fig. S2) testify that the bridged CAH hydrogen is more acidic than the other CAH hydrogen atoms, and all hydrogen atoms are slightly acidic. The dimerization mechanism of aziridine aldehyde free of any solvents was also illustrated in Fig. 1. Two aziridine aldehyde molecules are linked together via a weak O  HAC ‘‘hydrogen bond” (2.475 Å) to form IM1 due to the acidity of the hydrogen atom, and an n ? p⁄ interaction in N  C@O, decreasing the relative energy (DE) by 36.9 kJ/mol. In IM1, the frontier orbitals of

E (kJ/mol) 150.0

1.447

1.317

1.492

1.205

50.0

1.485

From the above-mentioned calculations it can be seen that there are various weak interactions and multiple reactive sites between two amphoteric aziridine aldehyde molecules, so the

1.616 1.159 1.141

1.456

1.500

1.437

1.357 1.667 1.303

1.480

TS1 125.6

R2 (11.2) 1.463

3.2. Reaction mechanisms of the dimer with CH2=CHCHO

1.451

1.495

100.0

HOMO-5 and HOMO-2 (Fig. 2) demonstrate these two weak interactions. Then NAH and C@O couple via TS1 with a four-membered ring to produce IM2 with a barrier of 162.5 kJ/mol. In this process, N links to C to form a new CAN bond, and H on N shifts to the C@O oxygen atom. From IM1 to IM2 via TS1, the CAN bond distance is shortened from 2.868 to 1.479 Å via 1.616 Å. In IM2, there is a seven-membered ring linked by hydrogen bonding (Fig. S3) and an apparent negative charge center (Fig. S4). IM2 is more stable than IM1 by 25.2 kJ/mol. IM2 will be further cyclized to the polycyclic dimer via TS2, where hydroxyl oxygen is bonded to the carbonyl carbon, and synchronously hydroxyl hydrogen transfers to the carbonyl oxygen. The barrier of this process is 157.8 kJ/mol, and the relative energy of dimer is further reduced by 41.1 kJ/mol compared with IM2. The pathway of IM1 ? TS1 is the ratedetermining step. To assess the validity of this theoretical model, IM1, TS1, IM2 and TS2 were re-optimized at B3LYP//MP2/6-31G (d, p) level in gas phase, and the barriers via TS1 and TS2 were estimated to be 168.5, 164.1 kJ/mol (B3LYP) and 170.0, 163.1 kJ/mol, which are in good agreement with the M06-2X ones. There is another reaction channel that the aziridine ring shifts from the C@O carbon to N and the hydroxyl hydrogen transfers concertedly to another O atom to lead to the rupture of the CAN bond and the ring opening of another aziridine group, producing an enol IM3 with only one three-membered ring, as illustrated in Fig. S5 of Supporting Information. Although IM3 is much more stable than the dimer by 66.1 kJ/mol, the energy barrier is as high as 242.5 kJ/mol, so this pathway is energetically unfavorable. The enol can be translated to its keto forms with slight energy decreases.

1.456

TS2 95.7

1.203

R1 (0.4)

Reaction coordinates

2R 0.0

0.0 1.464

-50.0

1.488

1.205

R (0.0)

-36.9 IM1

-62.1 IM2

2.475 1.211 2.868

-100.0 1.448

1.023

1.404 1.435

1.395 1.506 1.475

0.970

1.505 1.880 1.214

dimer -103.2

1.403

1.463 1.460

1.485

Fig. 1. Structural parameters and relative energies of R, R1 and R2 as well as the dimerization mechanism of R. The energies specified are the relative energies in kJ/mol (differences between the electronic and zero-point energies), and bond lengths are in Å.

107

X. Cheng / Computational and Theoretical Chemistry 1113 (2017) 105–109

HOMO

HOMO-2

HOMO-5

ESP

Fig. 2. Frontier orbitals and ESP surface of IM1.

reaction mechanisms between the polycyclic dimer and vinyl aldehyde CH2@CHCHO with multifunctional groups will be very complex but intriguing. Hili and Yudin [1] reported that the cyclization reactions of dimer and CH2@CHCHO occur in acetonitrile. Aziridine aldehyde (R) and its dimer were fully reoptimized, and their cyclization mechanisms will be discussed in acetonitrile, as illustrated in Figs. 3and 4. From Figs. 1, 3 and 4 it can be seen that the solvent can slightly influence their conformers. Firstly, CH2@CHCHO is connected to the dimer via a hydrogen bond and an n ? p⁄ interaction between N and the terminal C@C carbon, forming IM6. In IM6, the hydrogen atoms in NAH and OAH are ionized (Fig. S6), and NAH bond can couple to C@C bond via TS4 with a barrier of 107.8 kJ/mol, leading to IM7. From IM6 to IM7, the Mulliken charge of this N atom is enhanced from 0.548 to 0.451 e. The ionic OAH hydrogen can shift to the bridged O atom via TS5 to form IM8 with an energy barrier of 191.6 kJ/mol. In TS5, the bond distances between the shifting hydrogen and the two O atoms are 1.380 and 1.144 Å, and the breaking CAO bond

is extended to 1.618 Å. IM8 can resonate to its enol structure IM9, increasing the relative energy slightly by 14.3 kJ/mol. The relative energy of TS5 is 106.7 kJ/mol. This theoretical model shows that there is an alternative reaction channel from the enol form of IM7 to IM9. Although the enol IM10 is less stable than IM7 by 35.3 kJ/mol in energy, the barrier of the subsequent Htransfer process via TS6 still reaches 169.4 kJ/mol. Because the energy barrier via TS6 is lower than that via TS5 (191.6 kJ/mol), the process via TS6 is energetically more unfavorable, and the reaction channel of IM10 ? TS6 is its rate-determining step. IM10 and TS6 were re-optimized with the B3LYP and MP2 methods at same basis set level in acetonitrile, and energy barrier estimated by B3LYP and MP2 is 152.6 and 163.3 kJ/mol. The M06-2X barrier agrees well with the MP2 one. The hydroxyl hydrogen of IM9 can transfer to the adjacent N atom via TS7, leading to the CAN cleavage and the formation of and IM11. In IM11, the shifted hydrogen atom still connects with the O atom via a hydrogen bond with a bond length of 2.294 Å. If this bonding ruptures to IM12, the relative energy will increase

E (kJ/mol) 200.0 1.212

150.0

1.473

1.293

100.0

+

1.410 1.434

1.505

50.0

1.401

1.140

1.507

1.468

1.512

1.463

1.387

1.316

1.144 1.618

1.518

1.418

1.334

1.380

1.506

1.736

1.469

1.463

1.468

1.521

TS6 119.8

TS5 106.7

TS4 82.5

1.888

1.467

1.339

1.463

0.0

1.462

Reaction coordinates

-25.3 IM6

1.846

-84.9 IM7

1.017

1.505

1.479

2.025

1.504

1.334 1.508

1.399

1.467

-41.0 IM8

-49.6 IM10

1.335 3.299

-150.0

1.502

0.0

1.393

-100.0

1.219

1.483

dimer+CH2=CHCHO

1.476

-50.0

1.615 1.514

2.003 0.971

1.397

1.493

1.459

1.479 1.465

1.506

IM9 -26.7

1.482

1.464 Fig. 3. Structural parameters and potential energy surface (PES) of dimer + CH2 = CHCHO ? IM9. The relative energies (DE) in kJ/mol and bond lengths are in Å.

108

X. Cheng / Computational and Theoretical Chemistry 1113 (2017) 105–109

E (kJ/mol) 1.337

1.965

1.466 1.482

1.480

1.210

200.0 1.516

150.0

1.567

+ 1.164

1.338

1.440

TS7 142.8

1.882

1.478

1.478

100.0

50.0

1.179 1.282

0.0 IM9 -26.7 -50.0

1.491

58.3 TS8

1.513

1.356 1.336

1.501

1.020

1.489

IM13+R 58.3 1.200

IM12 23.3 3.7 IM11

2.159

1.459

-48.2 2.060 IM14

1.515 1.513

-100.0

1.577

Reaction coordinates 1.480

1.821

1.479

1.413

2.294

1.019

1.452 1.470

Fig. 4. Structural parameters and PES involved in the reaction mechanisms of IM9. The relative energies (DE) in kJ/mol and bond lengths are in Å.

by 19.6 kJ/mol. Then in the 5-(enolexo)-trig pathway, IM11 decomposes into the monomer aziridine aldehyde and IM13. However, the energy barrier of this process is too high (169.5 kJ/mol), and the subsequent cyclization of IM13 also do not reduce the relative energy evidently along the reaction coordinates, as illustrated in Fig. S7 of Supporting Information. The experimental investigation of Hili and A Yudin [1] also proved that there is no evidence for the 5-(enolexo)-trig pathway in trifluoroethanol. IM9 can also be cyclized to the 8-exo-trig aldol IM14 with an eight-membered ring via TS8. The energy barrier (85.0 kJ/mol) is much lower than that via TS7. The molecular orbitals (Fig. S8) show that, the CAO and C@O bonds are coupled together via the hydrogen bonding, forming a hexatomic ring. 4. Conclusion The dimerization mechanism of aziridine aldehyde and the addition and cyclization mechanisms of its dimer with CH2@CHCHO were investigated systematically in the present work. Two aziridine aldehyde molecules undergo an addition reaction (H-shift/CAN formation) and a cyclization process (H-shift/CAO formation) to form the stable dimer. There are prolific weak interactions in the cyclization system resulting from the amphoteric character of aziridine aldehyde, and the energy barriers of these two processes are 162.5 and 157.8 kJ/mol. In the reaction systems of the dimer of aziridine aldehyde with CH2@CHCHO, the N atom of the dimer attacks the terminal C@C carbon atom and the aziridine hydrogen transfers to another C

atom of C@C bond to form enol and keto isomers. The enol isomers can be decomposed via 5-(enolexo)-trig pathway, or be cyclized to 8-exo-trig aldol. However, the energy barrier of this 5-(enolexo)trig pathway is as high as 169.5 kJ/mol, so it is energetically unfavorable. Aziridine aldehyde and CH2@CHCHO are inclined to cyclize to polycyclic compounds. Conflict of interest The author declares no competing interest. Acknowledgements This work was supported by the National Natural Science Foundation of China (21502136 and 21571137). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.comptc.2017.05. 013. References [1] R. Hili, A.K. Yudin, J. Am. Chem. Soc. 131 (2009) 16404. [2] W. Zhao, Z. Wang, J. Sun, Angew. Chem. Int. Ed. 51 (2012) 6209. [3] D.B. Diaz, C.C.G. Scully, S.K. Liew, S. Adachi, P. Trinchera, J.D. St Denis, A.K. Yudin, Angew. Chem. Int. Ed. 55 (2016) 12659. [4] N.A. Afagh, A.K. Yudin, Angew. Chem. Int. Ed. 49 (2010) 262.

X. Cheng / Computational and Theoretical Chemistry 1113 (2017) 105–109 [5] C.-F. Xu, B.-H. Zheng, J.-J. Suo, C.-H. Ding, X.-L. Hou, Angew. Chem. Int. Ed. 54 (2015) 1604. [6] S. Celik, M. Cakici, H. Kilic, E. Sahin, Tetrahedron Asymmetry 26 (2015) 152. [7] L. Qi, Y. Chen, Angew. Chem. Int. Ed. 55 (2016) 13312. [8] X. Wang, W. Zhao, G. Li, G. Liu, J. Wang, M. Wang, L. Liu, Tetrahedron Asymmetry 26 (2015) 815. [9] M.A. Kuznetsov, L.M. Kuznetsova, A.S. Pankova, Tetrahedron Lett. 57 (2016) 3575. [10] Z. Wujkowska, S. Jarzyn´ski, A.M. Pieczonka, S. Les´niak, M. Rachwalski, Tetrahedron Asymmetry 27 (2016) 1238. [11] T.J. Senter, M.C. O’Reilly, K.M. Chong, G.A. Sulikowski, C.W. Lindsley, Tetrahedron Lett. 56 (2015) 1276. [12] C.S. Guan, T.R. Cundari, Comput. Theor. Chem. 1091 (2016) 64. [13] Á. Roxin, J. Chen, C.C.G. Scully, B.H. Rotstein, A.K. Yudin, G. Zheng, Bioconjugate Chem. 23 (2012) 1387. [14] S.K. Liew, Z. He, J.D. St Denis, A.K. Yudin, J. Org. Chem. 78 (2013) 11637. [15] S. Xing, J. Ren, K. Wang, H. Cui, W. Li, H. Yan, Tetrahedron 71 (2015) 6290. [16] C. Liu, P. Han, Y. Liu, M. Tang, Theor. Chem. Acc. 135 (2016) 59. [17] N.B. Heine, S.J. Kaldas, L. Belding, O. Shmatova, T. Dudding, V.G. Nenajdenko, A. Studer, A.K. Yudin, J. Org. Chem. 81 (2016) 5209. [18] N. Assem, R. Hili, Z. He, T. Kasahara, B.L. Inman, S. Decker, A.K. Yudin, J. Org. Chem. 77 (2012) 5613. [19] L. Belding, S. Zaretsky, B.H. Rotstein, A.K. Yudin, T. Dudding, J. Org. Chem. 79 (2014) 9465. [20] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,

[21] [22] [23] [24] [25] [26] [27] [28] [29]

109

X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, 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, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth, P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman, J.V. Ortiz, J. Cioslowski, D.J. Fox, GAUSSIAN 09, Revision C.01, Gaussian Inc., Wallingford CT, 2010. Y. Zhao, D. Truhlar, Theor. Chem. Acc. 120 (2008) 215. S. Schenker, C. Schneider, S.B. Tsogoeva, T. Clark, J. Chem. Theory Comput. 7 (2011) 3586. Y. Pan, C.W. Kee, Z. Jiang, T. Ma, Y. Zhao, Y. Yang, H. Xue, C.-H. Tan, Chem. Eur. J. 17 (2011) 8363. K.L. Jensen, C.F. Weise, G. Dickmeiss, F. Morana, R.L. Davis, K.A. Jørgensen, Chem. Eur. J. 18 (2012) 11913. D.K. Papayannis, A.M. Kosmas, Chem. Phys. 479 (2016) 53. J. Ghosh, A. Bhattacharya, Chem. Phys. 464 (2016) 26. J. Tomasi, R. Cammi, B. Mennucci, C. Cappelli, S. Corni, Phys. Chem. Chem. Phys. 4 (2002) 5697. J. Tomasi, B. Mennucci, R. Cammi, Chem. Rev. 105 (2005) 2999. A. Ishikawa, Y. Nakao, H. Sato, S. Sakaki, Dalton Trans. 39 (2010) 3279.