Density Functional Study of Nickel N-Heterocyclic ... - ACS Publications

Subscriber access provided by LAURENTIAN UNIV

Article

Density Functional Study of Nickel N-Heterocyclic Carbene Catalyzed C-O Bond Hydrogenolysis of Methyl Phenyl Ether: The Concerted #-H Transfer Mechanism Taveechai Wititsuwannakul, Yuthana Tantirungrotechai, and Panida Surawatanawong ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b02058 • Publication Date (Web): 19 Jan 2016 Downloaded from http://pubs.acs.org on January 26, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Density Functional Study of Nickel N-Heterocyclic Carbene Catalyzed C-O Bond Hydrogenolysis of Methyl Phenyl Ether: The Concerted β-H Transfer Mechanism Taveechai Wititsuwannakul,a Yuthana Tantirungrotechai,b and Panida Surawatanawong*a,c a

Department of Chemistry and Center of Excellence for Innovation in Chemistry,

Faculty of Science, Mahidol University, Bangkok 10400, Thailand. b

Department of Chemistry, Faculty of Science and Technology, Thammasat University,

Pathum Thani 12120, Thailand c

Center of Sustainable Energy and Green Material, Mahidol University, Salaya, Nakhon Pathom

73170, Thailand

ABSTRACT The catalytic C-O bond activation of aryl ethers attracts substantial interest as it is significant for the lignin degradation process. A nickel complex with N-heterocyclic carbene (Ni-SIPr) has been shown to selectively catalyze C-O bond hydrogenolysis of aryl methyl ether to obtain arene and alcohol as the only products. Here, the reaction mechanism of Ni-SIPr

ACS Paragon Plus Environment

1

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

catalyzed C-O bond hydrogenolysis of methyl phenyl ether (PhOMe) was studied using density functional theory. In the presence of H2, the catalytic cycle involves: (i) aromatic C-O bond oxidative addition of Ni(SIPr)(ƞ2-PhOMe) to form Ni(SIPr)(OMe)(Ph), (ii) β-H transfer from the methoxy to phenyl group in Ni(SIPr)(OMe)(Ph) via σ-complex-assisted metathesis (σ-CAM), which eliminates benzene and forms Ni(SIPr)(ƞ2-CH2O), (iii) H2 binding to form Ni(SIPr)(H2)(ƞ2-CH2O) prior to H-transfer from H2 to the formaldehyde carbon via σ-CAM to generate Ni(SIPr)(H)(OMe), and (iv) reductive elimination to form methanol and the binding of methyl phenyl ether to regenerate Ni(SIPr)(ƞ2-PhOMe). The tert-butoxide base could play a role to assist with the formation of Ni(SIPr)(ƞ2-PhOMe), the catalytically active species, and could bind to Ni(SIPr)(H)(OMe) before reductive elimination. A similar mechanism was found for the C-O bond hydrogenolysis of 2-methoxynaphthalene. Our study showed that the C-O bond oxidative addition is the rate-determining step and that the aromatic C-O bond cleavage to form Ni(SIPr)(OMe)(Ph) is more favorable than the aliphatic C-O bond cleavage to form Ni(SIPr)(OPh)(Me), consistent with the arene and alcohol products obtained from the experiment. Notably, the β-H transfer from the methoxy to phenyl group on Ni-SIPr is not a stepwise β-H elimination as generally perceived, but rather a concerted process that occurs via σ-CAM. This leads to benzene elimination before H2 binding, in accordance with the results of the isotope labeling experiment of C-O bond hydrogenolysis of 2-methoxynaphthalene. In the absence of H2, Ni(SIPr)(ƞ2-CH2O) tends to undergo C-H bond activation and α-H elimination to release H2 and generate nickel carbonyl complex, the catalytically inactive species. This was reflected by experimental results which demonstrated low conversion of 2-methoxynaphthalene in the absence of H2. Thus, H2 is crucial to the catalytic reaction through its role in suppressing the formation of the inactive nickel carbonyl species. Insights into the mechanisms of Ni-SIPr catalyzed conversion of methyl phenyl ether should benefit the development of catalysts for C-O bond activation. KEYWORDS: C-O bond activation, β-H transfer, nickel, N-heterocyclic carbene, density functional theory, aryl ether.


2

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

INTRODUCTION Catalysts for aromatic C-O bond activation have potential applications in the conversion of lignin to hydrocarbon feedstocks1,2 and in the conversion of aryl ethers into precursors for organic synthesis.3 Nickel catalysts have been developed for aromatic C-O bond cleavages of aryl ethers in coupled with carbon nucleophiles, e.g., Grignard reagent,4 organozinc,5 arylboronic acid.6 As hydrogen is less reactive than carbon nucleophiles, using hydrogen as a substrate for CO bond activation is more challenging. Heterogeneous catalysts are generally reactive for C-O bond hydrogenolysis of aryl ethers; however, aromatic rings often undergo further hydrogenation leading to the production of a mixture of cyclohexane and cyclohexanol, in addition to arene and aromatic alcohol products.7,8 Although homogeneous catalysts for aromatic C-O bond activation are less reactive, they are more selective than heterogeneous catalysts.1 In 2011, Sergeev and Hartwig reported CO bond hydrogenolysis of aryl ethers in the presence of tert-butoxide base under H2 using a Ni(COD)2 precursor and SIPr.HCl (Scheme 1).1 The C-O bond hydrogenolysis of diphenyl ether resulted in benzene and phenol as the only products. In the presence of both diphenyl ether and 4-tert-butylanisole, the primary observed products are derived from aromatic C-O bond cleavage (Scheme 1). These results implied that the Ni-SIPr catalytic system is more selective to aromatic C-O bond cleavage than to aliphatic C-O bond cleavage, and that the C-O bond hydrogenolysis of diphenyl ether is more favorable than that of methyl phenyl ether.1,

3d, 9

Thus to gain a better

understanding of the selectivity of the Ni-SIPr catalytic system, the reaction mechanisms for C-O bond hydrogenolysis of methyl phenyl ether should be investigated.

Scheme 1. C-O bond hydrogenolysis competition reaction of diphenyl ether and 4-tertbutylanisole.


3

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 28

Density functional theory (DFT) calculations have been successfully used to study the reaction mechanisms of nickel catalyzed cross-coupling of aryl esters;10 nickel with monodentate phosphine selectively activates C(aryl)-O over C(acyl)-O bond via a five-centered transition state, while nickel with bidentate phosphine selectively activates C(acyl)-O bond via a threecentered transition state. DFT study also provided insights into the mechanism of nickel Nheterocyclic carbene catalyzed C-O bond hydrogenolysis of diphenyl ether;11 Ni(SIPr)(ƞ2PhOPh) was found to be the active catalytic species for the reaction, and the energy barrier for the hydrogenation of the benzene products was found to be prohibitively high, in agreement with experimental results in which hydrogenation products were not observed.1 While the C-O bond hydrogenolysis of diphenyl ether involves the binding of H2 to Ni(SIPr)(ƞ2-PhOPh) to form Ni(SIPr)(OPh)(Ph)(H2) before H-transfer occurs via σ-complexassisted metathesis (σ-CAM)12 to finally generate benzene and phenol as the products,11 C-O bond hydrogenolysis of methyl phenyl ether possibly involves β-H elimination.13 In 2013, Agapie and coworkers performed C-O bond hydrogenolysis of 2-methoxynaphthalene, deuterated at the methoxy group (NaphOCD3), using a Ni(COD)2 precursor and SIPr.HCl in the presence of tert-butoxide base under H2 (Scheme 2).13 They obtained deuterated naphthalene (NaphD) as the main product (>90%), suggesting that the process proceeds via β-H elimination from the methoxy group.

Scheme 2. Isotopic labeling studies for C-O bond hydrogenolysis of deuterated 2methoxynaphthalene (NaphOCD3). Here, we used density functional theory to elucidate the reaction mechanism of Ni-SIPr catalyzed C-O bond hydrogenolysis of methyl phenyl ether (PhOMe) (Scheme 3) to gain insight into the reactivity and selectivity of the catalyst. We showed that the Ni(COD)2 precursor can proceed to ligand substitution with SIPr ligand and PhOMe substrate to generate Ni(SIPr)(ƞ2PhOMe), in a manner similar to the formation of Ni(SIPr)(ƞ2-PhOPh). The oxidative addition was found to preferably occur at the aromatic C-O bond. For methyl phenyl ether, H2 binding on


4

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Ni(SIPr)(OMe)(Ph) to form Ni(SIPr)(OMe)(Ph)(H2) before H-transfer to eliminate benzene is less likely. Instead, the Ni(SIPr)(OMe)(Ph) undergoes β-H transfer from the methoxy to phenyl group to generate benzene and Ni(SIPr)(ƞ2-CH2O). Notably, we found this process is not simply a stepwise β-H elimination, but rather a concerted σ-complex-assisted metathesis (σ-CAM) process. H2 then binds to Ni(SIPr)(ƞ2-CH2O) and the reaction proceeds to H-transfer leading to the reductive elimination of methanol. Understanding the mechanisms and the selectivity of the Ni-SIPr catalyst toward the aromatic C-O bond cleavage of methyl phenyl ether will greatly improve its use in lignin degradation and provide a means for using an inert methoxy leaving group for organic synthesis.3b-d

Scheme 3. Reaction mechanism for C-O bond hydrogenolysis of methyl phenyl ether catalyzed by nickel complex with N-heterocyclic carbene ligand (L=SIPr). The catalytic cycle involves: (i) oxidative addition, (ii) concerted β-H transfer, (iii) H2 addition, and (iv) reductive elimination.


5

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 28

COMPUTATIONAL DETAILS All calculations were performed with the Gaussian 09 program.14 Geometry optimizations and frequency calculations used B3LYP15 functional with basis set 1 (BS1). For BS1, the Stuttgart relativistic small core (RSC) 1997 ECP basis set16 was used for the Ni atom; 6-31++G(d,p)17 was used for the C, N, O, H atoms on the imidazoline ring of the N-heterocyclic carbene ligand (SIPr) and on the substrates, i.e., methyl phenyl ether, H2, and tert-butoxide base; 6-31G17 was used for all other C, N, O, and H atoms. Zero-point energies and thermodynamic functions were calculated at 393.15 K and 1 atm. M0618 functional has been successfully used to study reactions of related Ni catalysts.11, 19 Thus, M06 functional with basis set 2 (BS2) was used for a single point energy calculation on the gas-phase optimized structures and for the solvation energy correction. BS2 is the same as BS1, except that 6-31G(d)17 is used for the substituents on the imidazoline ring of SIPr. The solvation correction was performed by using the conductor-like polarizable continuum model (CPCM)20 with Bondi atomic radii and solvation parameters corresponding to m-xylene (ε = 2.348). The standard conditions were corrected to 1 M.11,

21

Unless specified otherwise, all energies mentioned throughout the article refer to the relative free energies with solvent correction in m-xylene calculated by M06/BS2//B3LYP/BS1.

RESULTS AND DISCUSSION We performed density functional calculations to investigate the mechanisms of C-O bond hydrogenolysis catalyzed by a nickel complex with N-heterocyclic carbene (SIPr) ligand in mxylene. Methyl phenyl ether was used as a model substrate to obtain insight into its reactivity towards aromatic C-O bond activation in comparison to diphenyl ether, and to investigate its CO bond hydrogenolysis mechanisms in the presence and absence of H2. The role of –OtBu base was elucidated. Then, the C-O bond hydrogenolysis of 2-methoxynaphthalene was also investigated. I.

Formation of Ni(SIPr)(ƞ2-PhOMe) and Ni(SIPr)(ƞ2-PhOPh) Similar to our previous study on diphenyl ether, Ni(COD)2 precursor can undergo ligand

substitution with SIPr to generate Ni(SIPr)(COD) (Figure 1). Although the second ligand substitution of Ni(SIPr)(COD) with SIPr to form Ni(SIPr)2 is more favorable by -8.1 kcal/mol,11 methyl phenyl ether is unlikely to bind onto Ni(SIPr)2 due to the steric hindrance of the SIPr


6

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

ligand. The formation and the accumulation of Ni(SIPr)2, which is not the key species, could impede the reaction, consistent with the findings that high temperature and a large amount of Ni(COD)2 precursor and SIPr ligand are required for the reaction.1 As the crystal structures of Ni0-arene complexes with ƞ6 and ƞ2–interactions are known,13, 22

we previously showed that in the presence of diphenyl ether, Ni(SIPr)(COD) can undergo

substitution reaction with diphenyl ether to form Ni(SIPr)(ƞ6-PhOPh), which easily rearranges to Ni(SIPr)(ƞ2-PhOPh) (IN1Ph) before proceeding to C-O bond oxidative addition.11 Thus, in the presence of methyl phenyl ether substrate, the formation of Ni(SIPr)(ƞ2-PhOMe) (IN1) is also expected. Note that Ni(SIPr)(ƞ6-m-xylene) could be formed for the reaction in m-xylene solvent; nevertheless, the ligand substitution with aryl ether substrates is energetically feasible to form Ni(SIPr)(ƞ2-PhOPh) and Ni(SIPr)(ƞ2-PhOMe) (Figure 1). While Ni(SIPr)(ƞ6-m-xylene) is catalytically inactive, Ni(SIPr)(ƞ2-PhOPh) and Ni(SIPr)(ƞ2-PhOMe), once formed, can readily proceed to C-O bond oxidative addition.

Figure 1. The ligand substitution of Ni(COD)2 precursor. Solvent corrected relative free energies in m-xylene are given in kcal/mol.


7

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 28

The formation of IN1 is less favorable than that of IN1Ph by +2.0 kcal/mol (Figure 1). The electronic structures of IN1 and IN1Ph revealed that, for both complexes, all five d-based orbitals are occupied (Figure S1). The dxy-based orbital represents the π-interaction with the phenyl group. Since the methoxy group is a stronger electron donating group than the phenoxy group, the back-bonding interaction from the Ni-SIPr to PhOMe is less favorable than that to PhOPh. Overall, the occupied MOs of IN1Ph are more stabilized than the corresponding MOs of IN1. The Ni-C(phenyl) bonds are also slightly stronger in IN1Ph (Figure S2). That is, in the presence of both diphenyl ether and methyl phenyl ether as substrates in Ni-SIPr catalytic system, IN1Ph should be more accessible than IN1.

II.

The C-O Bond Oxidative Addition of Ni(SIPr)(ƞ2-PhOMe) and Ni(SIPr)(ƞ2PhOPh) The aromatic C-O bond oxidative addition of Ni(SIPr)(ƞ2-PhOMe) (IN1) leads to three-

coordinate complex, Ni(SIPr)(OMe)(Ph) (IN2) via transition state TS1, similarly for the oxidative addition of Ni(SIPr)(ƞ2-PhOPh) (IN1Ph) to form IN2Ph, via TS1Ph. Although the free energy barriers for both complexes are comparable (~25 kcal/mol) (Figure 2), the formation of IN2Ph is thermodynamically more favorable than the formation of IN2 (-8.0 kcal/mol). In addition, IN1Ph is more accessible than IN1 (Figure 1). This corresponds to the competition reaction of diphenyl ether and 4-tert-butylphenyl methyl ether, which led to benzene and phenol as major products (94%) and to tert-butylbenzene as a minor product (2%).1


8

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Figure 2. The relative free energy profiles for the C-O bond oxidative addition of IN1 and IN1Ph. Solvent corrected relative free energies in m-xylene are given in kcal/mol. The aliphatic C-O bond oxidative addition of IN1 leads to a three-coordinate complex, Ni(SIPr)(OPh)(Me) (IN3). The reaction proceeds via transition state TS2 with a free energy barrier of +30.3 kcal/mol (Figure 2). Although IN3 is more stable than IN2, it involves a relatively higher energy barrier. We used the distortion/interaction model10b,

23

to gain insight

into the selectivity of C-O bond cleavage (Figure 3). The transition states TS1 and TS2 were separated into Ni-SIPr catalyst fragment and PhOMe substrate fragment for single point energy calculations. The distortion energy of Ni-SIPr catalyst (∆Edist-cat) is the difference between the distorted structure and the optimized structure of Ni-SIPr, while the distortion energy of PhOMe substrate (∆Edist-sub) is the difference between the distorted structure and the optimized structure of PhOMe. The interaction energy (∆Eint) is obtained from the difference between the activation energy (∆Eact) and the total distortion energy (∆Edist-cat+∆Edist-sub).


9

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

Figure 3. The distortion/interaction model for the activation of C(aryl)-O bond (TS1) and C(alkyl)-O bond (TS2). The distortion energy of the PhOMe fragment in TS1 is larger than that in TS2 (+41.6 and +31.5 kcal/mol), corresponding to a stronger C(aryl)-O bond relative to the C(alkyl)-O bond.24 Nevertheless, the interaction energy in TS1 is significantly more negative than that in TS2 (-46.7 and -27.1 kcal/mol). This could be due to the fact that π-back donation from Ni-SIPr to the C(aryl)-O bond with the available π* orbital is more favorable than that to the C(alkyl)-O bond with the available σ* orbital. Overall, the aromatic C-O bond cleavage is preferred for NiSIPr catalyzed conversion of methyl phenyl ether. The oxidative addition product, Ni(SIPr)(OMe)(Ph), has three isomers, i.e., IN2, IN4, and IN5 (Figure 2). The most stable isomer is IN2 with phenyl, the high trans influence ligand, opposite to the vacant site. IN4, with SIPr ligand trans to a C-H agostic interaction,25 has a higher energy than IN2 by +4.5 kcal/mol. The Ni-HC(methoxy) distance in IN4 is 1.77 Å (Figure 4). IN5 has two high trans influence ligands (phenyl and SIPr) opposite to each other; accordingly, the Ni-C(SIPr) bond in IN5 (2.02 Å) is weaker than those found in IN2 (1.90 Å) and IN4 (1.86 Å) (Figure 4), and IN5 has a higher energy than IN2 by +13.8 kcal/mol (Figure 2).


10

Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

IN1

TS1

TS2

IN2

IN4

IN5

Figure 4. The optimized geometries of the oxidative addition transition states (TS1 and TS2) and intermediates (IN1, IN2, IN4, and IN5). Calculated bond distances are shown in Å. All H atoms are omitted for clarity except for those from the methoxy. Ni atoms are shown in green, C atoms in grey, O atoms in red, N atoms in blue, and H atoms in purple. The role of tert-butoxide base in the formation and the C-O bond oxidative addition of Ni(SIPr)(ƞ2-PhOMe) The tert-butoxide base (–OtBu) in the reaction could be used to generate SIPr from SIPr•HCl for nickel catalyst.1, 11 Nolan and coworkers reported that with the excess amount of NaOtBu, Cu(IPr)(OtBu) can be generated from [Cu(IPr)2]BF4.26 Here, we also investigated the formation of [Ni(SIPr)(OtBu)]-, the isoelectronic complex to Cu(IPr)(OtBu). In fact, the ligand substitution of Ni(COD)2 precursor to form [Ni(SIPr)(OtBu)]- is quite favorable (-6.9 kcal/mol),


11

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

which is comparable to the formation of Ni(SIPr)2 (-8.1 kcal/mol) (Figure 1). [Ni(SIPr)(OtBu)]can bind PhOMe to form [Ni(SIPr)(OtBu)(η2-PhOMe)]- (IN6) (+15.9 kcal/mol) (Figure 5). Then, dissociation of –OtBu base from IN6 leads to Ni(SIPr)(η2-PhOMe) (IN1), the active species for the C-O bond oxidative addition. On the other hand, Ni(SIPr)2 is less likely to bind PhOMe substrate due to the steric hindrance. The SIPr ligand dissociation from Ni(SIPr)2 before PhOMe binding to form IN1 is less favorable than the association of the PhOMe via IN6 (Figure 5). Once formed, the IN1 proceeds to aromatic C-O bond oxidative addition with the energy barrier of +25.6 kcal/mol (Figure 5). We also explored the IN6 undergoing aromatic C-O bond oxidative addition via TS3 to yield [Ni(SIPr)(OtBu)(Ph)(OMe)]- (IN7); however, the energy barrier is higher than that via TS1 (Figure 5). Although the tert-butoxide base may not directly assist with the C-O bond oxidative addition via TS3, the excess amount of tert-butoxide base is important to shift the equilibrium position from Ni(SIPr)2 to [Ni(SIPr)(OtBu)]-, which facilitates the formation of IN1 (Figure 5).

Figure 5. The relative free energy profiles for the role of tert-butoxide base in the formation of IN1 and to the C-O bond oxidative addition of IN1. Solvent corrected relative free energies in mxylene are given in kcal/mol.


12

Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

III.

The conversion of methyl phenyl ether in the presence of H2

Path A. β-H transfer via σ-complex-assisted metathesis/H2 addition Several studies revealed that late transition metal alkoxide complexes can undergo β-H elimination.27 Here, IN4, the Ni(SIPr)(OMe)(Ph) that has a C-H agostic interaction from the methoxy to the nickel in the cis position to the phenyl (Figure 4) could proceed to β-H elimination. Several attempts have been made to locate the product of the β-H elimination involving a β-H transfer from the methoxy group to the Ni to form Ni(SIPr)(Ph)(H)(η2-CH2O) with a hydride cis to the phenyl; however, they were not successful. Instead, we obtained a fivecentered transition state, TS4 (Figure 6). The Ni-O distance is slightly lengthened from IN4 by 0.04 Å, while the Ni-C(methoxy) distance is shortened by 0.20 Å (Figure 4 and Figure 6). The βH moves away from the C(methoxy) (1.69 Å) and comes close to the ipso-carbon of the adjacent phenyl group (1.54 Å). Ni-H interaction was also found (1.49 Å).

TS4

IN8

IN9

IN10

Figure 6. The optimized geometries of the β-H transfer transition state via σ-complex-assisted metathesis (TS4) and intermediates (IN8, IN9, and IN10). Calculated bond distances are shown in Å. All H atoms are omitted for clarity except for those from the methoxy. The structure of TS4 suggests that the β-H transfer from the methoxy to the phenyl group proceeds via σ-complex-assisted metathesis12 (σ-CAM). To the best of our knowledge, the σCAM of the transition metal alkoxide complex has not been proposed. Other σ-CAM studies involve the σ-complexes of alkane, silane, borane, and dihydrogen.12, 28 In this reaction, IN4 has a phenyl group in close proximity to the C-H agostic interaction from the methoxy group,


13

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

facilitating the σ-CAM process to eliminate benzene and to form Ni(SIPr)(η2-CH2O) (IN8) (-13.7 kcal/mol). The energy barrier relative to IN2 is +12.8 kcal/mol (Figure 7). Although the three-coordinate complex IN2 can easily rearrange to IN9 with a C-H agostic interaction from the methoxy to the nickel in the cis position to the SIPr (90%).1,13 In fact, the formation of Ni(SIPr)(η2-NaphOMe) is more favorable than the formation of Ni(SIPr)(η2-PhOMe) (Figure S10). Correspondingly, the C-O bond


19

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 28

activation of 2-methoxynaphthalene is easier than that of methyl phenyl ether as commonly found in the experiments.4a, 6, 31 Generally, β-H elimination in transition metal (M) complexes involves the H transfer from the β-carbon of alkyl, amide, or alkoxide ligand to form an M-H bond and a π bond (Scheme 4).27a, 27b, 32 According to our density functional study, the mechanism for the H transfer in Ni-SIPr catalyzed C-O bond cleavage of methyl phenyl ether is not simply a β-H elimination as previously proposed,13 but rather a σ-complex-assisted metathesis, in which the β-H transfer from the methoxy to the phenyl occurs through a concerted process with Ni-H interaction (Scheme 4).

Scheme 4. A stepwise β-H elimination and a concerted β-H transfer via σ-complex-assisted metathesis (σ-CAM). IV.

The conversion of methyl phenyl ether in the absence of H2 Although Ni-SIPr catalyzed conversion of 2-methoxynaphthalene involved β-H transfer

from the methoxy group to form the naphthalene product, the reaction proceeded with only

Density Functional Study of Nickel N-Heterocyclic ... - ACS Publications

Recommend Documents