Mechanistic Insights Into the Transition‐Metal‐Catalyzed Dual Reactivity Mode of Vinyl Ethylene Carbonates With Anilines

ABSTRACT

A study into a catalyst‐controlled chemodivergent strategy to access 2‐vinyl indolines and indole‐2‐acetates is presented herein. This strategy overcomes the usual allylic reactivity of vinyl ethylene carbonates (VECs) via a tandem C─H/C─O bond activation sequence. The methodology provides useful molecular scaffolds by following two distinct pathways in a highly step‐ and atom‐economical manner and features a challenging β‐hydroxide elimination. In the Rh‐catalyzed transformation, unusual competing pathways were observed, such as a regioselective insertion of a Rh−OMe species into the “C═C” bond of the vinyl segment, a regioselective C─H activation of the vinyl indoline, and an interesting oxidative addition of MeOH into Rh(I). Detailed mechanistic insights through a combination of experimental and computational studies unveil a unique reactivity mode of VECs, thereby opening up a new avenue for transition metal catalysis. Synthetic utility of the annulated products demonstrates an additional advantage of the methodology.

Detailed mechanistic insights through a combination of experimental and computational studies have been carried out, which reveal an unusual dual reactivity mode of vinyl ethylene carbonates (VECs) during coupling with a wide array of anilines in the presence of mono‐N‐protected amino acid (MPAA) ligands. This protocol typically overcomes the usual allylic reactivity of vinyl ethylene carbonates (VECs) via a tandem C─H/C─O bond activation sequence and offers synthetically useful 2‐vinyl indolines and indole‐2‐acetates and their useful post‐synthetic precursors in a highly step‐ and atom‐economical manner.

1. Introduction

The challenge of synthesizing useful molecular scaffolds from simple and readily available starting materials has resulted in transition metal‐catalyzed C─H functionalization developing into a powerful synthetic tool over the last two decades [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11]. In this context, the development of novel and methodical strategies, including those that are catalyst‐controlled [12, 13], directing group (DG)‐enabled [14, 15], or by manipulating the reaction parameters [16, 17] to modify either chemo‐, regio‐, or stereoselectivity, has become increasingly better understood in recent times. Consequently, a coalescence of DG‐assisted C(sp ²)−H allylation and intramolecular annulation comes across as one of the promising solutions for the direct access to privileged structural motifs [18, 19].

As a smart organic synthon, cyclic carbonates (bearing a native functional group) have attracted increasing attention in several catalytic transformations via strain‐release‐driven decarboxylative ring‐opening processes [20, 21, 22, 23, 24, 25, 26]. Among cyclic carbonates, the reactivity of vinyl ethylene carbonates (VECs) has been well studied, leading to the development of various cycloaddition [27, 28] and allylic substitution reactions [29, 30, 31]. The dual reactivity mode of VECs can be envisioned via decarboxylative ring opening transformations as triggered by an external promoter, leading to the generation of: (i) a diene synthon and/or (ii) an α‐allyl ester synthon, which subsequently results in the formation of the coupling product (Scheme 1a). Notably, the role of VECs as a diene equivalent in catalytic reactions remains less explored, whereas their reactivity mode as an α‐allyl ester equivalent remains elusive to date (Scheme 1a). This unusual reactivity mode of VECs has garnered considerable attention from synthetic organic chemists, and these have been utilized as a reactive coupling partner in catalytic organic transformations, leading to the formation of important molecular frameworks, including heterocycles (Scheme 1b). In addition, VECs have been demonstrated to act as a transmuted allyl source to enable TM‐catalyzed C─H allylation of arenes through β‐oxygen elimination, leading to the formation of linear allylated arenes II (Scheme 1c, Step‐I) [32, 33, 34, 35, 36].

SCHEME 1.

Divergent reactivity profile of vinyl ethylene carbonates with anilines via tandem C─H/C─O activation sequence.

Despite impressive research in the area of VECs, accessing annulated products in a one‐pot fashion by employing a single catalytic system is quite challenging and remains rather underexplored (Scheme 1c). In this context, in 2014, Wang and coworkers reported a tandem catalytic system (Rh(III)/Pd(II)) that provided rapid access to vinyl‐substituted 5,6‐dihydropyridin‐2(1H)‐ones and 3,4‐dihydroisoquinolin‐1(2H)‐ones while employing VECs (Scheme 1c) [37]. The alkene effect was found to be the key factor for the success of the reaction. In 2017, Ackermann and coworkers used VECs to achieve a domino C─H/N─H allylation to access vinylated heterocycles under Co(III)‐catalysis (Scheme 1c) [38].

There are critical challenges that need to be considered during the coupling of VECs with arenes, After the formation of the crucial metallacyclic intermediate I via the migratory insertion and β‐oxygen elimination sequence (Scheme 1d, a), a variety of pathways could result: (i) It may result in the formation of linear allylated arene in which the ‐OH group remains intact after protodemetallation (Scheme 1d, b); (ii) Re‐insertion of the metal into the “C─C” bond of the allyl arene followed by protodemetallation can lead to the annulated product with the terminal ‐OH group (Scheme 1d, c); (iii) Intermediate I may also lead to the formation of an annulated product containing the isolated vinyl group after metal re‐insertion and β‐hydroxide elimination sequence thereafter, bypassing the protodemetalation step, and this is synthetically challenging (Scheme 1d, d); (iv) The resulting annulated product with the vinyl substituent may also react further leading to complex molecular scaffolds and this demands proper designing of substrates or reaction conditions and remains rather underexplored till date (Scheme 1d, e).

In this study, we present a catalyst‐controlled divergent reactivity profile of vinyl ethylene carbonates (VECs) with anilines [39, 40, 41, 42] leading to the formation of synthetically challenging 2‐vinyl indolines [43, 44, 45, 46, 47] under a Ru(II)/mono‐N‐protected amino acid (MPAA) catalytic system. In this case, the VEC acts as a diene equivalent [48], involving a β‐hydroxide elimination as the key step (Scheme 1e). On the other hand, under Rh(III)/MPAA catalysis, the VEC in combination with MeOH was found to act as an α‐allyl ester equivalent [49, 50, 51] to deliver indole‐2‐acetates [9, 52, 53, 54, 55] (Scheme 1e). In this case, a unique reactivity of VEC was observed, extending beyond the β‐hydroxide elimination step.

2. Results and Discussion

2.1. Optimization Studies

Our investigations were initiated by reacting N‐pyrimidyl aniline 1 and 4‐vinyl‐1,3‐dioxolan‐2‐one 2 under Ru(II) catalysis (Table 1). To our delight, the desired target compound 2‐vinyl indoline 3 was achieved in a chemoselective fashion. Encouraged by this outcome, further optimization led to the desired product with improved yields and after screening several reaction conditions, we then concluded that the optimum condition for the best yields of the corresponding 2‐vinyl indoline 3 was: [Ru(p‐cym)Cl₂]₂ (5 mol%), AgSbF₆ (10 mol%), N‐Ac‐Gly‐OH (20 mol%), and PivOH (1.5 equiv.) in MeOH at 80°C for 24 h (Table 1) (for a detailed optimization study, see Table S1). Replacing the Ru‐catalyst with the Rh‐catalyst hampered the desired transformation (Table 1, entry 2). Using AgNTf₂ instead of AgSbF₆ as the additive resulted in a reduced yield of the desired product (Table 1, entry 3). Notably, the efficiency of the reaction was significantly amplified by MPAA ligands (see Table S1 for details) and was optimal for N‐Ac‐Gly‐OH (Table 1, entry 4). MPAA ligands play a crucial role by acting as a proton acceptor in the C─H activation step and facilitating the C─H bond cleavage by forming MPAA‐bridged binuclear metal complexes [56, 57]. Several other additives were also tested, and it was found that PivOH was the best amongst the ones scanned (Table 1, entries 5–8). The desired product formation was not observed when the reaction was carried out in other polar aprotic solvents (Table 1, entry 9). Only a trace amount of product was detected when the temperature was reduced to 25°C or increased to 120°C (Table 1, entry 10). Reducing the catalyst loading resulted in a poor outcome (Table 1, entry 11).

TABLE 1.

Optimization of reaction conditions for Ru‐catalysis.

Entry	Deviation from standard conditions	Yield (%) a
1	None	66
2	[Cp*RhCl2]2 instead of [Ru(p‐cym)Cl2]2	Nd
3	AgNTf2 instead of AgSbF6	46
4	Piv‐Gly‐OH instead of N‐Ac‐Gly‐OH	Trace
5	NaOPiv.H2O instead of PivOH	Nd
6	Cu(OAc)2.H2O instead of PivOH	20
7	AcOH instead of PivOH	45
8	MesCO2H instead of PivOH	51
9	DCE or 1,4‐Dioxane instead of MeOH	Trace
10	25°C or 120°C instead of 80°C	Trace
11	2 mol% of [Ru(p‐cym)Cl2]2 instead of 5 mol%	40

^a Isolated yields.

nd = not detected.

In the Rh‐catalyzed transformation, we observed that the reactivity was not limited to the indoline formation step rather extended further to deliver indole‐2‐acetates 4 through a three‐component coupling (Table 2). The formation of the C2‐functionalized indole 4 was found to work best under the following conditions: [Cp*RhCl₂]₂ (2 mol %), AgSbF₆ (20 mol%), N‐Ac‐Gly‐OH (20 mol%), Cu(OAc)₂·H₂O (1.0 equiv.) in MeOH at 80°C for 24 h (Table 2, entry 1).

TABLE 2.

Optimization of reaction conditions for Rh‐catalysis.

Entry	Deviation from standard conditions	Yield (%) a
1	none	72
2	[Ru(p‐cym)Cl2]2 instead of [Cp*RhCl2]2	Nd
3	AgNTf2 instead of AgSbF6	52
4	without N‐Ac‐Gly‐OH	40
5	2.0 equiv. of Cu(OAc)2.H2O	59
6	without Cu(OAc)2.H2O	Nd
7	AcOH or PivOH instead of Cu(OAc)2.H2O	Nd
8	t BuOH/DCE instead of MeOH	Nd
9	25°C/120°C instead of 80°C	trace
10	5 mol% of [Cp*RhCl2]2 instead of 2 mol%	50

^a Isolated yields.

nd = not detected.

Here too, the reaction efficiency was sensitive toward the catalyst and the additives (Table 2, entries 2, 3). Notably, only a trace amount of indoline 3 was detected upon changing the catalyst from Rh to Ru (Table 2, entry 2). The productivity of the reaction was amplified by the MPAA ligand (Table 2, entry 4). The reaction efficiency was further improved by adding the Cu‐additive in a proper stoichiometry (Table 2, entries 5, 6). Other Brønsted acid additives resulted in poor outcomes (Table 2, entry 7). It should be noted that solvents other than MeOH did yield any product under the optimized conditions (Table 2, entry 8). This protocol was sensitive towards temperature, and it was found that 80°C was optimal for this reaction condition (Table 2, entry 9). The yield of the indole‐2‐acetate decreased upon increasing the loading of the catalyst (Table 2, entry 10) (for a detailed optimization study, see Table S2).

With the optimized conditions established, we then probed the scope of the reaction to explore the versatility of the transformation under Ru‐catalysis. The reaction was compatible with most of the anilines screened with good chemo‐ and regioselectivity and demonstrated good functional group tolerance. Various electronic and steric parameters were examined at the ortho‐, meta‐, and para‐positions of the aryl ring of N‐pyrimidyl anilines delivering the vinyl substituted indolines in moderate to good yields (3a−3s, Scheme 2a). Importantly, electron‐withdrawing functional groups (‐OCF₃, ‐CO₂Et, ‐SO₂Me) were well tolerated under this protocol and gave the corresponding products with excellent chemoselectivity (3f, 3g, 3k, Scheme 2a).

SCHEME 2.

Scope of 2‐vinyl indolines and indole‐2‐acetates.

Notably, a 2,4‐disubstituted aniline reacted best under the optimized reaction conditions to give 81% yield of the desired indoline (3a, Scheme 2a). The reaction was compatible with a variety of substituents, including halogens at the meta‐ and para‐positions of the aryl ring of N‐pyrimidyl anilines, yielding the desired indolines in good yields (3b−3e, 3h−3j, Scheme 2a).

The chemoselectivity was also retained in the case of polycyclic systems to afford the desired products in good yields (3l−3n, Scheme 2a). The structure of the indoline product was confirmed unambiguously by x‐ray crystallographic analysis of 3i (Scheme 2a). Importantly, the compatibility of the present transformation was further showcased by other directing groups to deliver the products with acceptable yields (3o‐‐3s, Scheme 2a). Notably, N‐phenyl acetamide remained unreacted under the standard reaction conditions (3t, Scheme 2a).

We next sought to probe the scope and generality of the Rh‐catalyzed transformation by testing a range of substituted anilines with varied electronic and steric properties at the aryl ring of the N‐pyridyl anilines (4a−4d, Scheme 2b). This protocol was compatible with halogens (‐Br, ‐Cl) at different positions of the aryl ring and provided ample opportunity for post‐synthetic modifications (4c, 4f, 4j, Scheme 2b). This methodology was also applied on a polycyclic system to yield the benzo‐fused indole‐2‐acetate in acceptable yields (4e, Scheme 2b). The reaction worked well with the assistance of pyrimidyl and acetyl directing groups too (4g−4l, Scheme 2b). The structure of the indole‐2‐acetate product was confirmed unambiguously by x‐ray crystallographic analysis of 4f (Scheme 2b). Notably, 4‐phenyl‐4‐vinyl‐1,3‐dioxolan‐2‐one 2’ was ineffective under the optimized reaction conditions for both the catalytic protocols.

2.2. Synthetic Utility

We then moved on to check the utility of the present transformation. Indolines (3) reacted with MeOH to afford indole‐2‐acetates (4) in moderate to good yields under oxidative conditions (4g, 4a, 4b, 4i, Scheme 3A). To further showcase the importance of the DG in the products, regioselective C7‐thiolation [58] and amidation [59] were carried out to deliver 6c and 8c, respectively, in good yields (Scheme 3B,C). Chemoselective reductions were also carried out by using the indoline 3c and indole‐2‐acetate 4a. Hydrogenation of 3c yielded 2‐ethyl indoline 9c in 78% yield (Scheme 3D–a). Synthesis of 2‐hydroxyethyl indole 10a was carried out by selectively reducing the ester group of 4a (Scheme 3D‐b). Moreover, to demonstrate the synthetic utility of the protocol, the pyridyl directing group was successfully removed from the N‐pyridyl indoline 3r using triethylsilane to deliver the oxidized product 2‐ethyl indole 13r in acceptable yields (Scheme 3E) [60, 61].

SCHEME 3.

Synthetic transformations.

2.3. Mechanistic Studies

Further, to gain insights into the pathway of the transformations, detailed mechanistic investigations were carried out (Scheme 4). To check the reversibility of C─H metalation under Ru‐catalysis, with added D₂O, in the absence of coupling partner 2, 75% D incorporation was observed at the ortho‐position of the recovered starting material (10% consumption of the SM, Scheme 4a‐i). When the study was carried out in the presence of the coupling partner 2, 70% deuteration was observed at the ortho‐position of the recovered starting material (Scheme 4a‐ii). A significant amount of deuteration was also observed (80% D incorporation without 2 and 40% D with 2) when we tested the reversibility of the C─H metalation step under the Rh‐catalysis conditions (Scheme 4b). A considerable amount of D to H exchange was also observed under both Ru‐catalysis (93% D to 17% D) (Scheme 4c) as well as [Rh]‐catalysis (93% D to 4% D) (Scheme 4d). Owing to the high reversibility of the C─H metallation, reliable results were not obtained when we performed studies to check for a KIE in the C−H activation step under both catalytic conditions. Additionally, in the conversion of 3 to 4 under Rh‐catalysis, the C─H metallation step was found to be mildly reversible, indicating that the subsequent steps were faster than the reverse step (Scheme 4e). To probe the origin of chemoselectivity in indoline and indole‐2‐acetate formation, control experiments were performed by using the pre‐formed indoline 3c (Scheme 5A). Deviation from the standard conditions resulted in a poor outcome. Further, to check the necessity of the MPAA ligand in the conversion of 3 to 4, the reaction was carried out in the absence of the MPAA ligand, and the reaction proceeded smoothly, which implies that the MPAA ligand has a significant role only up to the indoline formation and may not be involved in the formation of the indole‐2‐acetate 4 (Scheme 5A, entry 3).

SCHEME 4.

Mechanistic investigation‐I.

SCHEME 5.

Mechanistic Investigation‐II.

Notably, it was found that the Cu(OAc)₂.H₂O additive has a significant role in the indole‐2‐ acetate formation (Scheme 5A, entry 4). This may be due to the fact that the copper additive plays a key role in the oxidation step to give the final product 4. Polar protic solvents other than MeOH did not yield the desired product (Scheme 5A, entry 5). It is also important to note that the reactivity of VEC was restricted under Ru‐catalysis up to the indoline formation under the standard reaction conditions and did not proceed further to yield the indole‐2‐acetate (Scheme 5A, entry 6).

While a mechanism for the Ru‐catalyzed formation of the indoline could be predicted based on the mechanistic studies and literature precedent, it was the formation of the indole‐2‐acetates that intrigued us the most. We therefore moved on to check the mechanism of MeOH incorporation in the formation of indole‐2‐acetate 4. We reacted 3a with [Cp*RhCl₂]₂ (10 mol%), AgSbF₆ (20 mol%) in the presence of anhyd. NaOAc (1.0 equiv.) as an additive in CD₃OD (0.1 M) at 80°C and followed the reaction by ¹H NMR (Scheme 5B). The ¹H NMR spectrum of the reaction mixture showed a characteristic triplet in the upfield region at −8.71 ppm (J = 29 Hz), indicating a Cp*Rh‐dihydride complex with two equivalent hydride ligands, both coupling to the central ¹⁰³Rh‐nucleus (spin ½, ∼100% abundant) and hence producing a triplet pattern (1:2:1) [62, 63].

Additionally, an organometallic intermediate (cationic Rh(III)−OMe) was detected by ESI‐HRMS analysis (Scheme 5C, see the Supporting Information for further details). This result seems to indicate that the regioselective migratory insertion of Rh─OMe bond into the “C═C” bond of the vinyl segment may have taken place after the initial coordination with the pyrimidyl nitrogen [65]. Additionally, to confirm the role of the MPAA ligand in C─H activation under both Ru‐ and Rh‐catalysis, mass spectrometry studies were carried out (Scheme 5D‐i, ii). The generation of the Rh‐MPAA complex III was confirmed by reacting 1c with [Cp*RhCl₂]₂ (20 mol%), N‐Ac‐Gly‐OH (40 mol%) in the presence of AgSbF₆ (40 mol%) and Cu(OAc)₂.H₂O (3.0 equiv.) as an additive in MeOH (0.02 M) at 80°C and by following the reaction by HRMS analysis (Scheme 5D (i)). Similarly, the generation of Ru‐MPAA complex IV was also examined, and ESI‐HRMS indicated the formation of the Ru−MPAA species during the reaction (Scheme 5D (ii)). To further confirm the source of the ‐OMe group in the indole‐2‐acetate, reactions were carried out by using 1k and 3k under the standard [Rh]‐catalysis with CD₃OD (0.1 M) (Scheme 5E, 5F). Expectedly, we observed deuterium as well as ‐OCD₃ incorporation (>99% D) in the products 11k and 12k.

Further, to check for the involvement of single electron transfer pathways in the indole‐2‐acetate formation, two different reactions were carried out by using TEMPO as the radical quencher under the standard Rh‐catalysis conditions (Scheme 5G, 5H). The formation of indole‐2‐acetate 4g in good yields confirmed that the reaction did not proceed through a SET pathway in any of the intermediate steps of the catalytic cycle.

Next, to examine the participation of any [Rh−H] species in the catalytic cycle, a spike test was also performed where the NMR tube containing the reaction mixture was bubbled with H₂ gas. An enhanced signal for H₂ was observed at ∼δ 4.62 ppm, which further confirmed the release of H₂ gas in the reaction aliquot during the progress of the reaction (see the SI for details). This indicates that the Rh‐catalytic cycle may involve the generation of a transient [Rh−H] species as an intermediate.

2.4. Plausible Mechanism

Based on the aforementioned control experiments, mechanistic studies, and previous literature reports [32, 33, 34, 35, 36, 37, 38], plausible mechanisms for both the Ru‐ and Rh‐catalyzed transformations (up to the indoline product) are depicted in Scheme 6. For the Ru‐catalytic cycle, the reaction is proposed to be initiated with the formation of active Ru(II) species I. Next, a chelation‐assisted directed and reversible ortho‐C−H activation results in the formation of cationic intermediate II (Scheme 6).

SCHEME 6.

Ru and Rh‐catalytic cycle−up to the indoline intermediate.

Coordination followed by regioselective migratory insertion (via a [6,4]‐fused ruthenacycle III), β‐oxygen elimination [32, 33, 34, 35, 36], and subsequent regioselective re‐insertion of N−Ru species into the internal olefinic double bond of intermediate IV delivers intermediate V (Scheme 6(A)). Finally, a β‐hydroxide elimination [68] from V releases the desired indoline and regenerates the Ru‐catalyst via an active [Ru]−OH species VI. Notably, the intermediates I, II, V, and VI were detected in ESI‐HRMS (see the Supporting Information for details).

Under Rh‐catalysis, the in situ generated active Rh‐catalyst I engages in a reversible C─H activation to give the intermediate II (Scheme 6B). Further, coordination followed by regioselective migratory insertion and subsequent β‐oxygen elimination delivers the chelated organometallic species IV via a [6,4]‐fused intermediate III. It subsequently undergoes regioselective re‐insertion of Rh‐species [66, 67] followed by a β‐hydroxide elimination [68] to regenerate the active catalyst I via VI along with the release of indoline intermediate VII (Scheme 7).

SCHEME 7.

Rh‐catalytic cycle from the indoline intermediate to indole‐2‐acetate.

Further, two possible pathways have been proposed for the Rh‐catalyzed indole‐2‐acetate formation from the in situ generated indoline intermediate VI (Scheme 7). To begin with, species VI (by following the major “Path a”) readily undergoes a ligand exchange with MeOH [64] to form a transient cationic species VIII. This subsequently undergoes coordination with the pyrimidyl nitrogen, followed by a challenging regioselective migratory insertion of the [Rh]‐OMe species into the vinylic “C═C” bond [65] to generate a cationic intermediate X (detected by ESI‐HRMS analysis). Intermediate X then undergoes a syn β‐hydride elimination to form the reactive vinyl ether XII along with the release of cationic Rh(III)−H species XI (detected by ESI‐HRMS analysis), which subsequently reacts with AcOH to generate the cationic [Cp*Rh(OAc)]⁺ complex XIV [69] along with the release of H₂ and continues the catalytic cycle. On the other hand, intermediate XII in the presence of H⁺/H₂O undergoes hydration to generate transient hemiacetal intermediate XIII, which is finally oxidized either by Cu(II) or Cu(I) [70, 71] to release the desired indole‐2‐acetate product 4 (Scheme 7).

At the same time, indole‐2‐acetate 4 could also be achieved from the cationic intermediate VI by following the minor pathway, that is, “Path b.” It starts from VI through a ligand exchange with [Cu] to generate species XIV [72], which undergoes a regioselective vinylic C─H activation of the in situ generated indoline intermediate VII via a concerted‐metalation‐deprotonation (CMD) to form a cationic Rh(III)‐intermediate XV (detected by ESI‐HRMS). This then picks up MeOH as a ligand and a subsequent reductive elimination forms the vinyl ether species XII along with the generation of Rh(I)‐species XVII. Finally, a sequence of hydration and oxidation [70, 71] releases the final product 4. On the other hand, intermediate XVII undergoes oxidative addition into the O─H bond of MeOH [73, 74, 75] to generate Rh(III)−H species XVIII (confirmed by spike test) which upon a ligand exchange with the acetate affords species XIV [76]. This upon C─H activation with indoline VII regenerates intermediate XV and continues the catalytic cycle (Scheme 7). However, it should be emphasized that intermediate XVIII may generate a putative Cp*Rh(III)−dihydride species [62, 63] XIX, along with the release of HCHO XX (as detected by ¹H NMR studies, Scheme 5B) via a β‐hydride elimination pathway.

2.5. Computational Studies

Following the experimental observations and deuterium labelling studies, we performed detailed DFT calculations to gain more insights into the mechanism and to elucidate the most favorable pathway (see the Supporting Information for details). We successfully computed the catalytic cycles for both Ru‐ and Rh‐catalysis originating from intermediate VI to form the desired indole‐2‐acetate 4 (Scheme 8).

SCHEME 8.

Rh‐catalytic DFT analysis‐free energy profile and % V _Bur studies for the formation of 4. The values are at the M06(SMD, DMSO)/SDD(Rh, Ru)/def2‐TZVP//BP86/SDD(Rh, Ru)/def2‐SVP level of theory. Non‐essential hydrogen atoms are omitted for clarity.

The free energy profile diagram, as shown in Scheme 8, begins with the ligand exchange of VI with MeOH to generate the thermodynamically more stable intermediate VIII. The Rh−OMe species VIII forms the loosely bonded transient adduct TS1 through interaction with the in situ formed intermediate VII via coordination of pyrimidyl nitrogen with Rh, which is simultaneously involved in an interaction with the “C═C” bond of the vinyl functionality, thereby generating intermediate IX. This is followed by regioselective insertion of the Rh−OMe species into the “C═C” bond of the vinyl segment to form the cationic intermediate X (Scheme 8), which readily corresponds to the experimental observation as well (Scheme 5C). This step is associated with a barrier of 14.9 kcal/mol. Subsequently, syn β‐hydride transfer to the cationic Rh‐center and reconstitution of the olefinic double bond on the substrate is energetically most favorable to generate the loosely bound metal complex A via TS2 (barrier of 1.6 kcal/mol) (Scheme 8). Next, the addition of methoxide anion (⁻OMe) gives the intermediate complex XVI, which, after reductive elimination, leads to the formation of vinyl ether species XII, which is exergonic by 24.7 kcal/mol. Further hydrolysis of XII and subsequent oxidation assisted by Cu(II) or Cu(I) leads to the formation of the final product 4. On the other hand, β‐hydride elimination from species XVI generates the crucial intermediate Cp*RhH₂ (dihydride complex) XX, which is energetically favorable (stabilized by 19.3 kcal /mol) along with the formation of HCHO (XXI), which was also experimentally detected by ¹H NMR analysis (Schemes 5B and 8).

To gain a comprehensive understanding of the steric parameters influencing the reaction outcome, we next compared the %V_Bur and the energy of the transition state involved in the reaction (Scheme 8B). In the transition state, TS1_Ru required a higher energy barrier of almost 4.3 kcal/mol compared to the TS1, due to the high steric hindrance of the ligand (p‐cym) to the metal, as evidenced in the steric map at NE and NW quadrants (Scheme 8B; see the Supporting Information for details). For comparison, we examined the same substrate with the Ru catalyst and performed analogous DFT calculations, focusing primarily on the transition‐state barrier (see red‐colored barrier for Ru, Scheme 8). The Ru‐catalyzed mechanism begins with ligand exchange of complex VI_Ru with methoxy anion (⁻OMe), yielding intermediate IX_Ru. Subsequent coordination of substrate VII_Ru with the Ru‐metal forms the intermediate XVII_Ru, which then undergoes a re‐insertion of the Ru−OMe species into the “C═C” bond of the vinyl segment, which then proceeds to the intermediate B_Ru via the transition state TS1_Ru. This step features a barrier of 19.2 kcal/mol. Notably, this transition state for the insertion is 4.3 kcal/mol higher in energy for Ru‐catalysis as compared to the corresponding Rh‐catalyzed counterpart, highlighting the greater efficiency of Rh in promoting this transformation (Scheme 8A; see the Supporting Information for details). Additionally, we analyzed the steric influence of nucleophiles of varying sizes within the chiral pocket by calculating their buried volumes.

To probe the effect of different alcohols (MeOH, EtOH, CF₃CH₂OH, and CH(CF₃)₂ OH), we evaluated the transition‐state barriers (TS1) for the reactions, as shown in Scheme 9 (see the Supporting Information for details). As illustrated in Scheme 7, while the Rh–OMe species undergoes regioselective insertion into the “C═C” bond of the vinyl indoline, other alcohols display substantially higher transition‐state barriers, consistent with the experimental findings. A detailed quadrant analysis revealed that the SW and SE quadrants are nearly symmetrical (SW: 95.8%, 95.6%, 94.7%; SE: 96.7%, 96.7%, 96.8% for TS1_a, TS1_b, and TS1_c, respectively).

SCHEME 9.

Topographical steric maps of transition states (TS1_a, TS1_b, TS1_c) showing %V _bur per quadrant.

In contrast, the NW and NE quadrants show slight asymmetry (NW: 82.4%, 82.7%, 84.4%; NE: 60.4%, 60.5%, 58.3% for TS1_a, TS1_b, and TS1_c, respectively). Although the overall %V _Bur values are comparable, the observed differences in transition‐state energies can be rationalized by the steric challenges of fitting nucleophiles of different sizes into the restricted catalytic pocket of the Cp*Rh catalyst, which offers limited conformational flexibility.

2.6. Unsuccessful Validation of the Proposed Rh(III)/Rh(V) Pathway

We had initially considered a catalytic cycle involving a Rh(III)/Rh(V) redox manifold (Scheme 10). However, experimental attempts to detect or spectroscopically characterize any high‐valent Rh(V) species were unsuccessful. Moreover, kinetic studies and control experiments indicated that the reaction followed a pathway more consistent with a Rh(III)/Rh(I) catalytic cycle rather than a Rh(III)/Rh(V) one. Additionally, computational studies involving free‐energy profiles for Rh‐catalysis and %V _Bur studies did not provide sufficient evidence for the Rh(III)−Rh(V) catalytic pathway. This mechanistic divergence underscores the complexity of Rh‐mediated transformations involving vinyl ethylene carbonates and highlights the subtle interplay between the oxidation state, ligand environment, and substrate reactivity in governing proper catalytic pathways.

SCHEME 10.

An initially proposed Rh(III)−Rh(V) catalytic cycle that was later ruled out.

The proposed catalytic pathway involving a Rh(III)−Rh(V) pathway is depicted in Scheme 10. The formation of intermediate VI was the same as shown in Scheme 6B. Upon ligand exchange with Cu(OAc)₂.H₂O, it generates cationic Rh(III)‐intermediate XIV, which subsequently undergoes regioselective re‐insertion into the “C═C” bond of in situ generated vinyl intermediate VII to give a thermodynamically more stable intermediate B (following Path a). Species B may undergo a facile oxidative addition into MeOH to form a putative [Rh(V)]−H species C. Intermediate C may also be formed from XIV via regioselective C−H activation of indoline VII through an oxidative addition mechanism and followed by a ligand exchange with MeOH (following Path b). Intermediate C may also be generated from intermediate VIII (generated from species VI via a ligand exchange with MeOH) after regioselective C−H activation through an oxidative addition mechanism (following Path c). A fast reductive elimination of species C leads to XII and subsequent hydrolysis by H⁺/H₂O forms the corresponding hemiacetal XIII. Subsequently, oxidation, as assisted by Cu(II) or Cu(I), leads to the formation of the indole‐2‐acetate 4. The active [Rh(III)]‐catalyst XIV is regenerated from the Cp*Rh(III)−H species XI. The intermediate XIV subsequently undergoes ligand exchange with MPAA to regenerate species I and continue the catalytic cycle (Scheme 10). Unfortunately, this pathway could not be validated either experimentally or computationally.

3. Conclusions

To conclude, we have investigated the transition‐metal‐catalyzed dual reactivity of vinyl ethylene carbonates with a wide array of anilines. This leads to the formation of two different N‐heterocycles under two different catalytic conditions. Under Ru‐catalysis, the reactivity of VECs is restricted up to the β‐hydroxide elimination step, affording the 2‐vinyl indoline as the final product. Under Rh‐catalysis, a unified mechanism for the formation of indole‐2‐acetates was elucidated through a combination of experimental and computational studies. Detailed DFT studies, including free‐energy profiles for both Ru‐ & Rh‐catalysis, a comprehensive understanding of the steric parameters, steric influence of nucleophiles of varying sizes based on their %V_Bur studies were performed to gain deeper insights into the mechanism, and these correlate well to the experimental results wherein competing pathways (major and minor) were observed. These include a highly regioselective insertion of Rh−OMe species into the “C═C” bond of the vinyl segment, and an interesting oxidative addition of MeOH into Rh(I) and a Cp*Rh‐dihydride intermediate formation. The role of the Rh‐catalyst, MPAA‐ligand, and MeOH was also investigated in this protocol. Additionally, this methodology also demonstrates downstream synthetic applications of the annulated products.[77]

4. Experimental Section

General methods, synthetic procedures of starting materials, general method of the preparation of 2‐vinyl indolines, indole‐2‐acetates, and their further derivatization are described in detail in the Supporting Information. ¹H and ¹³C NMR spectra and x‐ray data, as well as details of the computational studies, are provided in the Supporting Information.

Conflicts of Interest

The authors declare no competing financial interest.

Supporting information

Supporting File: chem71132‐sup‐0001‐SuppMat.pdf

Data Availability Statement

The data that support the findings of this study are available in the Supporting Information of this article.