Theoretical study of the mechanism of two successive N-methylene C–H bond activations on a phosphine-tethered N-heterocyclic carbene on a triruthenium carbonyl cluster

Abstract

Recently the Kennedy group reported a unique successive N-methylene C–H bond activation on N-heterocyclic triruthenium carbene complexes. Here, density functional theory calculations have been performed on this reaction to clarify the molecular-level mechanism of the two C–H bond activations. The calculated results indicated that the reaction occurs sequentially through the following steps: phosphine ligand dissociation from the Ru center followed by rearrangement of ligands on Ru center, the first C–H bond oxidative addition to Ru, the elimination of the first CO ligand with recoordination of the phosphine ligand, the second CO ligand elimination followed by the second C–H bond activation, and hydride migrations between Ru centers. The rate-determining step is the first C–H bond activation, which needs to overcome a barrier of 37.9 kcal mol⁻¹.

---

1. Introduction

N-heterocyclic carbenes (NHCs) have attracted considerable attention because of their significant potential as versatile ligands for homogeneous catalysis.^1,2 In recent decades, researchers have contributed a great deal to the coordination chemistry of NHCs. Numerous breakthroughs, made by many chemists including Grubbs,^3–5 Herrmann,^6,7 Nolan,^8,9 and Fürstner,^10,11 indicate the importance of NHCs as supporting ligands on various metal complexes. Metal–NHC complexes as superior homogeneous catalysts can be widely used in stoichiometric or catalytic reactions,¹² and also show extensive applications in some significant organic synthesis reactions, including hydrogenation reactions of alkenes,¹³ olefin metathesis reactions,^14,15 C–C cross-coupling reactions,^16–18 hydrosilylation reactions,^19,20 allylic substitution reactions,²¹ and hydroboration reactions.²²

Compared with some traditional ligands, such as phosphines,²³ NHCs, as strong σ-donor ligands, can form stronger coordination bonds with metal centers. Moreover, the fence or fanlike steric environment of NHCs differentiates them substantially from tertiary phosphines. This feature makes them usually act as supporting spectator ligands on metal centers.^7,24–26 However, recent studies found that the C–H,^27–32 C–C,^33,34 and C–N³⁵ bonds of substitutional groups bounded to the nitrogen atoms of metal–NHC complexes can be activated under specific conditions. These processes provide valuable information for designing efficient organometallic catalysts.^36–38

Recently, a series of work related to the N-alkyl C–H bond activation on triruthenium carbene complexes have been studied by the researcher group of Cabeza et al.^29,39–42 Scheme 1 shows an example of their work, where triruthenium carbonyl complex (1) is converted to its dihydrido derivative (2) through intramolecular successive C–H bond activations upon gentle warming (∼70 °C) in tetrahydrofuran (THF).

Scheme 1 Schematic diagram for the successive activations of two N-methylene C–H bonds on a NHC triruthenium complex reported by Cabeza et al.⁴⁰ Black points represent COs, and two methylene H atoms in 1 and hydrides in 2 are highlighted by colors for easy identification.

In order to better understand the mechanism of N-alkyl C–H bond activations, Cabeza et al.^43,44 and our group⁴⁵ have performed density functional theory (DFT) studies on several triruthenium carbonyl complexes, which provide fundamental information of the mechanistic and energetic aspects. However, the overall reaction profile and mechanism details of the transformation from 1 to 2 are still not clear, which spurs us on to perform a detailed theoretical study at the molecular level. In this work, our aim is to show a clear mechanistic picture of the transformation from 1 to 2 via two successive C–H bond activations and hence understand the experimentally observed transformation.

2. Computational details

In this study, all structures were first optimized in the gas phase and then reoptimized by considering the solvent effect of THF with the IEFPCM model using Truhlar's SMD model parameters.⁴⁶ The calculations were carried out by using the M06-2X functional⁴⁷ with the 6-31G(d,p) basis set⁴⁸ for C, H, O, N atoms and the effective core potential basis set (LANL2DZ)^49–51 for Ru and P atoms. The functional and basis set used in this work have been confirmed to predicate accurate relative energies.⁵² Frequency calculations were carried out by using the same method to characterize the stationary points as intermediates or transition states and to get Gibbs free energies. The intrinsic reaction coordinate (IRC) calculations were also carried out to examine that whether or not each transition state connects to the corresponding reactant and product. All reported relative energies are Gibbs free energies. GAUSSIAN 09 package is utilized to perform all DFT calculations.⁵³ The optimized structures and their corresponding Cartesian coordinates are given in the ESI.†

3. Results and discussion

Table 1 compares the calculated structural parameters of complex 1 (Scheme 1) with corresponding crystallographic data.⁴⁰ It is found that the theoretical structural parameters are compatible with the experimental values, indicating that the ruthenium complex can be appropriately described by the theoretical method used in this paper (M06-2X/6-31G(d,p)/LANL2DZ).⁵²

Table 1 Calculated and experimental structural parameters of complex 1^a

		Calc	Exp
a Bond lengths in angstroms.
	Ru1–Ru2	2.928	2.903
	Ru1–Ru3	2.887	2.850
	Ru2–Ru3	2.875	2.834
	Ru1–C2	2.101	2.112
	C2–N1	1.360	1.362
	C2–N2	1.357	1.368
	C1–N1	1.458	1.459
	C1–C3	1.521	1.512
	Ru2–P	2.370	2.338

---

For the conversion of reactant 1 to product 2 (Scheme 1), the elementary steps identified from the present calculations are as follows: phosphine ligand dissociation from the Ru center followed by rearrangement of ligands, the first C–H bond activation, the first CO ligand elimination with recoordination of the phosphine ligand, the second CO ligand followed by the second C–H bond activation, and hydride migrations between Ru centers. The following discussion shows the mechanism details of each elementary step.

As shown in Fig. 1, the first C–H bond activation starts from dissociation of phosphine ligand from Ru2 center through transition state TS1, leading to intermediate IM1 with a vacant site on Ru2. The barrier of this step in free energy is calculated to be 16.1 kcal mol⁻¹ (1 → TS1). Then intermediate IM1 isomerizes to a more stable intermediate IM2 through a single bond twist. Subsequently, rearrangement of ligands on Ru1 center occurs through transition state TS2, which makes the NHC fragment site in IM2 is migrating from the initial horizontal coordination mode to the vertical coordination mode. Such a ligand reorganization process is essential for the C–H bond activation. The rearrangement of ligands on Ru1 center in IM2 forms intermediate IM3, with an activation barrier of 18.3 kcal mol⁻¹. In order to make a hydrogen atom (H1) of N-methylene approach Ru2, intermediate IM3 converts to the suitable intermediate IM3-a through a clockwise single bond rotation around Ru1–C2 bond. In IM3-a, the Ru2–H1 distance is 2.689 Å, implying that there is a weak agnostic interactions between Ru2 and H1 atoms. The calculated NBO charges on Ru2 and H1 are −0.232 and 0.221 e respectively, indicating that the weak agnostic interaction is facilitated by the coulombic attraction. With the aid of the weak agostic interaction, the process for the oxidative addition of C1–H1 bond to Ru2 occurs via transition state TS3-a. The C1–H1 bond activation affords IM4 and the calculated free energy barrier from IM3-a to TS3-a is 25.2 kcal mol⁻¹. The optimized stationary structures involved in the first C–H bond activation are given in Fig. S1 in the ESI.†

Fig. 1 Calculated free energy profile (a) and the schematic diagrams for the first C–H bond activation process (b). Bond lengths in angstroms.

In addition, we also studied another possible pathway for the C1–H1 bond activation, and the calculated energy profile and mechanism details are given in Fig. S2 in the ESI.† In this pathway, the CO ligand elimination from IM3-a occurs through transition state TS3-b, affording intermediate IM4-b which has a vacant site for subsequent C1–H1 bond oxidative addition to Ru3. The calculated barrier for CO ligand elimination is 13.3 kcal mol⁻¹. The C1–H1 bond activation occurs via transition state TS4-b, which is located to lie above the reaction entrance by 52.2 kcal mol⁻¹, implying that this pathway is unaccessible under the experiment conditions. The optimized structures in this pathway are shown in Fig. S3 in the ESI.†

The second C–H bond activation occurs sequentially through the first CO ligand elimination with recoordination of the phosphine ligand, and the second CO ligand elimination followed by the C1–H2 bond oxidative addition. The calculated mechanism details and relative energy profile are given in Fig. 2 and the optimized structures are shown in Fig. S4 in the ESI.†

Fig. 2 Calculated free energy profile (a) and the schematic diagrams for the second C–H bond activation process (b). The red and blue points denote the CO that will be released or is being released from the metal center. Bond lengths in angstroms.

To make phosphine ligand approach Ru2 center, intermediate IM4 evolves to a more stable intermediate IM5 by rotation of the Ph₂PPh group along the C1–C3 single bond. The first CO ligand elimination from Ru2 was accompanied by the recoordination of the phosphine ligand. This process occurs via transition state TS5, leading to intermediate IM6. The transformation from IM5 to IM6 needs to overcome a free barrier of 15.9 kcal mol⁻¹. To accomplish the second C–H bond activation, the second CO ligand elimination from Ru3 center first occurs via transition state TS6, affording intermediate IM7, where Ru3–H2 distance is 2.042 Å, implying that a weak agostic interaction exists between Ru3 and H2 atoms. With the aid of the weak agostic interaction, the C1–H2 bond is activated through TS7, affording IM8 with a barrier of 16.1 kcal mol⁻¹.

The C1–H2 bond activation proceeds through TS7, resulting in intermediate IM8, where Ru3–H2 bond is formed. IM8 converts to the final dihydrido derivative 2 through two sequential hydride ligand migration steps. Fig. 3 shows the calculated energy profile for these two hydride migration steps, and the optimized structures are given in Fig. S5 in the ESI.†H2 first migrates to the bridging position of Ru1 and Ru3, as indicated by IM9, via TS8 with a free barrier of 7.0 kcal mol⁻¹. Subsequently, IM9 converts to the final product 2 through TS9 with a barrier of only 3.1 kcal mol⁻¹. This smaller barrier implies that hydride ligand migrations occur easily under thermal reaction condition.

Fig. 3 Calculated free energy profile (a) and the schematic diagrams for the process of the hydride ligand migration (b). Bond lengths in angstroms.

Alternatively, we located two alternative pathways for the second C–H bond activation. The calculated energy profile and optimized structure are given in Fig. S6 and S7 in the ESI† respectively. The common feature of these two pathways is that the initial CO elimination is from Ru3 through TS5-1 rather than from Ru2 through TS5 shown in Fig. 2. As seen in Fig. S6,† the relative free energy of TS5-1 is slightly lower (∼1.2 kcal mol⁻¹) than that of TS5. However, the relative energies of subsequent transition states, TS6-1 (corresponding to the second CO elimination) and TS6-2 (corresponding the C1–H2 oxidative addition to Ru3), along both the two alternative pathways are found to be energetically less favorable than the transition states shown in Fig. 2. Thus the reaction is not expected to occur along the two pathways shown in Fig. S6 in the ESI.†

From the above results, it is clear that the overall maximum on the free energy surface corresponds to TS3-a, where the C1–H1 oxidative addition to Ru2. The entire reaction is calculated to be endothermic by 8.6 kcal mol⁻¹ with an overall free energy barrier of 37.9 kcal mol⁻¹. To refine the calculated energies, we performed single-point energy calculations using the more extensive basis sets, i.e. the Def2TZVP basis set for Rh, and the 6-311+G(d,p) basis set for all other atoms. The overall barrier is found to be 37.1 kcal mol⁻¹, which is not remarkably different from the calculated result (37.9 kcal mol⁻¹) using the LANL2DZ-6-31G(d,p) basis set. This indicates that the reaction could intrinsically involve a barrier of ∼38 kcal mol⁻¹. Such a barrier seems to be somewhat higher than expected for the reaction under consideration which was carried out at 66 °C. However, it should been noted that as shown in Scheme 1, the overall process involves two CO-elimination steps that are irreversible because the reaction was performed in an open system, where the released CO was continuously purged with an inert gas.⁴⁰ Thus although the barrier of the reaction is relatively higher, the irreversible elimination of CO from the system drive the reaction proceed smoothly. In addition, it should be noted that some transition states only involving bond rotations (for example, from IM3 to IM3-a and IM4 to IM5) were not located in this work because they are expected to have low barriers and are not the rate-determining states of reaction.

4. Conclusions

In conclusion, the mechanism details of the N-methylene C–H bond activations on N-heterocyclic triruthenium carbonyl complexes have been theoretically studied by carrying out DFT calculations. The dihydrido derivative is formed through the following steps: phosphine ligand dissociation followed by rearrangement of ligands, the first C–H bond activation, the elimination of the first CO ligand with recoordination of the phosphine ligand, the second CO ligand followed by the second C–H bond activation, and hydride migrations between Ru centers. CO ligand elimination is an essential process for the metal center approaching the C–H bond and the first C–H bond activation is the rate-determining step. The entire reaction is calculated to be endothermic by 8.6 kcal mol⁻¹, and the C–H activation transition state TS3-a lies above the reaction entrance by 37.9 kcal mol⁻¹. Such a moderately high barrier is compatible with the experimentally observed easy transformation from the NHC carbene triruthenium carbonyl complex to its dihydrido derivative under thermal reaction conditions. The present theoretical results provide a clear picture of mechanistic and energetic aspects of unique C–H bond activation processes.

Acknowledgements

We acknowledge National Basic Research Program of China (973 Program, 2013CB934301) and National Natural Science Foundation of China (No. 21433006 and 21273131) for financial support for this work.

References

1. D. Cardin, J. B. Cetinkaya and M. F. Lappert, Chem. Rev., 1972, 72, 545–547.
2. T. M. Trnka and R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18–29.
3. M. S. Sanford, J. A. Love and R. H. Grubbs, J. Am. Chem. Soc., 2001, 123, 6543–6554.
4. T. M. Trnka, J. P. Morgan, M. S. Sanford, T. E. Wilhelm, M. Scholl, T. Choi, S. Ding, M. W. Day and R. H. Grubbs, J. Am. Chem. Soc., 2003, 125, 2546–2558.
5. J. A. Love, M. S. Sanford, M. W. Day and R. H. Grubbs, J. Am. Chem. Soc., 2003, 125, 10103–10109.
6. T. Weskamp, W. C. Schattenmann, M. Spiegler and W. A. Herrmann, Angew. Chem., Int. Ed., 1998, 37, 2490–2493.
7. W. A. Herrmann, Angew. Chem., Int. Ed., 2002, 41, 1290–1309.
8. R. Dorta, E. D. Stevens and S. P. Nolan, J. Am. Chem. Soc., 2004, 126, 5054–5055.
9. N. M. Scott, R. Dorta, E. D. Stevens, A. Correa, L. Cavallo and S. P. Nolan, J. Am. Chem. Soc., 2005, 127, 3516–3526.
10. A. Fürstner, O. R. Thiel and C. W. Lehmann, Organometallics, 2002, 21, 331–335.
11. S. Pruehs, C. W. Lehmann and A. Fürstner, Organometallics, 2004, 23, 280–287.
12. N. M. Scott and S. P. Nolan, Eur. J. Inorg. Chem., 2005, 2005, 1815–1828.
13. H. M. Lee, D. C. Smith Jr, Z. J. He, E. D. Stevens, C. S. Yi and S. P. Nolan, Organometallics, 2001, 20, 794–797.
14. S. J. Connon and S. Blechert, Top. Organomet. Chem., 2004, 11, 93–124.
15. H. Clavier, K. Grela, A. Kirschning, M. Mauduit and S. P. Nolan, Angew. Chem., Int. Ed., 2007, 46, 6786–6801.
16. F. Glorius, Top. Organomet. Chem., 2007, 21, 1–20.
17. O. Navarro, N. Marion, J. G. Mei and S. P. Nolan, Chem.–Eur. J., 2006, 12, 5142–5148.
18. M. G. Gardiner, W. A. Herrmann, C.-P. Reisinger and M. Spiegler, J. Organomet. Chem., 1999, 572, 239–247.
19. H. Kaur, F. K. Zinn, E. D. Stevens and S. P. Nolan, Organometallics, 2004, 23, 1157–1160.
20. I. E. Marko, S. Sebastien, O. Buisine, G. Mignani, P. Branlard, B. Tinant and J.-P. Declercq, Science, 2002, 298, 204–206.
21. Y. Sato, T. Yoshino and M. Mori, Org. Lett., 2003, 5, 31–33.
22. G. A. Grasa, Z. Moore, K. L. Martin, E. D. Stevens, S. P. Nolan, V. Paquet and H. Lebel, J. Organomet. Chem., 2002, 658, 126–131.
23. M. Melaimi, M. Soleilhavoup and G. Bertrand, Angew. Chem., Int. Ed., 2010, 49, 8810–8849.
24. W. A. Herrmann and C. Kocher, Angew. Chem., Int. Ed., 1997, 36, 2162–2187.
25. A. J. Arduengo, Acc. Chem. Res., 1999, 32, 913–921.
26. D. Enders and H. J. Gielen, J. Organomet. Chem., 2001, 617, 70–80.
27. R. Cariou, C. Fischmeister, L. Toupet and P. H. Dixneuf, Organometallics, 2006, 25, 2126–2128.
28. J. A. Cabeza, I. del Río, D. Miguel and M. G. Sánchez-Vega, Chem. Commun., 2005, 3956–3958.
29. J. A. Cabeza, I. del Rıo, D. Miguel, E. Perez-Carreno and M. G. Sanchez-Vega, Dalton Trans., 2008, 1937–1942.
30. J. A. Cabeza, I. del Río, D. Miguel and M. G. Sánchez-Vega, Angew. Chem., Int. Ed., 2008, 47, 1920–1922.
31. S. Burling, B. M. Paine, D. Nama, V. S. Brown, M. F. Mahon, T. J. Prior, P. S. Pregosin, M. K. Whittlesey and J. M. J. Williams, J. Am. Chem. Soc., 2007, 129, 1987–1995.
32. S. Burling, M. F. Mahon, B. M. Paine, M. K. Whittlesey and J. M. J. Williams, Organometallics, 2004, 23, 4537–4539.
33. R. A. Diggle, S. A. Macgregor and M. K. Whittlesey, Organometallics, 2008, 27, 617–625.
34. R. F. R. Jazzar, S. A. Macgregor, M. F. Mahon, S. P. Richards and M. K. Whittlesey, J. Am. Chem. Soc., 2002, 124, 4944–4945.
35. S. Burling, M. F. Mahon, R. E. Powell, M. K. Whittlesey and J. M. J. Williams, J. Am. Chem. Soc., 2006, 128, 13702–13703.
36. R. Corberán, M. Sanaú and E. Peris, J. Am. Chem. Soc., 2006, 128, 3974–3979.
37. M. Sholl, S. Ding, C. W. Lee and R. H. Grubbs, Org. Lett., 1999, 1, 953–956.
38. L. Ackermann, A. Furstner, T. Weskamp, F. J. Kohl and W. A. Herrmann, Tetrahedron Lett., 1999, 40, 4787–4790.
39. J. A. Cabeza, M. Damonte and E. Pérez-Carreño, Organometallics, 2012, 31, 8355–8359.
40. J. A. Cabeza, M. Damonte, P. García-Álvarez, M. G. Hernández-Cruz and A. R. Kennedy, Organometallics, 2012, 31, 327–334.
41. J. A. Cabeza, M. Damonte, P. García-Álvarez, A. R. Kennedy and E. Pérez-Carreño, Organometallics, 2011, 30, 826–833.
42. J. A. Cabeza, M. Damonte and P. García-Álvarez, Organometallics, 2011, 30, 2371–2376.
43. J. A. Cabeza and E. Pérez-Carreño, Organometallics, 2008, 27, 4697–4720.
44. J. A. Cabeza, M. Damonte and E. Pérez-Carreño, Organometallics, 2012, 31, 8114–8120.
45. J. Yan, Z. Han, D. J. Zhang and C. B. Liu, Theor. Chem. Acc., 2015, 134, 137.
46. A. V. Marenich, C. J. Cramer and D. G. Truhlar, J. Phys. Chem. B, 2008, 113, 6378–6396.
47. Y. Zhao and D. G. Truhlar, Acc. Chem. Res., 2008, 41, 157–167.
48. P. C. Hariharan and J. A. Pople, Theor. Chim. Acta, 1973, 28, 213–222.
49. P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 270–283.
50. P. J. Hay and W. R. Wadt, J. Chem. Phys., 1985, 82, 299–310.
51. W. R. Wadt and P. J. Hay, J. Chem. Phys., 1985, 82, 284–298.
52. S. C. Chmely, S. Kim, P. N. Ciesielski, G. Jimenéznez-Osé, R. S. Paton and G. T. Beckham, ACS Catal., 2013, 3, 963–974.
53. 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, 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, 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 and D. J. Fox, Gaussian 09, Revision B.01, Gaussian, Inc., Wallingford CT, 2010.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19787f

---