Au-Catalyzed Hexannulation and Pt-Catalyzed Pentannulation of

This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article Cite This: ACS Omega 2018, 3, 1159−1169

Au-Catalyzed Hexannulation and Pt-Catalyzed Pentannulation of Propargylic Ester Bearing a 2‑Alkynyl-phenyl Substituent: A Comparative DFT Study Animesh Ghosh, Atanu Basak, Kuheli Chakrabarty, Sonjoy Mondal, Arpita Chatterjee, and Gourab Kanti Das* Department of Chemistry, Visva-Bharati, Santiniketan 731235, West Bengal, India S Supporting Information *

ABSTRACT: The mechanistic pathways of metal-catalyzed pentannulation and hexannulation of aromatic enediyne were studied quantum mechanically with Pt and Au salts. In agreement with the experimental facts, our result shows that the pentannulation favors over the hexannulation under Pt-catalyzed conditions and the reverse possibility favors when the Pt salt is replaced with an Au one. The Ptcatalyzed reaction involves a long-range acyl migration that follows the cyclization step. Our study reveals that such migration takes place under the assistance of a ligand of the metal atom. Moreover, the variation of aromaticity (probed by the change of the nucleus-independent chemical shift (0) value) in the cyclization steps shows that both processes maintain the development of the aromatic character of the generated intermediate during the progress of the reaction.

■

INTRODUCTION Propargylic esters serve as an important starting material in the chemistry of platinum- and gold-catalyzed reactions because they are easily available in the form of densely functionalized derivatives and offer a general route of [1,2]- or [1,3]-ester migration to form metal carbenoids or metal-coordinated allene intermediates. The rearranged intermediates may further undergo some interesting reactions depending upon the reactive moiety present in the environment.1 A tandem sequence of rearrangement reactions to produce useful carbon skeletal motifs is found in a large number of recent contributions. The selectivity of initial [1,2]- or [1,3]-migration of ester depends on the substituent present in the reactant and the further evolution of the intermediate thus generated.2 One class of the reaction pattern is the intramolecular nucleophilic attack by an atom such as carbon,3 oxygen,4 or nitrogen.5 Among them, a notable example has been reported by Zhao et al.,6 where benzannulated products are produced from diynol esters by following the sequence of the [3,3] sigmatropic rearrangement/cycloisomerization reaction under Au- or Agcatalyzed conditions (Scheme 1). The cyclization may be considered as a transition metal-catalyzed version of Myers− Saito cyclization7 for the synthesis of fused aromatic ketones. In this sequence, the generated intermediate enyne−allene undergoes a 6-endo-dig cyclization through the metalpromoted activation of an alkyne moiety (Scheme 1). Though the behavior of the Pt metal has been found to be similar to that of the Au metal in catalyzing several cycloisomerization reactions,8 a recent report of Chen et al.9 shows that the alteration of the catalytic condition by replacing Au or Ag salt © 2018 American Chemical Society

with Pt one follows a different pathway. Similar substrates undergo an unexpected 5-exo-dig cyclization under PtCl2catalyzed conditions and furnish a pentannulated benzofulvene derivative (Scheme 1). Moreover, the reaction follows an unusual long-range migration of the acyl group to generate a methylene-benzofulvenediketone. While rationalizing the mechanistic steps, the authors assumed an initial Pt-catalyzed [3,3] sigmatropic rearrangement of a propargylic ester to a reactive carboxyallene species that finally undergoes 5-exo-dig cyclization to generate the pentannulated compound. The final [1,5]-acyl migration furnishes a single stereoisomeric Zconfigured end product. The variation in reaction pathways under two different catalytic conditions caught our attention and promoted us to make a detailed theoretical investigation on the mechanism for locating the guiding factors that stimulate the formation of different products under two environments. A previous report on dynamics study on uncatalyzed versions of enyne−allene cyclization shows that the reaction may be interpreted in terms of formation of a highly asynchronous transition state near the concerted/stepwise boundary.10 However, the catalytic process should follow a different pathway and necessitates the reinvestigation of the possible mechanisms under the presence of metal salts that catalyze the reaction. In addition, our previous report on Au-catalyzed acyl migration shows an unusual result, where the migrating group Received: November 30, 2017 Accepted: January 16, 2018 Published: January 29, 2018 1159

DOI: 10.1021/acsomega.7b01889 ACS Omega 2018, 3, 1159−1169

Article

ACS Omega Scheme 1. Divergent Products Generated in Two Different Metal-Catalyzed Conditions

Scheme 2. Possible Pathways for Pentannulation and Hexannulation under the Catalysis of PtCl2a

a

The colored curly arrows show the migrations of the acyl group rather than the electronic movement.

was found to take the assistance of a ligand of the Au complex to generate the final product.11 Observing the unusual longrange acyl migration in the present problem of Pt-catalyzed cyclization, we suspected a similar type of ligand-assisted mechanism for such rearrangement reactions that take place through a multistep process. Searching and analyzing several possibilities reveal that such migration occurs through the assistance of ligands to follow an energetically favorable pathway which was not evident before this investigation. Moreover, the paradox of favoring a nonaromatic-fused benzofulvene derivative rather than a stable aromatic naphthyl ring structure under Pt-catalyzed conditions has also been addressed by the study of the developing aromaticity of the whole structure during the cyclization process. Analysis of the

molecular orbitals also interprets the requirement of a relatively high activation barrier for platinum-catalyzed cyclization steps, which is consistent with the experimental condition required for two different metal-catalyzing environments. The Results and Discussion section has been organized according to the following sequence: (1) computation and analysis of the possible pathways in the Pt-catalyzed enediyne rearrangement to the pentannulated benzofulvene derivative. (2) Comparative study of the Pt-catalyzed pentannulation to that of hexannulation. (3) Computation and comparative study on the Au-catalyzed pentannulation and hexannulation of the enediyne. (4) Analysis of the divergent pathway of Pt- and Aucatalyzed annulation on the basis of electronic distribution of key stationary points on the favorable pathways. 1160


Article

ACS Omega

Figure 1. Energy profile showing stationary points in pathway-1 (magenta), 2 (blue), and 3 (red) for pentannulation under Pt-catalyzed conditions.

■

RESULTS AND DISCUSSION For the study on all the metal-catalyzed cyclizations, we have chosen the benzenediyne ester (1 in Scheme 2 and R1 = R2 = R = CH3 in Scheme 1) as a model structure in which the metal center binds the triple bonds of the two substituents on the benzene nucleus. Computation and Analysis of Pt-Catalyzed Pentannulation Pathways. As discussed in the introduction, our exploration starts by designing the most probable pathways through which the reaction under metal-catalyzed conditions is taking place. A short schematic presentation of the studied pathways is shown in Scheme 2. Each pathway starts from a PtCl2-bound substrate structure (1), in which the metal ion exists in a square planar geometry by forming a chelate complex with the two triple bonds of the substrate. The first step involves a [3,3] sigmatropic shift to produce a metal-bound allene intermediate (2). In pathway-1, the allene intermediate undergoes a 5-exo dig attack by the triple bond to achieve pentannulation, thus forming the basic benzofulvene skeleton (3). This intermediate finally converges to the Pt-bound alkyne system (P) through a direct acyl migration from the acetate oxygen to its ultimate destination (indicated by the blue arrow in pathway-1). This pathway shows maximum resemblance to the proposed pathway,9 where the final migration involves an [1,5] shift of the acyl group. Pathway-2 also shows a sequence similar to that of pathway-1 during the formation of the benzofulvene derivative (3), that is, allene formation followed by cyclization. However, this pathway deviates from pathway-1 when the acyl migration takes place with the assistance of several atoms in a step wise fashion. During the acyl transfer, the migrating acyl group first jumps to the chlorine ligand (shown by the blue

arrow in pathway-2) and then reaches to the carbon atom adjacent to the final destination (shown by green arrow). The last migration step (shown by red arrow) results in the final product P. Pathway-3 differs from pathway-1 and 2 in the sequence of cyclization and acyl migration steps. Here, the allene derivative undergoes an acyl migration process (2 to 4) (may be characterized as a [1,9] shift of the acyl group), followed by the final cyclization step to generate the product (P). Shifting of the acyl group from oxygen to the alkyne carbon occurs through two steps (shown by blue and green arrows on pathway-3) in which the intermediate shows a bond between the migrating acyl group and the metal atom. Detailed potential energy surfaces of the studied pathways are shown in Figure 1 (for clarity, the geometry of the stationary points associated with pathway-2 only are shown in this diagram as this pathway was found to be energetically the most favorable one. Geometries of pathway-1 and 3 are shown in the Supporting Information). In all three pathways, the initial two steps represent the allene formation consisting of two transition structures (1-TS1 and 1-TS2) and an intermediate (1-I2) between them (Figure 1). In the first step, the metalbound substrate 1-I1 transforms into a six-membered cyclic intermediate 1-I2 (through 1-TS1) by the nucleophilic attack of the acetoxy oxygen on the distal terminus of the triple bond. This step is followed by the ring-opening process through 1TS2, the end product of which is the allene bound metal complex (1-I3). In pathway-1, the intermediate 1-I3 undergoes a metal-catalyzed cyclization step through a 5-exo-dig mode (1TS3) in which the middle carbon atom of the allenic moiety joins with the acetylenic carbon atom (connected directly to the aromatic ring) and generates pentannulated intermediate 1I4 (structure 3 in Scheme 2). The activation free energy 1161


Article

ACS Omega

Table 1. Comparison of Free Energy Barriers of the Several Segments of Different Pathways and the Global activation Free Energy in kcal mol−1 free activation energy required in several segments in each pathway pathway path-1 (Figure 1) (Pt-pentannulation) path-2 (Figure 1) (Pt-pentannulation) path-3 (Figure 1) (Pt-pentannulation) path-4 (Figure 2) (Pt-benzannulation) path-5 (Figure 3) (Au-pentannulation) path-6 (Figure 3) (Au-benzannulation) path-7 (Figure 3) (Au-pentannulation)

allene formation (af)

global activation free energy

relative free energy of the highest transition structure

ring formation (rf)

acyl migration (am)

17.42

22.47

38.61

38.61

36.24, 1-TS4

17.42

22.47

30.36, 2-I8−2-TS8

30.36, 2-I8−2-TS8

19.28, 1-TS3

17.42

5.01, 3-I8−3-TS8

33.89, 3-I6−3-TS7

35.34, 1-I3−3-TS7

32.15, 3-TS7

17.42

50.08, 1-I3−4-TS4

50.08

46.89, 4-TS4

14.66, 5-I1−5-TS1

22.09, 5-I5−5-TS5

22.09, 5-I5−5-TS5

23.16, 5-TS5

14.66

15.11, 5-I5−6-TS5

16.18, 5-I1−6-TS5

16.18, 6-TS5

14.66

32.09 (5-I3−7-TS3), acyloxy migration

37.39, 5-I1−7-TS3

37.39, 7-TS3

17.15, 5-I6−5-TS6

Figure 2. Structures of the stationary points on pathway-2 showing the migration of the acyl group.

required in such process is 22.47 kcal mol−1 (Figure 1 and Table 1). This step is followed by a direct acyl migration (through 1-TS4, magenta color) from the allenyl oxygen to the newly formed exocyclic double bond (see also the blue arrow in pathway-1 of Scheme 2) with an activation barrier of 38.61 kcal mol−1. The comparison of the energy requirement, as summarized in Table 1, clearly suggests that the highest barrier involved in pathway-1 is the acyl migration process (38.61 kcal mol−1) and acts as the rate-determining step. In pathway-2, the allene intermediate 1-I3 follows a cyclization step similar to that in pathway-1 to generate intermediate 1-I4. However, the cyclized intermediate then undergoes a geometrical rearrangement of the metal-complex structure (blue color path) by rearranging the ligands with respect to the metal center that places the two chlorine atoms in the opposite faces of the molecular plane bound to the platinum ion in between them (see Figure 2, structure 2-I6). Such configurational alteration of ligands starts from intermediate 1-I4 and passes through 2-TS4, 2-I5, and 2-TS5 successively to give a bipyramidal structure of the metal complex (2-I6), with one coordination site of the metal atom remaining vacant. The overall change in the geometry results in the placement of one chlorine ligand at close proximity with the acetoxy group attached to the allene moiety. Next step involves a ligandassisted acyl migration, where the initial movement of the acyl group takes place from the allenic oxygen to the proximal chlorine ligand (2-I6 → 2-TS6 → 2-I7; free energy barrier: 21.98 kcal mol−1) to generate 2-I7 (see the blue colored arrow

in pathway-2 of Scheme 2). Our search for the final movement of the acyl group results in two more transition structures (2TS7 and 2-TS8) through which the acyl group first jumps to a carbon atom of the newly formed five-membered ring (2-I8) (see the green colored arrow in pathway-2 of Scheme 2) followed by a second migration to the exocyclic carbon to generate the intermediate 2-I9 (see the red arrow in pathway-2 of Scheme 2). Final product 2-I10 is obtained by a conformational change of 2-I9. The comparison of activation energy requirements (Table 1) in various steps clearly suggests that the highest barrier results from the migration of the acyl group from one carbon of the pentacyclic ring to its final destination (2-I8 to 2-TS8, 30.36 kcal mol−1). The global activation barrier (30.36 kcal mol−1) as well as the relative energy of the highest transition structure (1-TS3, 19.28 kcal mol−1) necessarily indicates the favorability of this mechanism over pathway-1. Pathway-3 starts from the deviation of pathway-1 when the allene-bound Pt complex (1-I3) generates an η1 system (3-I4) from an η2 one (1-I3) by crossing the transition structure 3TS3. This makes a geometry, where one coordination site of Pt is occupied by a single allenic carbon atom. This arrangement leads to a facile movement of the acyl group from the allene oxygen to the metal center (3-I4 to 3-I5 through 3-TS4). In the first step of the acyl migration process, the migrating acyl group forms a bond with the metal atom, generating a square pyramidal geometry of the complex. The intermediate is then transformed into another one by the final acyl migration from a metal center to the acetylenic carbon atom. Finally, a 1162


Article

ACS Omega

Figure 3. Energy profile for the comparison of pentannulation and benzannulation under Pt-catalyzed conditions.

Scheme 3. General Scheme for the Study of the Reactions Pathways

Comparative Study of Pt-Catalyzed Pentannulation and Hexannulation. Figure 3 represents the studied energy profile for this hypothetical pathway to generate the fused naphthyl derivative (violet color). For easy comparison, the most favorable pathway for pentannulation (pathway-2) is also shown in blue color. The pathway for the formation of the naphthyl derivative originates from the diversion of pathway-2 at the intermediate 1-I3. Here, this intermediate undergoes a hypothetical 6-endodig attack through 4-TS3 to generate cyclic intermediate 4-I4. The energy of this intermediate is very close to the precursor transition structure 4-TS3, and the generated ring is not totally planar in nature. The Pt metal is positioned below the plane of the newly generated ring. In the next step, the PtCl2 moiety

cyclization step results in the formation of a pentannulated ring system as the end product. The overall activation barrier shows a high energy requirement (35.34 kcal mol−1, Table 1) for this process in comparison to that of pathway-2 and thus considered as an unfavorable one. In addition to the activation barrier of the fragmented steps, a comparison of global activation energy barriers associated with three different pathways is also shown in Table 1. It reveals clearly that pathway-2 requires a minimum energy of activation (30.36 kcal mol−1) and may be considered as a favorable pathway. With this pathway in hand, we have explored the hypothetical 6-endo-dig cyclization catalyzed by the Pt salt that would generate the naphthalene derivative. 1163


Article

ACS Omega

Figure 4. Potential energy surfaces PESs) for pathway-5, 6, and 7 under Au-catalyzed conditions.

moves close to one side of the ring, passing through 4-TS4. The intermediate 4-I5 (generated from 4-TS4) shows a tricoordinated Pt metal in which one coordination site of the metal is occupied by just one unsaturated portion of the newly generated ring. In the next step, the metal atom moves further and comes close to a single carbon atom of the newly formed ring, maintaining its orientation on the same plane of the aromatic system (4-I6). Comparison of the relative free energy (Table 1) reveals that the activation barrier required for generating the six-membered ring (4-TS3) is 18.79 kcal mol−1 higher than that for generating the five-membered ring (through 1-TS3). To stabilize this intermediate 4-I4, an additional 7.35 kcal mol−1 energy is needed in the next step (4-I4 → 4-TS4). However, the intermediate 1-I4 in path-2 does not require this stabilization energy. The necessity of this excess energy in pathway-4 makes it unfavorable with respect to pathway-2 and thus does not allow the formation of naphthyl moiety (hexannulation_), which is consistent with the experimentally observed fact.9 Comparative Study of the Au-Catalyzed Pentannulation and Hexannulation. After identifying two plausible pathways for Pt-catalyzed cycloisomerization for pentannulation and hexannulation, we have explored the Au-catalyzed cyclization for generating pentannulated and hexannulated rings. An overview of the studied pathways of Au-catalyzed annulations is shown in Scheme 3. As usual, the propargyl

moiety undergoes a common rearrangement of the acetyl group migration to form the metal-bound allene derivative. Diversion to several pathways occurs from this intermediate. In pathway-5 and 6, the gold ion migrates from the allene to alkyne moiety and follows pentannulated (green) and hexannulated (red) cyclization to generate benzofulvene and naphthalene derivatives, respectively. In another process (violet), the hexannulation was studied without migrating the metal center from the allene moiety and is designated here as pathway-7. The detailed PESs of all these pathways are shown in Figure 4. The pentannulation process of pathway-5 starts from the gold-bound diynol−acetate complex (5-I1) that undergoes an usual [3,3] sigmatropic rearrangement in two-step processes (through intermediate 5-I2 and transition states 5-TS1 and 5TS2). To proceed through pathway-5, the generated metalbound allene intermediate 5-I3 rearranges to change the coordination site of Au through two transition states (5-TS3 and 5-TS4). In between them, the intermediate 5-I4 has an unusual geometry where the Au atom appears to coordinate with three ligands. The intermediate 5-I5 undergoes a 5-exo-dig attack to generate the pentannulated product 5-I6 (activation barrier 22.09 kcal mol−1). The final acyl migration from the acetate oxygen to the gold-bound unsaturated system results in the formation of product 5-I7 (activation barrier 17.15 kcal mol−1). Comparison of the activation free energies of different steps (Table 1) clearly reveals that the cyclization process is the rate-determining step of the whole pathway. 1164


Article

ACS Omega

Figure 5. Intrinsic Reaction Coordinate (IRC) calculation for the annulation step of the allene intermediate under Pt (a,b)- and Au (c,d)-catalyzed conditions. The solid black line describes the changes in energy. The panels (a,c) represent the pentannulation path, whereas the panels (b,d) represent the hexannulation pathway. Change in NICS(0) values at different geometric centers are shown by various points.

which is 18.79 kcal mol−1 lower in free energy than the transition structure 4-TS3. On the other hand, under Aucatalyzed conditions, the intermediate 5-I5 follows the favorable 6-endo-dig cyclization through 6-TS5 with a favorable free energy difference of 6.98 kcal mol−1. To pinpoint the guiding factors, we further study the development of aromaticity of the generated complexes under two different metal-catalyzed conditions. Analysis of the Divergent Pathways of Annulations under Pt- and Au-Catalyzed Conditions. Apparently, the formation of the benzofulvene derivative through cyclization under platinum-catalyzed conditions bypasses the formation of, the so-called more aromatic, naphthalene derivative. To get a proper chemical insight on the favorability of formation of an apparently less-stabilized, nonaromatic pentacyclic ring structure, we analyzed the development of aromaticity by probing the nucleus-independent chemical shift (NICS) (0) value (see Computational Methods) at several ring centers on the changing structures along the reaction path of the annulation step under two different metal-catalyzed conditions. Our studied results have been summarized in Figure 5 in which the upper two panels, (a,b), represent the pentannulation and hexannulation pathways under platinum-catalyzed conditions, whereas the lower panels, (c,d), represent the results of

Pathway-5 may be diverted to pathway-6, when the intermediate 5-I5 undergoes 6-endo-dig attack to generate the hexannulated product 6-I6. The energy profile in Figure 4 reveals that this step requires 15.11 kcal mol−1 energy and controls the global activation barrier (16.18 kcal mol−1) required for the whole process. Comparison of the highest free energy barriers of pentannulation (pathway-5) reveals that the energy requirement for pathway-5 is about 7 kcal mol−1 higher than that required for pathway-6. Hence, the experimental report for the generation of a six-membered naphthyl derivative6 is totally consistent with our results for Aucatalyzed reactions. Energetically costly pathway-7 starts from intermediate 5-I3 and cyclizes to the naphthyl ring (7-I4) through 7-TS3. However, the possibility of this mechanism can be discarded because of the requirement of high energy (37.39 kcal mol−1) to cross the global activation barrier of the process. Comparative study of the pathways for generating five- and six-membered ring structures under Pt-catalyzed (pathways-2 and 4) and Au-catalyzed (pathways-5 and 6) conditions reveals that the pathways are diverged from two critical stationary points, 1-I3 and 5-I5. These intermediates may generate a fiveor six-membered ring structure through 5-exo-dig or 6-endo-dig cyclization. In the case of Pt-catalyzed conditions, the favorable pathway follows 5-exo-dig through the transition state 1-TS3, 1165


Article

ACS Omega the two respective annulations under gold-catalyzed conditions. As evident from the four figures, the NICS(0) values at the geometric center of the existing benzene ring (marked by blue diamond) show the least variation in all the four cyclization processes. Under platinum-catalyzed conditions (panels a and b in Figure 5), the center at the forming carbocyclic ring (marked by pink circle) shows a small decrease in NICS values, indicating the increase of aromaticity to a slight extent in both pentannulation and hexannulation processes. The center of the ring containing the Pt atom (marked by a black inverted triangle) shows an overall increase in the NICS value for pentannulation though an overall decrease is noted in the hexannulation pathway. The most notable change in the NICS value for pentannulation is observed at the ring junction of the carbocyclic and carboplatina ring (marked by brown square), where a sharp decrease in NICS values take place at the vicinity of the transition structure. When compared with the change in values for the hexannulation pathway, it appears that during the pentannulation process a strong ring current develops on the forming bicyclic ring-containing Pt metal. Absence of such effects in hexannulation clearly indicates that the pathway of pentannulation under Pt-catalyzed conditions generates a relatively higher aromatic character and dictates the favorability of the process. The geometric center between the existing and forming ring junction (marked by red triangle) shows a variation for the pentannulation under Au-catalyzed conditions (Figure 5c). However, a more decreased value in NICS is noted during the hexannulation pathway (Figure 5d). The lower values clearly correlate with our common sense for the development of an aromatic system while generating the naphthalene ring. The center of the forming ring (marked by pink circle) decreases more rapidly in Figure 5d than in Figure 5c, indicating the generation of more aromaticity, while forming the naphthyl ring structure under Au-catalyzed conditions. Hence, the overall development of the aromaticity in the Aucatalyzed system is higher for the hexannulation process, consistent with the experimentally observed fact.6 It may further be noted that the activation barrier required for the platinum-catalyzed cyclization is much higher than that required for the Au-catalyzed reaction (Table 1). The optimum conditions as reported in the literature also support the fact that the platinum-catalyzed reaction requires a relatively more drastic condition (80 °C, 12 h) with respect to that required for the gold-catalyzed reaction (rt, 11 h).6,9 Close inspection on the molecular orbitals of the stationary point reveals that the πorbitals associated with the alkyne moiety may fall under two categories which are orthogonal to each other (Figure 6). One set is perpendicular to the plane of the ring, whereas the other set remains parallel to the ring. For the convenience of our discussion, the π-orbitals of the alkyne π-system that are coplanar or nearly coplanar to the aromatic ring are termed as periplanar π-orbitals, whereas those perpendicular or nearly perpendicular to the aromatic ring are designated as clinal πorbitals. It is evident from the figure that the periplanar π-orbitals associated with the alkyne and allene moiety are suitable to generate σ-bonds for their correct orientation to overlap with the allene orbitals. The periplanar π-orbitals of the alkyne moiety that is proximal to the aromatic ring, when overlaps with the periplanar π-orbitals of the central carbon atom of the allene system, result in pentannulation through 5-exo-dig attack (path a), whereas the corresponding distal p-orbital of alkyne

Figure 6. Orientation of the two orthogonal sets of the π-system in the allene and alkyne moiety of enyne−allene.

gives the benzannulated product through 6-endo-dig attack (path b). Analysis of the molecular orbitals associated with the intermediate, transition state, and product of the cyclization steps reveals that the Au-catalyzed cyclization progresses through the activation of the periplanar π-orbitals of the alkyne moiety while forming the transition structure. This results in the reduction of the activation barrier. The unsaturated clinal πorbitals get stabilized by forming a part of the conjugated aromatic π-orbital of the generated ring system (Figure 7).

Figure 7. Orbital activation through the coordination of the Au ion.

However, the molecular orbitals associated with the stationary points on Pt-catalyzed reactions reveal some different pictures. The geometry of 1-I3 reveals that the Pt metal forms a chelate complex by coordinating with two unsaturated πorbitals of the alkyne and allene system that are formed from the clinal π-orbitals. While adopting the transition structure geometry (5-exo-dig or 6-endo-dig), the periplanar π-orbitals are not directly activated by the metal center and the forming σbonds leaves an unshared orbital at the end of the process (Figure 8). While forming the six-membered ring system, the two unsaturated π-orbitals that are already coordinated to the metal ion would try to get stabilized by involving themselves in generating π-orbitals of the newly generated aromatic ring. Obviously, the process of cyclization should weaken the coordinate bond between the unsaturation and the metal ion, and thus, the square planar geometry of the complex is effectively destabilized. This effect is reflected in the high activation barrier in the energy profile diagram (4-TS3 in Figure 3), and the cyclized intermediate (4-I4 in Figure 3) has an energy very close to those of the transition structures that generate it. 1166


Article

ACS Omega

set for other atoms.19 Effect of the solvent has been understood with the polarizable continuum model20 calculation, and this has been done on the single point geometry of the gas phase optimized structure. ortho-Xylene has been used for Ptcatalyzed reactions, and dichloromethane is used for Aucatalyzed reactions. Solvation free energies were calculated by adding the solvation energies to the computed gas phase relative free energies (ΔG353 for Pt-catalyzed cyclizations and ΔG298 for Au-catalyzed cyclizations, where the superscripts indicate the temperature in kelvin). To test the developed aromatic character of the intermediate formed during the cyclization steps under two different metal-catalyzed conditions, the variation of the nucleus-independent chemical shift (NICS(0)) was examined along the reaction pathways.21 The NICS(0) method, introduced by Schleyer, is probably the most common and convenient way to measure the aromaticity. Literature data show their relevance and success for characterizing the aromatic behaviors of a variety of chemical structures of intermediates and transition states and act as a tool to study the developing aromaticity in a process. In the present paper, all NICS(0) values were calculated for transition structures and species on the IRC paths at the PBE0 level. While computing the NICS, we placed a ghost atom in the position where we would like to compute the NICS. The output result is then obtained using the NMR keyword in the Gaussian input file. The isotropic magnetic shielding value at the ghost atom, when multiplied by −1, gives us the NICS value. Because of numerous ambiguities in selecting reference points for calculating NICS values, we have measured this parameter at several single and fused ring centers of the changing structures and analyzed the concerned problem by selecting the one that gives interpretable result.

Figure 8. Orientation of the Pt metal while occurring in 6-endo-dig and 5-exo-trig attack of the enyne−allene system.

■

CONCLUSIONS In summary, we have made a comparative study of the pathways of pentannulation and hexannulation of the aromatic enediyne system, under platinum(II) and gold(I) catalytic conditions. In agreement with the reported experimental results, our study reveals that the platinum-catalyzed pentannulation is a more favorable process than the hexannulation pathway, whereas the gold-catalyzed reaction favors hexannulation over pentannulation. In both the cases, the guiding factor has been elucidated to be the increasing aromaticity. The aromatic character of the naphthyl derivative, which is generated by the hexannulation process in goldcatalyzed conditions, though, is quite apparent, such behavior of the intermediate generated in platinum-catalyzed conditions is quite difficult to recognize. Monitoring the change of NICS(0) values with the change in the structure during the pentannulation process under platinum-catalyzed conditions reveals that the metal complex of the pentannulated ring gains more aromaticity than the structure generated by the hexannulation pathway. Besides this, our study also reveals that the acyl migration after pentannulation in platinumcatalyzed conditions takes place through a favorable multistep process, in which the assistance of the ligand of the metal is also involved. This study reveals several intricate details of the mechanism of the annulation reactions and also increases the chemical insight on the studied pathway.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.7b01889. PES of all pathways, Cartesian coordinates, absolute energies, frequencies, and detailed thermodynamic parameters for all the computed stationary points along with the structure and ball and stick model (PDF)

■

COMPUTATIONAL METHODS For the study of all pathways and analysis of the properties of the individual species, we have used the density functional theory quantum mechanical method implemented in the Gaussian suite.12 While choosing the proper functional to carry out the search for stationary points, we were motivated by our previous success of the study on other metal-catalyzed reactions.13 Our previous experience shows that the hybrid density functional PBE014 gave consistent results for Aucatalyzed reactions, and its utility had already been endorsed by other reports. For selecting a favorable pathway, we presently use the same functional (equivalent to PBE1PBE) and LANL2DZ15 basis set for metal atoms. The other atoms are treated with the 6-31G(d,p) basis set.16 Optimizations and energy calculations were done by considering the singlet multiplicity for each structure, and the restricted method was used. The vibrational frequency calculation of the optimized stationary points reveals whether the species is a minimum or transition structure on the PES of the system. Some transition structures were further confirmed with IRC calculation to relate the reactant and the product. Further refinement of the energy barriers were made by single point energy calculations on the optimized structures using the M06-2X17,18 functional and def2-QZVP basis set for metal atoms and the 6-311+G** basis

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (G.K.D.). ORCID

Gourab Kanti Das: 0000-0002-1235-3298 Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are thankful to the DST, New Delhi, India for providing financial assistance to one of our author (K.C.) in the form of the research project (SR/WOS-A/CS-153/2016). A.C. thanks UGC for providing research fellowship. We are also thankful to Visva-Bharati for providing us the necessary infrastructure facility to perform the research work. 1167


Article

ACS Omega

■

(h) Nun, P.; Gaillard, S.; Poater, A.; Cavallo, L.; Nolan, S. P. A combined mechanistic and computational study of the gold(I)catalyzed formation of substituted indenes. Org. Biomol. Chem. 2011, 9, 101−104. See the following review for other examples (i) Gorin, D. J.; Sherry, B. D.; Toste, F. D. Ligand Effects in Homogeneous Au Catalysis. Chem. Rev. 2008, 108, 3351−3378. (4) For Pt-catalyzed reaction, see (a) Pujanauski, B. G.; Prasad, B. A. B.; Sarpong, R. Pt-catalyzed tandem epoxide fragmentation/ pentannulation of propargylic Esters. J. Am. Chem. Soc. 2006, 128, 6786−6787. (b) González-Pérez, A. B.; Vaz, B.; Faza, O. N.; de Lera, Á . R. Mechanism and stereochemical on the Pt-catalysed Rearrangement of Oxiranyl-Propargylic Esters to Cyclopentenones. J. Org. Chem. 2012, 77, 8733−8743. For Au-catalyzed reaction see (c) Cordonnier, M.-C.; Blanc, A.; Pale, P. Gold(I)-catalyzed rearrangement of alkynyloxiranes: A mild access to divinyl ketones. Org. Lett. 2008, 10, 1569−1572. For a theoretical study see (d) González Pérez, A.; López, C. S.; Marco-Contelles, J.; Faza, O. N.; Soriano, E.; de Lera, A. R. Mechanism of the Gold-catalyzed rearrangement of (3-Acyloxyprop-1-ynyl) oxiranes: A dual role of the catalyst. J. Org. Chem. 2009, 74, 2982−2991. (5) For Pt-catalyzed reaction, see (a) Motamed, M.; Bunnelle, E. M.; Singaram, S. W.; Sarpong, R. Pt(II)-catalyzed synthesis of 1,2dihydropyridines from aziridinyl propargylic esters. Org. Lett. 2007, 9, 2167−2170. (b) Smith, C. R.; Bunnelle, E. M.; Rhodes, A. J.; Sarpong, R. Pt-catalyzed cyclisation/1,2-migration for the synthesis of indolizines, pyrrolones, and indolizinones. Org. Lett. 2007, 9, 1169− 1171. (c) Motamed, M.; Bunnelle, E. M.; Singaram, S. W.; Sarpong, R. Pt(II)-catalyzed synthesis of 1,2-dihydropyridines from aziridinyl propargylic esters. Org. Lett. 2007, 9, 2167−2170. For Au-catalyzed reaction, see (d) Xia, Y.; Dudnik, A. S.; Li, Y.; Gevorgyan, V. On the validity of Au-vinylidened in the Gold-catalyzed 1,2-migratory cycloisimerization of Skipped prpargylpyridines. Org. Lett. 2010, 12, 5538−5541. (e) Kern, N.; Blanc, A.; Weibel, J.-M.; Pale, P. Gold(I)catalyzed rearrangement of aryl alkynylaziridines to spiro [isochroman4,2′-pyrrolines]. Chem. Commun. 2011, 47, 6665−6667. (6) Zhao, J.; Hughes, C. O.; Toste, F. D. Synthesis of aromatic ketones by a transition metal-catalyzed tandem sequence. J. Am. Chem. Soc. 2006, 128, 7436−7437. (7) (a) Myers, A. G.; Harrington, P. M.; Kwon, B. M. Evidence for aminoglycoside participation in thiol activation of neocarzinostatin chromophore. Synthesis and reactivity of the epoxy dienediyne core. J. Am. Chem. Soc. 1992, 114, 1086−1087. (b) Sugiyama, H.; Fujiwara, T.; Kawabata, H.; Yoda, N.; Hirayama, N.; Saito, I. Chemistry of neocarzinostatin-mediated cleavage of oligonucleotides. Competitive ribose C5’ and C4’ hydroxylation. J. Am. Chem. Soc. 1992, 114, 5573− 5578. By products obtained by mediation of Mo-mediated carbonylation of allenyl arene may also be considered as the result of Myers−Saito rearrangement, see (c) Datta, S.; Liu, R.-S. Molybdenum-mediated cyclocarbonylation of 1-ethyl-2-allenylbenzenes to 1H-cyclopenta[a]inden-2-ones. Tetrahedron Lett. 2005, 46, 7985−7988. (8) Sun, J.; Conley, P. M.; Zhang, L.; Kozmin, S. A. Au- and Ptcatalyzed cycloisomerizations of 1,5-enynes to cyclohexadienes with a broad alkyne scope. J. Am. Chem. Soc. 2006, 128, 9705−9710. (9) Chen, Z.; Jia, X.; Huang, J.; Yuan, J. Platinum-Catalyzed Tandem cycloisomerization reaction of benzoendiynyl esters: Regioselective long-range 1,5 acyl migration. J. Org. Chem. 2014, 79, 10674−10681. (10) Bekele, T.; Christian, C. F.; Lipton, M. A.; Singleton, D. A. “Concerted” transition state, stepwise mechanism. Dynamics effects in C2-C6 enyne allene cyclizations. J. Am. Chem. Soc. 2005, 127, 9216− 9223. (11) Ghosh, A.; Basak, A.; Chakrabarty, K.; Ghosh, B.; Das, G. K. Ligand assisted acyl migration in Au catalyzed isomerization of propargylic ester to diketone: A DFT study. J. Org. Chem. 2014, 79, 5652−5663. (12) Frisch, M. J.; et al. Gaussian 09, Revision C.01; Gaussian Inc.: Wallingford CT, 2010 (Full reference is given in Supporting Information).

REFERENCES

(1) (a) Marion, N.; Nolan, S. P. Propargylic esters in Gold catalysis: Access to diversity. Angew. Chem., Int. Ed. 2007, 46, 2750−2752. (b) Peng, Y.; Cui, L.; Zhang, G.; Zhang, L. Gold-catalyzed homogeneneous oxidative C−O bond formation: Efficient synthesis of 1-benzoxyvinyl ketones. J. Am. Chem. Soc. 2009, 131, 5062−5063. (c) Cran, J. W.; Krafft, M. E. Regioselective cyclizations utilizing a Gold catalyzed [3,3] propargyl ester rearrangement. Angew Chem., Int. Ed. 2012, 51, 9398−9402. (d) Zhang, L. Tandem Au-catalyzed 3,3rearrangement-[2+2] cycloadditions of propargylic esters: Expeditious access to highly functionalized 2,3-indoline-fused cyclobutanes. J. Am. Chem. Soc. 2005, 127, 16804−16805. (e) Correa, A.; Marion, N.; Fensterbank, L.; Malacria, M.; Nolan, S. P.; Cavallo, L. Golden carousel in catalysis: The cationic gold/propargylic ester cycle. Angew. Chem. 2008, 120, 730−733. (f) Shiroodi, R. K.; Gevorgyan, V. Metalcatalyzed double migratory cascade reactions of propargylic esters and phosphates. Chem. Soc. Rev. 2013, 42, 4991−5001. (g) Rettenmeier, E.; Schuster, A. M.; Rudolph, M.; Rominger, F.; Gade, C. A.; Hashmi, A. S. K. Gold catalysis: Highly functionalized cyclopentadienes prepared by intermolecular cyclization of ynamides and propargylic carboxylates. Angew. Chem., Int. Ed. 2013, 52, 5880−5884. (h) Fensterbank, L.; Malacria, M. Molecular complexity from polyunsaturated substrates: The gold catalysis approach. Acc. Chem. Res. 2014, 47, 953−965. . See also the book chapters of Computational Mechanisms of Au and Pt Catalyzed Reactions; Soriano, E., MarcoContelles, J. Eds.; Springer: Berlin, Heidelberg, 2011; Vol. 302. (2) For 1,2-acyloxy migration some recent publications are (a) Shu, X.-Z.; Li, X.; Shu, D.; Huang, S.; Schienebeck, C. M.; Zhou, X.; Robichaux, P. J.; Tang, W. Rhodium-catalyzed intra- and intermolecular [5 + 2] cycloaddition of 3-acyloxy-1,4-enyne and alkyne with concomitant 1,2-acyloxy migration. J. Am. Chem. Soc. 2012, 134, 5211−5221. (b) Zheng, H.; Zheng, J.; Yu, B.; Chen, Q.; Wang, X.; He, Y.; Yang, Z.; She, X. Pt-catalyzed tandem 1,2-acyloxy migration/ intramolecular cycloaddition of enynyl esters. J. Am. Chem. Soc. 2010, 132, 1788−1789. (c) Lauterbach, T.; Ganschow, M.; Hussong, M. W.; Rudolph, M.; Rominger, F.; Hashmi, A. S. K. Gold-catalyzed synthesis of 1-Naphthylcarbenoids and their synthetic utilization in cyclopropanation reactions. Adv. Synth. Catal. 2014, 356, 680−686. For 1,3-acyloxy migration some recent articles are (d) Lebœuf, D.; Simonneau, A.; Aubert, C.; Malacria, M.; Gandon, V.; Fensterbank, L. Gold-catalyzed 1,3-acyloxy migration/5-exo-dig cyclization/1,5-acyl migration of diynyl esters. Angew. Chem., Int. Ed. 2011, 50, 6868− 6871. (e) Shu, D.; Li, X.; Zhang, M.; Robichaux, P. J.; Guzei, I. A.; Tang, W. Rhodium-catalyzed carbonylation of cyclopropyl substituted propargyl esters: A Tandem 1,3-acyloxy migration [5+1] cycloaddition. J. Org. Chem. 2012, 77, 6463−6472. (f) Wang, L.-J.; Zhu, H.T.; Wang, A.-Q.; Qiu, Y.-F.; Liu, X.-Y.; Liang, Y.-M. Gold catalyzed tandem [3,3]-propargyl ester rearrangement leading to (E)-1H-Inden1-ones. J. Org. Chem. 2014, 79, 204−212. (3) For Pt and Au catalyzed cyclization see (a) Soriano, E.; MarcoContelles, J. Mechanistic insights on the cycloisomerization of polyunsaturated precursors catalyzed by platinum and gold complexes. Acc. Chem. Res. 2009, 42, 1026−1036. (b) Zhang, L.; Sun, J.; Kozmin, S. A. Gold and platinum catalysis of enyene cycloisomerization. Adv. Synth. Catal. 2006, 348, 2271. (c) Aubert, C.; Buisine, O.; Malacria, M. The behavior of 1,n-enyenes in the presence of transition metals. Chem. Rev. 2002, 102, 813. For Pt-catalyzed reaction, see (d) Prasad, B. A. B.; Yoshimoto, F. K.; Sarpong, R. Pt-Catalyzed pentannulations from in situ generated metallo-carbenoids utilizing propargylic esters. J. Am. Chem. Soc. 2005, 127, 12468−12469. (e) Zhao, J.; Clark, D. A. Regiodivergent synthesis of functionalized Indene derivatives via Ptcatalyzed rautenstrauch reaction of propargyl carbonates. Org. Lett. 2012, 14, 1668−1671. For Au-catalyzed cyclization, see (f) Marion, N.; de Frémont, P.; Lemière, G.; Stevens, E. D.; Fensterbank, L.; Malacria, M.; Nolan, S. P. AuI-catalyzed cycloisomerization of 1,5enynes bearing a propargylic acetate: formation of unexpected bicycle [3.1.0] hexene. Chem. Commun. 2006, 2048−2050. (g) JiménezNúñez, E.; Echavarren, A. M. Gold catalyzed cycloisomerizations of enynes: A mechanistic perspective. Chem. Rev. 2008, 108, 3326−3350. 1168


Article

ACS Omega (13) Basak, A.; Chakrabarty, K.; Ghosh, A.; Das, G. K. Mechanism of the Gold (III) catalyzed isomerization of substituted allenes to conjugated dienes: A DFT Study. J. Org. Chem. 2013, 78, 9715−9724. (14) Kang, R.; Chen, H.; Shaik, S.; Yao, J. Assessment of theoretical methods for complexes of gold(1) and gold(III) with unsaturated aliphatic hydrocarbon: which density functional should we choose? J. Chem. Theory Comput. 2011, 7, 4002−4011. (15) (a) Hay, P. J.; Wadt, W. R. Ab initio effective core potentials for molecular calculations. Potential for the transition metal atoms Sc to Hg. J. Chem. Phys. 1985, 82, 270−283. (b) Wadt, W. R.; Hay, P. J. Ab initio effective core potentials for molecular calculations. Potential for main group elements Na to Bi. J. Chem. Phys. 1985, 82, 284−298. (c) Hay, P. J.; Wadt, W. R. Ab initio effective core potentials for molecular calculations. Potentials for K to Au including the outermost core orbitals. J. Chem. Phys. 1985, 82, 299−310. (16) (a) Ditchfield, R.; Hehre, W. J.; Pople, J. A. Self-consistent molecular-orbital methods. IX. An extended Gaussian-type basis for molecular-orbital studies of organic molecules. J. Chem. Phys. 1971, 54, 724−728. (b) Hehre, W. J.; Ditchfield, R.; Pople, J. A. Self-consistent molecular-orbital methods. XII. Further extensions of Gaussian-type basis for use in molecular orbital studies of organic molecules. J. Chem. Phys. 1972, 56, 2257−2261. (c) Hariharan, P. C.; Pople, J. A. The influence of polarization functions on molecular orbital hydrogenation energies. Theor. Chim. Acta 1973, 28, 213−222. (d) Hariharan, P. C.; Pople, J. A. Accuracy of AHn equilibrium geometries by single determinant molecular orbital theory. Mol. Phys. 1974, 27, 209−214. (e) Dill, J. D.; Pople, J. A. Self-consistent molecular-orbital methods. XV. Extented Gaussian-type basis sets for lithium, beryllium, and boron. J. Chem. Phys. 1975, 62, 2921−2923. (f) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon, M. S.; DeFrees, D. J.; Pople, J. A. Self-consistent molecular orbital methods. XXII. A polarization type basis set for second row elements. J. Chem. Phys. 1982, 77, 3654− 3665. (g) Rassolov, V. A.; Pople, J. A.; Ratner, M. A.; Windus, T. L. 631G* basis set for atoms K through Zn. J. Chem. Phys. 1998, 109, 1223−1229. (17) (a) Latouche, C.; Skouteris, D.; Palazzetti, F.; Barone, V. TDDFT Benchmark on Inorganic Pt(II) and Ir(III) Complexes. J. Chem. Theory Comput. 2015, 11, 3281−3289. (b) Walker, M.; Harvey, A. J. A.; Sen, A.; Dessent, C. E. H. Performance of M06, M06-2X, and M06HF Density Functionals for Conformationally Flexible Anionic Clusters: M06 Functionals Perform Better than B3LYP for a Model System with Dispersion and Ionic Hydrogen-Bonding Interactions. J. Phys. Chem. A 2013, 117, 12590−12600. (18) Zhao, Y.; Truhlar, D. G. Density Functionals with Broad Applicability in Chemistry. Acc. Chem. Res. 2008, 41, 157−167. (19) Ciancaleoni, G.; Rampino, S.; Zuccaccia, D.; Tarantelli, F.; Belanzoni, P.; Belpassi, L. An ab initio benchmark and DFT validation study on gold(I) catalyzed hydroamination of Alkynes. J. Chem. Theory Comput. 2014, 10, 1021−1034. (20) Tomasi, J.; Persico, M. Molecular Interactions in solution: an overview of methods based on continuous distributions of the solvent. Chem. Rev. 1994, 94, 2027−2094. (21) (a) Jiao, H.; von Ragué Schleyer, P. A detailed theoretical analysis of the 1,7-sigmatropic hydrogen shift: The Möbius character of the eight-electron transition structure. Angew. Chem., Int. Ed. Engl. 1993, 32, 1763−1765. (b) Herges, R.; Jiao, H.; von Ragué Schleyer, P. Magnetic properties of aromatic transition states: The Diels-Alder reactions. Angew. Chem., Int. Ed. Engl. 1994, 33, 1376−1378. (c) Twyman, L. J.; Beezer, A. E.; Mitchell, J. C. An approach for the rapid synthesis of moderately sized dendritic macromolecules. J. Chem. Soc., Perkin Trans. 1 1994, 407−411. (d) Jiao, H.; von Ragué Schleyer, P. Introductory lecture. Electrostatic acceleration of the 1,5-H shifts in cyclopentadiene and in penta 1,3-diene by Li+-complexation: aromaticity of the transition structures. J. Chem. Soc., Faraday Trans. 1994, 90, 1559−1567. (e) Jiao, H.; von Ragué Schleyer, P. Electroststic acceleration of electrolytic reactions by metal cation complexation: The cyclization of 1,3-cis-5-Hexatriene into 1,3cyclohexadiene and the 1,5-Hydrogen shift in cyclopentadiene. The aromaticity of the transition structures. J. Am. Chem. Soc. 1995, 117,

11529−11535. (f) Jiao, H.; von Ragué Schleyer, P. The Cope rearrangement transition structure is not diradicaloid, but is it aromatic? Angew. Chem., Int. Ed. Engl. 1995, 34, 334−337.

1169


Au-Catalyzed Hexannulation and Pt-Catalyzed Pentannulation of

Recommend Documents