How Does the Catalyst Affect the Reaction Pathway? DFT Analysis of

Article Cite This: Organometallics XXXX, XXX, XXX−XXX

How Does the Catalyst Affect the Reaction Pathway? DFT Analysis of the Mechanism and Selectivity in the 1,6-Diyne Ester Cycloisomerization Yunhe Li,† Peng-Cheng Tu,† Lin Zhou,† Alexander M. Kirillov,‡ Ran Fang,*,† and Lizi Yang*,† †

State Key Laboratory of Applied Organic Chemistry and Key Laboratory of Nonferrous Metals Chemistry and Resources Utilization of Gansu Province, College of Chemistry and Chemical Engineering, Lanzhou University, Lanzhou 730000, People’s Republic of China ‡ Centro de Química Estrutural, Complexo I, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-001 Lisbon, Portugal S Supporting Information *

ABSTRACT: The present study reports a detailed theoretical analysis of the mechanistic and chemoselectivity features in 1,6diyne ester cycloisomerization. The energy profiles for three different catalysts, namely, [Au I (PPhMe 2 )(NCMe)] + , [AuIII(Cl)2(pic)] (pic = 2-picolinate), and PtCl2, were investigated. The DFT calculations reveal that all of these catalysts entail similar 1,3-acyloxy migration and 5-exo-dig cyclization steps, whereas completely distinct reaction pathways are observed after the formation of the putative vinyl metal complex intermediates. In the [AuI(PPhMe2)(NCMe)]+ system, the configuration of the phosphine ligand can explain the exclusive chemoselectivity of the Friedel−Crafts reaction over the 1,5-acyl migration. On the other hand, in the [AuIII(Cl)2(pic)] and PtCl2 systems, the 1,5-acyl migration is assisted by the chloride ligand, offering an alternative mechanism that can justify a reasonable activation barrier and the corresponding stereochemical feature in the reaction. Moreover, the [AuI(PPhMe2)(NCMe)]+ complex with soft and carbophilic character represents an electron-deficient catalyst with a linear structure; it is particularly unsuitable for the 1,5-acyl migration. In contrast, the [AuIII(Cl)2(pic)] catalyst reveals a distortedsquare-planar geometry that satisfies the condition to form a square-planar moiety with an acyl functionality. Thus, the obtained theoretical results not only well rationalize the experimental observations but provide insights into the details of the 1,5-acyl migration.

1. INTRODUCTION Transition-metal-catalyzed tandem cycloisomerization of unsaturated π-conjugated propargylic esters is an elegant synthetic tool for the assembly of complex polyclic scaffolds.1 This type of transformation can be driven by homogeneous gold catalysis, which has received significant consideration in synthetic organic chemistry, given the mild reaction conditions and atom economy.2 In particular, 1,n-enyne esters were employed to synthesize a variety of targets.3 Generally, there are two possible pathways (1,3 or 1,2-acyloxy migration) for the rearrangement of 1,n-enyne esters forming allenes or vinyl gold carbenoid intermediates;4 both of them can undergo further transformations.5 Chan and co-workers described an excellent synthetic method to prepare 2,4-dihydro-1H-fluorenes via a gold(I)catalyzed 1,2-acyloxy migration, cyclopropenation, and Nazarov cyclization of 1,6-diyne carbonates.6 Then, they presented a notable synthetic protocol to chemoselectively prepare 1Hcyclopenta[b]naphthalenes and cis-cyclopenten-2-yl δ-dike© XXXX American Chemical Society

tones with excellent yields, using the gold-catalyzed cycloisomerization of 1,6-diyne esters (Scheme 1).7 These studies showed that efficient and divergent chemoselectivity is possible by using the advantage of the electronic and steric differences in the respective gold(I) and gold(III) complex catalysts, which open up many possibilities for the development of new synthetic protocols. Meanwhile, Chen and co-workers also developed a novel PtCl2-catalyzed sequential pentannulation of propargylic diynyl esters that gives benzofulvene diketones (Scheme 1).8 Additionally, metal-catalyzed isomerization of propargylic esters to diketone derivatives is a fascinating example of C−C bond generation in gold organometallic chemistry, since it is one of the few reactions that exploit the nucleophilicity of organoaurates to a migrating acyl group. The selective transformation of various substrates using different homogeneous catalysts under individually optimized reaction Received: November 7, 2017

A

DOI: 10.1021/acs.organomet.7b00813 Organometallics XXXX, XXX, XXX−XXX

Article

Organometallics Scheme 1. Au(I)-, Au(III)-, and Pt(II)-Catalyzed 1,6-Diyne Ester Cycloisomerization

Scheme 2. Proposed Mechanism for the Au(I)-, Au(III)-, and Pt(II)-Catalyzed Cycloisomerization of the 1,6-Diyne Ester 1

conditions to give structurally distinct products is a powerful approach to generate diverse molecular scaffolds. For example, Chan et al. achieved a divergence in product selectivity by finetuning the steric nature of ligands in gold(I) catalysts.9 Fensterbank et al. also examined the nature of ligands, counterions, and metal centers that govern distinct products in such transformations.10 Thus, a great challenge of Chan’s studies on chemoselectivity consists of identifying the details of the reaction mechanisms. On the basis of their experimental observations, we can conclude that a phosphine-gold(I) catalyst leads to the formation of tricyclic 1H-cyclopenta[b]naphthalene, while a gold(III) complex and PtCl2 as catalysts generate cis-cyclopenten-2-yl δ-diketone. The proposed mechanisms for these reactions are depicted in Scheme 2. The reaction pathway in the phosphine-gold(I)-catalyzed process may initially proceed by activation of the acetate alkynyl moiety in the substrate to give the gold(I) complex A.

This undergoes a 1,3-acyloxy migration and leads to the allene intermediate C and then the 1,3-dioxin-1-ium species B. A subsequent 5-exo-dig cyclization of C involving an anti addition of the allenic moiety to the CC bond in the ester would then provide the allene intermediate D.11 The next step is a Friedel− Crafts-type reaction of D involving an attack of the pendant phenyl group at the oxocarbenium moiety, leading to the generation of E (Scheme 2, pathway a). Protodeauration followed by rearomatization of the ensuing Wheland intermediate F would then furnish the product G. However, the use of the more Lewis acidic gold(III) catalyst12 and PtCl2 leads to the chemoselective transformation of D to ciscyclopenten-2-yl δ-diketone H via 1,5-acyl migration (Scheme 2, pathway b). Unfortunately, some key issues in the catalytic activity and selectivity remain obscure. (1) Although the mechanism for gold(I)- and gold(III)-catalyzed cycloisomerizations of 1,6-diyne esters has been proposed,13 it has still not been rationalized. (2) The origin of the divergent chemoB


Article

Organometallics

Figure 1. Calculated free energy profile for the consecutive 1,3-acyloxy migration and 5-exo-dig cyclization in the Au(I)- and Au(III)-catalyzed cycloisomerization of the 1,6-diyne ester 1. Formulas of the Au(I) and Au(III) complex catalysts are given at the top left and right corners, respectively. Relative energies are given in kcal/mol.

Figure 2. Selected structures for different pathways shown in Figure 1. Representative bond lengths (in Å) are indicated. with the reactants and products. Thus, the relative energies were corrected to the vibrational zero-point energies (ZPE, not scaled). The solvent effect was taken into consideration by using the M06-2X/6311++G(d,p) level. For Au, the small-core Los Alamos (LANL2TZ(f)) pseudopotentials and basis sets (these include the Dunning− Huzinaga full TZ and Los Alamos ECPs plus TZ) were employed with an extra f polarization function (BSII),20 using a single-point calculation with the integral equation formalism polarizable continuum model (IEF-PCM) in CH2Cl2 (ε = 8.9). The radii and nonelectrostatic terms were taken from the universal solvation model (SMD).21 Unless stated otherwise, all of the energies discussed in the main text are relative solvation-free energies (ΔGsol), which were obtained by adding the solvation corrections to the computed gas-phase relative free energies (ΔG298). It should be noted that a recent work comparing different DFT methods for studying gold-catalyzed reactions suggested that the M06 functional performs better than other DFT methods.22 To compare the results of the calculation methods, other DFT functionals, such as M06 and B3LYP, were also applied in our calculations, which led to the same conclusion (Figures S5−S11 in the Supporting Information).

selectivity of different catalysts is not clear, despite being of great importance for designing new reactions. (3) A presumably important role of the chloride moieties in the [AuIII(Cl)2(pic)] (pic = 2-picolinate) and PtCl2 catalysts toward the formation of cis-cyclopenten-2-yl δ-diketones has not been confirmed but can add new insight into the current understanding of the electronic effect of the catalysts. Thus, our motivation in this work is to clarify the details of the reaction mechanisms and explain the experimentally observed chemoselectivity by an in-depth mechanistic study.

2. COMPUTATIONAL DETAILS All calculations were conducted using the Gaussian 09 program package.14 The geometries of all species were fully optimized by DFT15 with the M06-2X16 method and using the 6-31G(d,p) basis set for all atoms except for Au; the Lanl2dz17 basis set (BS1) was applied for Au. This computational method was successfully applied in various mechanistic studies.18 Vibrational frequency calculations conducted at the M06-2X/6-31G(d,p) theoretical level were used to characterize all of the stationary points as either minima (number of imaginary frequencies, NIMAG = 0) or transition states (NIMAG = 1). In several significant cases, the calculation of intrinsic reaction coordinates (IRC)19 was performed to unambiguously connect the transition states

3. RESULTS AND DISCUSSION According to Scheme 2, both of the metal-catalyzed reactions of the 1,6-diyne ester cycloisomerization entail similar 1,3-acyloxy C


Article

Organometallics

Figure 3. Calculated free energy profile for the Friedel−Crafts-type mechanism (in blue) vs 1,5-acyl migration reaction (in black) catalyzed by [AuI(PPhMe2)(NCMe)]+.

mechanisms (1,3- or 1,2-acyloxy migration25) were located in our study. In A-1 or B-1, the coordination of the CC bond to the gold atom induces a cyclization of the carbonyl oxygen onto the triple bond. Thus, new and stable oxonium ion structures A-1a and B-1a are formed through the six-membered-ring transition states A-TSa1 and B-TSa1 (1,3-acyloxy migration). Such transition states were also found in the Au-catalyzed reactions of skeletal rearrangements and cycloadditions of enynes and cyclopropyl propargylic carboxylates.4,5 Figure 2 indicates that the Au and alkyne C1 atoms are connected in ATSa1 and B-TSa1 (Au−C1 distances are 2.161 and 2.097 Å, respectively). Nucleophilic attack of the O1 atom on the positively charged C2 atom leads to C2−O1 bond formation in A-TSa1 and B-TSa1, having distances of 2.086 and 2.150 Å, respectively. The calculated activation energies for this step are 13.8 and 19.6 kcal/mol for A-TSa1 and B-TSa1, respectively (Figure 1). The free energies of the reaction for the formation of A-1a and B-1a intermediates are 1.6 and −0.1 kcal/mol with respect to A-1 and B-1. A higher barrier for B-TSa1 over ATSa1 can be attributed to the following reasons. By examining the structural features of transition states and intermediates, we found that the torsional strain along the forming O1−C2 bond and the ring strain within the generated six-membered ring influence the activation barrier. For example, the corresponding O1−C2−C9−C10 dihedral angles in A-TSa1 and B-TSa1 are

migration and 5-exo-dig cyclization steps, while different reaction pathways are observed for the formation of vinyl metal complex intermediates. Along with the model 1,6-diyne ester substrate 1 (Scheme 2), mononuclear gold(I) [Au(PPhMe2)(NCMe)]+ and gold(III) [Au(Cl)2(pic)] (pic = 2picolinate)23 complexes as well as PtCl2 were used as model catalysts in DFT calculations. 3.1. Consecutive 1,3-Acyloxy Migration and 5-exo-dig Cyclization in the Au(I)- and Au(III)-Catalyzed Reactions. Energy profiles for the 1,3-acyloxy migration and 5-exo-dig cyclization reaction pathways are shown in Figure 1. The selected optimized geometries for the reactants, intermediates, transition states, and reaction products are schematically shown in Figure 2, along with important bond distances. Unless otherwise noted, the relative free energies discussed in subsequent sections refer to the values in CH2Cl2. The detailed structural parameters and energies for various structures are described in the Supporting Information. From the energy profile, it is evident that the first step of the 1,3-acyloxy migration involves the formation of a preliminary intermediate, namely A-1 for Au(I) or B-1 for Au(III). Figure 1 shows that A-1 and B-1 are stabilized by the coordination of the Au atom to the CC bond.24 Furthermore, A-1 or B-1 can be formed exothermically (energies of −7.8 and −7.1 kcal/mol, respectively). After the generation of A-1 or B-1, two general D


Article

Organometallics

Figure 4. Calculated free energy profile for the Friedel−Crafts-type mechanism (in blue) vs 1,5-acyl migration reaction (in black) catalyzed by [AuIII(Cl)2(pic)].

provide the putative vinyl gold complexes A-3a and B-3a26 through the A-TSa3 and B-TSa3 transition states, respectively.27 Figure 2 indicates that the C1−C4 distances for A-TSa3 and B-TSa3 are 2.225 and 2.295 Å, respectively. For the A and B series, the calculated activation energies for this step are 12.3 and 13.5 kcal/mol, respectively. A lower free energy of activation for A-TSa3 vs B-TSa3 indicates that the 5-exo-dig cyclization is easier in the Au(I)-catalyzed reaction. In this case, the σ-acceptor character of the two catalysts is the main reason for the calculated differences in activation energy. When the Au(I) catalyst is used, the charged species activate more effectively the alkyne toward nucleophilic attack (5-exo-dig cyclization).23 Furthermore, different charge distributions in A2a and B-2a may account for this phenomenon. The NBO charges for the C1 and C2 atoms of A-2a and B-2a are 0.113, 0.152 au and 0.256, 0.063 au, respectively. A larger difference between the NBO charges of the C1 and C2 atoms in B-2a indicates that the Au(III) atom has a more pronounced effect on the charge distribution. In addition, a more negative charge of the C1 atom in B-2a makes the 5-exo-dig cyclization more difficult for series B over series A.

132.7 and 68.5°, respectively, implying that the torsional strain in B-TSa1 is much more significant than that in A-TSa1. In A1a and B-1a, it is evident that the C1−C2 triple bond completes its change from a triple bond to a double bond (1.332 Å) and the C2−O1 bond becomes completely formed (1.447 Å). In A1a or B-1a, the C3−O2 and the C2−O1 bonds are weaker than a normal C−O single bond (1.43 Å), as demonstrated by their rather long distances of 1.478 and 1.445 Å, respectively. The fact that the C3−O2 bond is breaking and the C2−O1 bond is forming may account for an interesting phenomenon. Due to the strain of a six-membered ring, the oxonium ion structure A1a or B-1a converts to an allenyl structure A-2a or B-2a via the four-electron rearrangement transition state A-TSa2 or BTSa2. In A-TSa2 and B-TSa2, the distances of the breaking C3−O2 bonds are 2.000 and 2.077 Å, respectively. The activation free energy for the second step is 5.0 and 11.0 kcal/mol, and the formation of A-2a is an exothermic process (the free energy of the reaction for A-2a is −1.7 kcal/mol with respect to A-1a). However, the formation of B-2a is endothermic by 0.1 kcal/mol. In the following step, the 5exo-dig cyclization of A-2a and B-2a involving an anti addition of the allenic moiety to the CC bond in the ester would then E


Article

Organometallics

Figure 5. Selected structures for different pathways shown in Figures 3 and 4. Representative bond lengths (in Å) are indicated.

3.2. Friedel−Crafts Mechanism vs 1,5-Acyl Migration for Au(I)- and Au(III)-Catalyzed Reactions. After the formation of A-3a or B-3a, at least two possible pathways can be envisioned. Pathway a is associated with a direct Friedel−Crafts-type reaction involving an attack of the pendant group to the oxocarbenium moiety followed by a series of proton transfer steps. An alternative pathway b consists of generating an intermediate A-4b or B-4b with a rotation of an oxonium side chain. cis-Cyclopenten-2-yl δ-diketone is formed after the 1,5-carbonyl migration that is assisted by chloride ligands. Energy profiles for the Au(I)- and Au(III)-catalyzed cycloisomerizations catalyzed by [AuI(PPhMe2)(NCMe)]+ and [AuIII(Cl)2(pic)] are given in Figures 3 and 4, respectively. The selected optimized geometries for the reactants, intermediates, transition states, and reaction products are depicted in Figure 5. For the series A catalyzed by [AuI(PPhMe2)(NCMe)]+, pathway a proceeds via the transition structure A-TSa4 with an energy barrier of 7.3 kcal/mol to afford Wheland intermediate A-4a. Then, the release of AcOH should proceed preferentially via the following transformation: A-4a (−25.0 kcal/mol) → ATSa5 (−16.7 kcal/mol) → A-5a (−53.3 kcal/mol). A subsequent deprotonation proceeds by crossing barriers of 10.8 and −2.0 kcal/mol (through A-TSa6 and A-TSa7) for A6a and A-7a, respectively. The whole catalytic process is exothermic and lower in energy by 79.8 kcal/mol in comparison to the starting reactants. In pathway b, the rotation of an oxonium side chain would give A-4b, which is energetically less stable by 2.8 kcal/mol than A-3a. This indicates that the rotation of the oxonium side chain in B-3a is thermodynamically unfavored. A subsequent 1,5-acyl migration generates the diketone skeleton A-5b, which would need 17.2

kcal/mol to overcome the corresponding transition state ATSb5. By a review of the two aforementioned mechanistic pathways, 1,5-acyl migration can be predicted to be the ratedetermining step. A higher activation energy for A-TSb5 indicates that the formation of the B-6b product from B-1 via pathway b should be disfavored. These calculation results suggest that the first step of the reaction catalyzed by [AuI(PPhMe2)(NCMe)]+ will cause an initial 1,3-acyloxy migration via a six-membered-ring transition state to form an allene intermediate. A subsequent 5-exo-dig cyclization of this intermediate then provides the putative vinyl gold complex. Next, the Friedel−Crafts-type reaction will proceed and two consecutive migrations of the hydrogen atom (assisted by AcOH) would induce the formation of the 1H-cyclopenta[b]naphthalene and the regeneration of the catalyst. For the series B catalyzed by [AuIII(Cl)2(pic)], the cyclic intermediate B-3a also undergoes the Friedel−Crafts reaction and proton transfer assisted by AcOH (pathway a). The Gibbs activation barriers for the transition structures B-TSa5, B-TSa6, and B-TSa7 are 13.4, 13.3, and 13.6 kcal/mol, respectively. For pathway b, rotation of the oxonium side chain would also give B-4b. In contrast to the series A, B-4b is 0.7 kcal/mol inferior in energy to B-3a. Hence, the rotation of the oxonium side chain in B-3a can be thermodynamically feasible. In addition, we do not find transition structures which could be responsible for the direct 1,5-acyl migration. On the contrary, a stepwise acyl group migration process assisted by chloride ligands was identified in our calculation results. The first step involves the migration of an acyl group to the chloride ligand in the catalyst, while the second step corresponds to the migration of the chloride moiety to the carbon coordinated to gold; this affords F


Article

Organometallics

Figure 6. Calculated free energy profile for the Friedel−Crafts-type mechanism (in blue) vs 1,5-acyl migration (in black) in the Pt(II)-catalyzed cycloisomerization of the 1,6-diyne ester. Relative energies are given in kcal/mol.

the final product B-6b, which is accompanied by the regeneration of the gold(III) catalyst. It should be mentioned that a ligand-assisted acyl migration in the Au-catalyzed isomerization of propargylic ester has also been reported by Das et al.28 Calculations indicate that the activation free energies for the present acyl group migration are 9.4 and 9.1 kcal/mol, respectively. Overall, the two mechanistic pathways of the series B reveal lower activation free energies for an acyl group migration process, thus indicating that the formation of B-7a from B-1 via pathway a should be disfavored. These calculation results suggest that the first step of the reaction catalyzed by [Au(Cl)2(pic)] will cause an initial 1,3-acyloxy migration via a six-membered-ring transition state to form an allene intermediate. Subsequent 5-exo-dig cyclization of this intermediate then provides the putative vinyl gold complex. Next, the 1,5-acyl migration assisted by chloride ligands would induce the formation of the cis-cyclopenten-2-yl δ-diketone and the regeneration of the catalyst. Our calculation results are consistent with the experimental data of Chan and co-workers.7 3.3. Chemoselectivity in the Au(I)- and Au(III)Catalyzed Reactions. The obtained calculation results for the Au(I)-catalyzed reaction (series A) indicate that the ATSa4 and A-TSb5 transition states play a vital role in the 1,6diyne ester cycloisomerization. Moreover, the relative energy difference between the activation energies of A-TSa4 and ATSb5 is 10.4 kcal/mol, which suggests that the first step of the reaction cycle will cause an initial 1,3-acyloxy migration via a six-membered-ring transition state to form an allene

intermediate. Subsequent 5-exo-dig cyclization of this intermediate then provides the putative vinyl gold complex. Next, the Friedel−Crafts-type reaction will proceed and two consecutive migrations of the hydrogen atom (assisted by AcOH) would induce the formation of the final product and the regeneration of the catalyst. When the Au(III) catalyst is used (series B), the B-TSa4 and B-TSb5 transition states will account for the formation of the final product. The relative energy difference between the activation energies of B-TSa4 and B-TSb5 is 4.0 kcal/mol. This indicates that the Au(III) catalyst leads to chemoselective formation of cis-cyclopenten-2yl δ-diketone via the 1,5-acyl migration, followed by the 1,3acyloxymigration/5-exo-dig cyclization. Very recently, Jiang et al. reported a comprehensive utilization of both electronic and steric properties of ligands in homogeneous gold catalysis in the regiodivergent intramolecular hydroarylation of alkynes and proposed the concepts of “ligand-compensated π-system-pulling” and “ligand-directed π-system-pushing”.29 Then, Waldmann et al. investigated a ligand-directed synthetic approach for the gold(I)-catalyzed cycloisomerization of oxindole-derived 1,6-enynes that affords distinct molecular scaffolds as a result of different catalytic pathways.30 Hence, our investigations commenced by utilizing the ligand effect in the cycloisomerization of 1,6-diyne esters, with an objective to computationally understand how ligands can steer the catalytic behavior and lead to divergent catalytic approaches toward structural and functional scaffold diversity. Electrophilic G


Article

Organometallics

4. CONCLUSIONS In summary, the mechanisms and chemoselectivity in the Au(I)-, Au(III)-, and Pt(II)-catalyzed cycloisomerizations of the model 1,6-diyne ester substrate 1 were investigated in detail by DFT calculations. The experimental chemoselectivites were confirmed by the calculations of energy profiles for the cycloisomerization of 1. On the basis of our calculation results, the following conclusions can be drawn: (1) Both the Au(I) and Au(III) complex catalysts and PtCl2 entail similar 1,3-acyloxy migration and 5-exo-dig cyclization steps, whereas different reaction pathways are observed for the formation of putative vinyl gold complex intermediates. (2) In the [AuI(PPhMe2)(NCMe)]+-catalyzed reaction, an electron-deficient phosphine ligand would be a π system pulling process, resulting in a Friedel−Crafts reaction and generating 1H-cyclopenta[b]naphthalenes. (3) However, for the more Lewis acidic gold(III) catalyst [AuIII(Cl)2(pic)] bearing electron-rich ligands, a π system pushing would be achieved. The putative vinyl metal complex intermediate endures a rotation of the oxonium moiety to generate a new intermediate. After the 1,5-carbonyl migration assisted by chloride ligands, cis-cyclopenten-2-yl δ-diketone is formed. (4) The obtained computational results are in very good agreement with the experimental data, suggesting that the chemoselectivity in such cycloisomerization reactions can be attributed to the electronic properties of the different catalysts and steric interactions between the two side chains. We believe that these new mechanistic insights revealed by the DFT calculations in the present study will contribute to a better understanding of the mechanisms and chemoselectivity features in other types of related organic transformations.

gold complexes, tuned by electron-deficient ligands, will permit coordination with a directing group (DG), which simultaneously “pulls” an Au-coordinated π system to the sterically hindered ortho position on aromatic rings for further cyclization. Alternatively, the rigid, bulky, and electron-rich ligands supplement steric hindrance and electronic properties, which eventually “push” the π system to the para position.29 In Chan’s experimental research, the divergent chemoselectivity was possible by harnessing the inherent differences in electronic and steric properties of the gold complexes.7 On the basis of the concepts proposed by Jiang et al.,29 an electron-deficient phosphine ligand in [AuI(PPhMe2)(NCMe)]+ would be a πsystem pulling process, resulting in Friedel−Crafts reaction and generating 1H-cyclopenta[b]naphthalenes. However, for the more Lewis acidic gold(III) catalyst [AuIII(Cl)2(pic)] bearing electron-rich ligands, a π-system pushing would be achieved. Therefore, the chloride ligands of the catalyst play a ligandcompensating role, pushing the reaction to another direction that is an acyl migration to give the cis-cyclopenten-2-yl δdiketones. 3.4. PtCl2-Catalyzed Reaction. According to experimental results by the groups of Chan and Chen,7,8 a similar mechanism can also be calculated for the ligand-assisted 1,5-acyl migration catalyzed by PtCl2. From Figure 6, it is evident that the first step of the 1,3-acyloxy migration also involves the formation of the preliminary intermediate C-1. Then C-1 follows a mechanism similar to that described above for A-1 and B-1, involving two sections: (i) consecutive 1,3-acyloxy migration and 5-exo-dig cyclization and (ii) Friedel−Crafts reaction vs 1,5acyl migration. The PtCl2-catalyzed cycloisomerization cycle starts with a two-step 1,3-acyloxy migration, generating C-2a through the C-TSa1 or C-TSa2 transition states with activation energies of 3.0 and 8.9 kcal/mol, respectively. Similarly, the intermediate C-2a undergoes 5-exo-dig cyclization leading to the formation of C-3a via C-TSa3; this is a rate-determining step of the reaction with an activation free energy of 18.5 kcal/ mol. In addition, the formation of C-3a is strongly exergonic by 44.7 kcal/mol. Then, we investigated the mechanisms for competitive generations of C-7a or C-6b. Intermediate C-3a undergoes a Friedel−Crafts-type reaction, affording the Wheland intermediate C-4a with a barrier of 10.8 kcal/mol. Similar to the case for the above systems, the subsequent steps are a series of hydrogen migration reactions assisted by AcOH. These are the steps C-4a → C-5a, C-5a → C-6a, and C-6a → C-7a, having barriers of 7.4, 14.8, and 16.2 kcal/mol, respectively. On the other hand, pathway b that leads to the C-6b product and is accompanied by a rotation of the oxonium side chain in C-3a can now take place more readily to give C-4b. C-4b is 0.7 kcal/ mol inferior in energy to C-3a. Similar to the series B, a stepwise acyl group migration process assisted by chloride ligands was also located in our calculation results. The migrating acyl group prefers to attach itself to the chloride moiety in the first step (C-4b → C-5b). In the next step, it migrates to the carbon atom bonded to the platinum center (C5b → C-6b). The step of acyl migration to form C-5b and C6b through C-TSb5 and C-TSb6 involves activation barriers of 9.1 and 6.1 kcal/mol, respectively. Hence, the chloride atom placed at the vicinity of the acyl group (in C-4b and C-5b) mediates the shifting process by acting as a platform atom during the migration of the acyl group from the oxygen to the carbon atom.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.organomet.7b00813. Additional calculated energy parameters, optimized structures, energy profiles, and thermodynamic properties of structures (PDF) Optimized Cartesian coordinates with the self-consistent field (SCF) energies and the imaginary frequencies of transition states (XYZ)

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail for R.F.: [email protected]. *E-mail for L.Y.: [email protected]. ORCID

Ran Fang: 0000-0001-6804-6572 Lizi Yang: 0000-0002-2855-0149 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research has been supported by the National Natural Science Foundation of China (Nos. 21672090, 21301080, and 21203080) and the Fundamental Research Funds for the H


Article

Organometallics

Gimbert, Y.; Gandon, V.; Fensterbank, L. ACS Catal. 2016, 6, 5146− 5160. (11) (a) Hashmi, A. S. K.; Schuster, A. M.; Rominger, F. Angew. Chem., Int. Ed. 2009, 48, 8247−8249. (b) Hashmi, A. S. K.; Ramamurthi, T. D.; Rominger, F. Adv. Synth. Catal. 2010, 352, 971−975. (12) (a) Bolm, C.; Kasyan, A.; Drauz, K.; Günther, K.; Raabe, G. Angew. Chem., Int. Ed. 2000, 39, 2288−2290. (b) Hashmi, A. S. K.; Frost, T. M.; Bats, J. W. J. Am. Chem. Soc. 2000, 122, 11553−11554. (c) Fukuda, Y.; Utimoto, K. J. Org. Chem. 1991, 56, 3729−3731. (13) (a) Pernpointner, M.; Hashmi, A. S. K. J. Chem. Theory Comput. 2009, 5, 2717−2725. (b) Lein, M.; Rudolph, M.; Hashmi, A. S. K.; Schwerdtfeger, P. Organometallics 2010, 29, 2206−2210. (14) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision D.01; Gaussian, Inc., Wallingford, CT, 2009. (15) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 1372−1377. (b) Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 785−789. (c) Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200−206. (16) (a) Zhao, Y.; Schultz, N. E.; Truhlar, D. G. J. Chem. Theory Comput. 2006, 2, 364−382. (b) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2006, 110, 13126−13130. (17) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270−283. (b) Wadt, W. R.; Hay, P. J. J. Chem. Phys. 1985, 82, 284−298. (18) (a) Zhao, Y.; Truhlar, D. G. J. Phys. Chem. A 2008, 112, 1095− 1099. (b) Jacquemin, D.; Perpète, E. A.; Ciofini, I.; Adamo, C.; Valero, R.; Zhao, Y.; Truhlar, D. J. Chem. Theory Comput. 2010, 6, 2071−2085. (c) Zhao, Y.; Truhlar, D. G. J. Chem. Theory Comput. 2008, 4, 1849− 1868. (19) (a) Fukui, K. J. Phys. Chem. 1970, 74, 4161−4163. (b) Gonzalez, C.; Schlegel, H. B. J. Chem. Phys. 1989, 90, 2154−2161. (c) Gonzalez, C.; Schlegel, H. B. J. Phys. Chem. 1990, 94, 5523−5527. (20) Roy, L. E.; Hay, P. J.; Martin, R. L. J. Chem. Theory Comput. 2008, 4, 1029−1031. (21) (a) Marenich, A. V.; Cramer, C. J. J. Phys. Chem. B 2009, 113, 6378−6396. (b) Ribeiro, R. F.; Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. J. Phys. Chem. B 2011, 115, 14556−14562. (22) Kang, R.; Chen, H.; Shaik, S.; Yao, J. J. Chem. Theory Comput. 2011, 7, 4002−4011. (23) Hashmi, A. S. K.; Weyrauch, J. P.; Rudolph, M.; Kurpejović, E. Angew. Chem., Int. Ed. 2004, 43, 6545−6547. (24) For a detailed discussion of the coordination bond between alkynes and gold catalysts, see: (a) Bistoni, G.; Belanzoni, P.; Belpassi, L.; Tarantelli, F. J. Phys. Chem. A 2016, 120, 5239−5247. For fundamental studies on the activation of alkynes by gold(I) or gold(III) catalysts and a comparison with the isoelectronic platinum(II) catalysts, see: (b) Pernpointner, M.; Hashmi, A. S. K. J. Chem. Theory Comput. 2009, 5, 2717−2725. (c) Krauter, C. M.; Hashmi, A. S. K.; Pernpointner, M. ChemCatChem 2010, 2, 1226− 1230. (d) Lein, M.; Rudolph, M.; Hashmi, A. S. K.; Schwerdtfeger, P. Organometallics 2010, 29, 2206−2210. (25) Apart from the 1,3-acyloxy migration, the Rautenstrauch-type 1,2-acyloxy migration via an oxirenium intermediate gives rise to another possible reaction pathway. For the A and B series, the free energies of activation for the RDS are 19.7 and 22.5 kcal/mol,

Central Universities (Grant No. lzujbky-2017-108). A.M.K. acknowledges the FCT of Portugal (UID/QUI/00100/2013).

■

REFERENCES

(1) (a) Jia, M.; Bandini, M. ACS Catal. 2015, 5, 1638−1652. (b) Friend, C. M.; Hashmi, A. S. K. Acc. Chem. Res. 2014, 47, 729− 730. (c) Braun, I.; Asiri, A. M.; Hashmi, A. S. K. ACS Catal. 2013, 3, 1902−1907. (d) Krause, N.; Winter, C. Chem. Rev. 2011, 111, 1994− 2009. (e) Dorel, R.; Echavarren, A. M. J. Org. Chem. 2015, 80, 7321− 7332. (f) Rao, W.; Susanti, D. W.; Ayers, B. J.; Chan, P. W. H. J. Am. Chem. Soc. 2015, 137, 6350−6355. (g) Gulevich, A. V.; Dudnik, A. S.; Chernyak, N.; Gevorgyan, V. Chem. Rev. 2013, 113, 3084−3213. (h) Schmidbaur, H.; Raubenheimer, H. G.; Dobrzanska, L. Chem. Soc. Rev. 2014, 43, 345−380. (i) Pflästerer, A.; Hashmi, A. S. K. Chem. Soc. Rev. 2016, 45, 1331−1367. (2) (a) Wei, Y.; Shi, M. ACS Catal. 2016, 6, 2515−2524. (b) Frutos, M.; Avello, M. G.; Viso, A.; Pradilla, R. F. Org. Lett. 2016, 18, 3570− 3573. (c) Hosseyni, S.; Wojtas, L.; Li, M.; Shi, X. D. J. Am. Chem. Soc. 2016, 138, 3994−3997. (d) Corma, A.; Perez, A. L.; Sabater, M. J. Chem. Rev. 2011, 111, 1657−1712. (e) Chen, K.; Shu, Q.; Schmittel, M. Chem. Soc. Rev. 2015, 44, 136−160. (f) Hansmann, M. M.; Rominger, F.; Hashmi, A. S. K. Chem. Sci. 2013, 4, 1552−1559. Rudolph, M.; Hashmi, A. S. K. Chem. Soc. Rev. 2012, 41, 2448−2462. (3) (a) He, C.; Guo, S.; Ke, J.; Hao, J.; Xu, H.; Chen, H.; Lei, A. W. J. Am. Chem. Soc. 2012, 134, 5766−5769. (b) Studer, A.; Curran, D. P. Angew. Chem., Int. Ed. 2016, 55, 58−102. (c) Mothe, S. R.; Lauw, S. J.; Kothandaraman, P.; Chan, P. W. H. J. Org. Chem. 2012, 77, 6937− 6947. (d) Susanti, D.; Koh, F.; Kusuma, J. A.; Kothandaraman, P.; Chan, P. W. H. J. Org. Chem. 2012, 77, 7166−7175. (e) Teo, W. T.; Rao, W.; Ng, C. J. H.; Koh, S. W. Y.; Chan, P. W. H. Org. Lett. 2014, 16, 1248−1251. (4) (a) Rao, W.; Chan, P. W. H. Chem. - Eur. J. 2014, 20, 713−718. (b) Lauterbach, T.; Ganschow, M.; Hussong, M.; Rudolph, M.; Rominger, F.; Hashmi, A. S. K. Adv. Synth. Catal. 2014, 356, 680−686. (c) Yan, J.; Tay, G. L.; Neo, C.; Lee, B. R.; Chan, P. W. H. Org. Lett. 2015, 17, 4176−4179. (d) Hashmi, A. S. K.; Yang, W.; Yu, Y.; Hausmann, M. M.; Rudolph, M.; Rominger, F. Angew. Chem., Int. Ed. 2013, 52, 1329−1332. (e) Yu, Y.; Yang, W.; Rominger, F.; Hashmi, A. S. K. Angew. Chem., Int. Ed. 2013, 52, 7586−7589. (f) Fang, R.; Yang, L.; Wang, Y. Org. Biomol. Chem. 2011, 9, 2760−2770. (g) Yang, L.; Fang, R. J. Mol. Catal. A: Chem. 2013, 379, 197−205. (h) Li, Y.; Kirillov, A. M.; Fang, R.; Yang, L. Organometallics 2017, 36, 1164− 1172. (i) Xia, Y.; Huang, G. J. Org. Chem. 2010, 75, 7842−7854. (j) Huang, G.; Cheng, B.; Xu, L.; Li, Y.; Xia, Y. Chem. - Eur. J. 2012, 18, 5401−5415. (k) Cheng, B.; Huang, G.; Xu, L.; Xia, Y. Org. Biomol. Chem. 2012, 10, 4417−4423. (l) Zhou, T.; Xu, L.; Xia, Y. Org. Lett. 2013, 15, 6074−6077. (5) (a) Nicolaou, K. C.; Chen, J. S. Chem. Soc. Rev. 2009, 38, 2993− 3009. (b) Soriano, E.; Fernandez, I. Chem. Soc. Rev. 2014, 43, 3041− 3105. (c) Ting, C.-M.; Hsu, Y.-L.; Liu, R.-S. Chem. Commun. 2012, 48, 6577−6579. (d) Basak, A.; Chakrabarty, K.; Ghosh, A.; Das, G. K. J. Org. Chem. 2013, 78, 9715−9724. (e) Yang, W.; Yu, Y.; Zhang, J.; Hansmann, M. M.; Pflästerer, D.; Hashmi, A. S. K. Adv. Synth. Catal. 2013, 355, 2037−2043. (f) Oh, C. H.; Kim, J. H.; Piao, L.; Yu, J.; Kim, S. Y. Chem. - Eur. J. 2013, 19, 10501−10505. (g) Rao, W.; Koh, M. J.; Kothandaraman, P.; Chan, P. W. H. J. Am. Chem. Soc. 2012, 134, 10811−10814. (h) Zhang, D.-H.; Zhang, Z.; Shi, M. Chem. Commun. 2012, 48, 10271−10279. (6) Rao, W.; Koh, M. J.; Li, D.; Hirao, H.; Chan, P. W. H. J. Am. Chem. Soc. 2013, 135, 7926−7932. (7) Li, D.; Rao, W.; Tay, G. L.; Ayers, B. J.; Chan, P. W. H. J. Org. Chem. 2014, 79, 11301−11315. (8) Chen, Z.; Jia, X.; Huang, J.; Yuan, J. J. Org. Chem. 2014, 79, 10674−10681. (9) Rao, W.; Susanti, D.; Ayers, B. J.; Chan, P. W. H. J. Am. Chem. Soc. 2015, 137, 6350−6355. (10) Jaroschik, F.; Simonneau, A.; Lemière, G.; Cariou, K.; Agenet, N.; Amouri, H.; Aubert, C.; Goddard, J.-P.; Lesage, D.; Malacria, M.; I


Article

Organometallics respectively. These values indicate that the 1,2-acyloxy migration pathway is unfavorable. See the Supporting Information for details. (26) (a) Hashmi, A. S. K. Gold Bull. 2009, 42, 275−279. (b) Hashmi, A. S. K.; Schuster, A. M.; Rominger, F. Angew. Chem., Int. Ed. 2009, 48, 8247−8249. (c) Hashmi, A. S. K.; Ramamurthi, T. D.; Rominger, F. Adv. Synth. Catal. 2010, 352, 971−975. (27) There is also a possibility of dual activation of the allene intermediate. For [Au(Cl)2(pic)], the free energy of activation for such a dual activation is 33.8 kcal/mol. This value indicates that the pathway involving the dual activation of the allene intermediate is unfavorable. (28) Ghosh, A.; Basak, A.; Chakrabarty, K.; Ghosh, B.; Das, G. K. J. Org. Chem. 2014, 79, 5652−5663. (29) Ding, D.; Mou, T.; Feng, M.; Jiang, X. J. Am. Chem. Soc. 2016, 138, 5218−5221. (30) Lee, Y.-C.; Patil, S.; Golz, C.; Strohmann, C.; Ziegler, S.; Kumar, K.; Waldmann, H. Nat. Commun. 2017, 8, 14043−14054.

J


How Does the Catalyst Affect the Reaction Pathway? DFT Analysis of

Recommend Documents