Rhodium(III) Iodide Catalyzed Carboamination of Alkynes through C–N Bond Activation

Abstract

Here, we report a RhI₃-catalyzed intramolecular carboamination reaction to access polysubstituted indoles. The protocol features a broad substrate scope (>20 examples), good functional-group compatibility, and a low catalyst loading (5 mol% Rh). Good to excellent yields (up to 98%) were obtained. An unprecedented C–N bond-cleavage mode via a six-membered transition state σ-bond metathesis mechanism was proposed based on control experiments. A series of C3-allylated indole derivatives were accessed, proving that the system provides an alternative catalytic route to polysubstituted indoles.

The intramolecular carboamination of alkynes is an important strategy for synthesizing polysubstituted indoles and their analogues, which are key molecular frameworks found in both pharmaceuticals and bioactive natural products (Scheme [1a]).[1] The carboamination reaction is usually triggered by electrophilic cyclization from an ortho-alkynylaniline I (Scheme [1b]) through triple-bond activation by coordination of a π-acidic transition metal. This generates a metalated zwitterionic intermediate II that undergoes a formal 1,3-migration of the Y group from the nitrogen to the C3 position, affording the 2,3-disubstituted indole skeleton III. Since the seminal reports by the groups of Cacchi[2] and Yamamoto,[3] who used Pd(II) and Pt(II) catalysts, respectively, a series of transition metals have been reported to catalyze such processes.[4] [5] Astonishingly, no rhodium complex has been reported to engage in the carboamination of alkynylanilines.[6] It has been suggested that, unlike the coinage metals, Rh is not intrinsically alkynophilic. However, the Rh(I)-mediated olefin migration/hydrolysis process for formally deprotecting an allyl group on a nitrogen atom is a well-known protocol (Scheme [1c]).[7] Here, we report a Rh-catalyzed C–N bond-activation strategy to effect the carboamination reaction, providing access to polysubstituted indole frameworks (Scheme [1d]). The reaction features a low catalyst loading (5 mol% of Rh), a broad substrate scope (24 examples), and good to excellent yields (47–98%), besides (and most importantly) its mechanistically novel C–N bond-activation mode.

Our investigation began with the preparation of the ortho-alkynylaniline 1a, which was readily synthesized in two steps from the commercially available feedstock N-methyl-2-iodoaniline [see the Supporting Information (SI) for details]. We found that the desired carboamination reaction could be realized in the presence of 2.5 mol% of [Cp*RhCl₂]₂ (Cp* = η⁵-pentamethylcyclopenta-1,3-diene) in 1,4-dioxane at 150 °C without any additive, giving the 2,3-disubstituted indole 2a in 94% yield (Table [1], entry 1); these conditions are henceforth referred to as the ‘standard conditions’. We also ran a few control experiments under nonstandard conditions in an attempt to identify the optimal conditions. We found that Cp*Rh(MeCN)₂(SbF₆)₂ (10 mol%) and Rh(acac)₃ (10 mol%) also catalyzed the transformation with similar efficacy (entries 2 and 3). Additionally, the reaction gave an almost quantitative yield (98%) in the presence of only 5 mol% of RhI₃ at a lower temperature (120 °C) (entry 4). In contrast, Wilkinson’s catalyst only led to decomposition of the starting material (entry 5). The reaction of substrate 6 did not proceed in the absence of a Rh catalyst, and the desired product 7 was not obtained (entry 6). A Brønsted acid (HCl) drove the cyclization reaction, but gave the deallylated product 5 in a low 19% yield (entry 7). The reaction failed with substrate 3, which lacked an allyl pendent group (entry 8), indicating the allylamine might be the kick-off point for the C–N bond-cleavage step. Under the standard conditions, reactant 4 gave a 77% yield of the allyl derivative 5 (entry 9), the structure of which was verified by X-ray crystallography.[8]

Table 1 Selected Optimization Studies^a

Entry	Deviation from standard conditions	Conversionb (%)	Product	Yieldc (%)
1	–	100	2a	94
2	Cp*Rh(MeCN)2(SbF6)2 (10 mol%)	100	2a	94
3	Rh(acac)3 (10 mol%)	69	2a	68
4	RhI3 (5 mol%) at 120 °C	100	2a	98
5	Wilkinson’s catalyst (10 mol%)	44	–	–
6	substrate 6; no catalyst	–	–	–
7	1 M HCl in 1,4-dioxane	48	5	19
8	substrate 3	–	–	–
9	substrate 4	83	5	77

^a All reactions were run with 2.5 mol% precatalyst on a 20 mg scale in toluene at 150 °C for 12 h unless otherwise noted.

^b Determined based on the recycled substrate.

^c Isolated yield.

With the optimal reaction conditions in hand (Table [1], entry 4), we were ready to explore the scope and functional-group compatibility of the reaction (Table [2]). First, we examined the effects of the olefin substitution pattern. It appeared that steric bulk and electronic variations were both tolerated, as the desired indoles 2a–g were obtained under the optimized conditions in yields of 88–98%, with the general trend that alkyl substituents were favored over aryl groups. Steric effects were not quite as obvious, as both the isopropylated (2b, 91%) and cyclopentylated products (2d, 96%) were obtained in high yields. Hetaryl groups were also explored, and products 2h (91%) and 2i (77%) were both obtained in reasonable yields.

Table 2 Substrate Scope^a,b

^a All reactions were run with 5 mol% RhI₃ on a 20 mg scale in toluene at 120 °C for 12 h, unless otherwise noted.

^b Isolated yield.

^c brsm = based on recovered starting material.

^d [Cp*RhCl₂]₂ and 150 °C was used.

A counterintuitive observation is that the 1,4-diene groups obtained in products 2e–i were well preserved under the reaction conditions. No olefin-isomerization byproducts were detected although these were assumed to be more thermodynamically stable, as a larger conjugation would have resulted. Nevertheless, this observation also indicated that Rh(III) does not trigger an olefin migration for products such as 2.

Next, the electronic effects of substituents on the aniline were examined. Pleasingly, C5 and C6 halo- and trifluoromethyl-substituted substrates were all viable substrates and gave the corresponding products 2j–q in yields of 81–96%. Terminal olefins also proved to be suitable substrates, but the olefin migration was completely suppressed, affording 2r in 47% yield [96% based on the recovered starting material (brsm)].

Alkyne substituents were then investigated. The ortho sterically hindered substrate 1s afforded the desired product 2s in 53% yield (85% brsm), providing a further indication that the Rh catalyst might not activate the alkyne moiety directly. Alkyl substituents such as methyl groups were perfectly compatible, and product 2t was obtained in 56% yield (85% brsm). When a substrate with an electron-withdrawing chloro group attached to the alkyne was tested, the desired product 2u was smoothly obtained in an excellent yield (78%).

Finally, substrates containing a 1,2-disubstituted olefin group (1v) or a trisubstituted olefin group (1w) were studied; in both cases, the branched product (2v or 2w) was obtained exclusively in yields of 52 and 80%, respectively. A terminal alkyne (1x) and an N-allylated substrate (6) were both viable under standard conditions, and afforded corresponding products 2x and 7 in satisfactory yields (57 and 72%, respectively).

We then carried out a control experiment to explore the C–N bond-cleavage step and to probe the reaction mechanism (Scheme [2]A). When we performed a crossover reaction with 1c and 1l as substrates, unsurprisingly both 2c and 2l were obtained in decent yields. What was more informative was that we also identified a trace amount of the crossover product 2cl; this product could be detected by LC/MS only (see SI for details). In addition, when we used 10 mol% of RhI₃ to catalyze the intramolecular carboamination of 1a, besides the desired product 2a, we also identified the m/z peak of an allyl iodide by GC/MS analysis (see SI for details). We then carried out a gram-scale reaction with 1a using only 2.0 mol% of RhI₃, and we isolated the desired product 2a in 42% yield (96% brsm), indicating an excellent mass balance (Scheme [2]B).

Based on the control experiments and our group’s previous results with Rh catalysis,[9] [10] we tentatively propose the reaction mechanism shown in Scheme [3]. C–N bond activation is catalyzed by RhI₃ by a concerted ligand-to-ligand exchange to form a N–Rh(III) bond and 3-iodo-2-methylprop-1-ene. The resulting intermediate B undergoes a 5-endo-dig migratory insertion, affording the indole–rhodium intermediate C, which engages in alkylation with the 3-iodo-2-methylprop-1-ene formed in situ to give product 2a, with restoration of RhI₃. Moreover, a very small amount of the protonation byproduct 5 was detected by HPLC (see SI for details).

To conclude, we have developed a RhI₃-catalyzed intramolecular carboamination reaction to access polysubstituted indoles.[11] The reaction features a broad substrate scope and excellent functional-group compatibility. Good to excellent yields were obtained for more than 20 examples. This reaction provides an alternative catalytic system to access 2,3-disubstituted indoles.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgment

We thank the whole pharmacy department of the Ocean University of China for general support.

• References and Notes

• 1a Makarieva TN, Ilyin SG, Stonik VA, Lyssenko KA, Denisenko VA. Tetrahedron Lett. 1999; 40: 1591
  CrossrefSearch in Google Scholar

For a review on recent developments of indole-containing antiviral agents, see:
• 1b Zhang M.-Z, Chen Q, Yang G.-F. Eur. J. Med. Chem. 2015; 89: 421
  CrossrefPubMedSearch in Google Scholar
• 1c Hamid H A, Ramli AN. M, Yusoff MM. Front. Pharmacol. 2017; 8: 96
  PubMedSearch in Google Scholar
• 1d Zhang Y, Li X, Xu T. Org. Biomol. Chem. 2021; 19: 9138
  CrossrefPubMedSearch in Google Scholar
• 1e Yu H, Li X, Jia Y, Zhang D, Xu T. Tetrahedron 2021; 91:  132240
  CrossrefSearch in Google Scholar
• 2a Cacchi S, Fabrizi G, Pace P. J. Org. Chem. 1998; 63: 1001
  CrossrefSearch in Google Scholar
• 2b Cacchi S. J. Organomet. Chem. 1999; 576: 42
  CrossrefSearch in Google Scholar
• 3a Kamijo S, Yamamoto Y. J. Am. Chem. Soc. 2002; 124:  11940
  CrossrefPubMedSearch in Google Scholar
• 3b Shimada T, Nakamura I, Yamamoto Y. J. Am. Chem. Soc. 2004; 126: 10546 ; and references cited therein
  CrossrefPubMedSearch in Google Scholar
• 4 For a stoichiometric carboamination using a Cr complex, see: Rudler H, Parlier A, Bezennine-Lafollée S, Vaissermann J. Eur. J. Org. Chem. 1999; 2825
  CrossrefPubMed

For representative reports on carboaminations to access 2,3-difunctionalized indoles through a N to C3 translocation, see (Pd):
• 5a Zhao F, Zhang D, Nian Y, Zhang L, Yang W, Liu H. Org. Lett. 2014; 16: 5124
  CrossrefPubMedSearch in Google Scholar

(Au):
• 5b Zeng X, Kinjo R, Donnadieu B, Bertrand G. Angew. Chem. Int. Ed. 2010; 49: 942
  CrossrefPubMedSearch in Google Scholar
• 5c Nakamura I, Yamagishi U, Song D, Konta S, Yamamoto T. Angew. Chem. Int. Ed. 2007; 46: 2284
  CrossrefPubMedSearch in Google Scholar
• 5d Chong E, Blum SA. J. Am. Chem. Soc. 2015; 137:  10144
  CrossrefPubMedSearch in Google Scholar

(Pt):
• 5e Nakamura I, Sato Y, Konta S, Terada M. Tetrahedron Lett. 2009; 50: 2075
  CrossrefSearch in Google Scholar
• 5f Patra SR, Sangma SW, Padhy AK, Bhunia S. J. Org. Chem. 2022; 87: 9714
  CrossrefPubMedSearch in Google Scholar

(Ru):
• 5g Chiang P.-Y, Lin Y.-C, Wang Y, Liu Y.-H. Organometallics 2010; 29:  5776
  CrossrefSearch in Google Scholar
• 5h Wu C.-Y, Hu M, Liu Y, Song R.-J, Lei Y, Tang B.-X, Li R.-J, Li J.-H. Chem. Commun. 2012; 48: 3197
  CrossrefPubMedSearch in Google Scholar

(Re):
• 5i Rong M.-G, Qin T.-Z, Zi W. Org. Lett. 2019; 21: 5421
  CrossrefPubMedSearch in Google Scholar

(Ir):
• 5j Badart MP, Hawkins BC. Synthesis 2021; 53: 1683
  Thieme ConnectSearch in Google Scholar
• 6 For a benzofuran synthesis by a Rh-mediated enyne cycloisomerization, see: Sakiyama N, Noguchi K, Tanaka K. Angew. Chem. Int. Ed. 2012; 51: 5976
  CrossrefPubMedSearch in Google Scholar
• 7 For a representative review, see: Escoubet S, Gastaldi S, Bertrand M. Eur. J. Org. Chem. 2005; 2005: 3855
  CrossrefPubMedSearch in Google Scholar
• 8 CCDC 2215689 contains the supplementary crystallographic data for compound 7. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures
• 9 Zhang J, Wang X, Xu T. Nat. Commun. 2021; 12: 3022
  CrossrefPubMedSearch in Google Scholar
• 10 Wang Y, Qiu B, Hu L, Lu G, Xu T. ACS Catal. 2021; 11: 9136
  CrossrefSearch in Google Scholar
• 11 Qiu B, Li X.-T, Zhang J.-Y, Zhan J.-L, Huang S.-P, Xu T. Org. Lett. 2018; 20: 7689
  CrossrefPubMedSearch in Google Scholar
• 12 Wang X, Liu F, Xu T. Chin. Chem. Lett. 2023; 34: 107624
  CrossrefPubMedSearch in Google Scholar
• 13a Zhang Y, Shen S, Fang H, Xu T. Org. Lett. 2020; 22: 1244
  CrossrefPubMedSearch in Google Scholar
• 13b Sun T, Zhang Y, Qiu B, Wang Y, Qin Y, Dong G, Xu T. Angew. Chem. Int. Ed. 2018; 57: 2859
  CrossrefPubMedSearch in Google Scholar
• 14 1-Methyl-3-(2-methylprop-2-en-1-yl)-2-phenyl-1H-indole (2a); Typical Procedure In a nitrogen-filled glove box, an oven-dried 4 mL vial was charged with amine 1a (23.7 mg, 0.091 mmol) and RhI₃ (2.3 mg, 0.005 mmol). Toluene (1 mL) was added, and the mixture was stirred at rt for 5 minutes until the solids fully dissolved. The vial was then sealed with a PTFE-lined cap and the mixture was stirred for 12 h on a pie-block preheated to 120 °C. The mixture was then directly filtered through Celite and silica gel, which were washed with EtOAc (20 mL). The solvent was removed under reduced pressure, and the crude residue was purified by flash column chromatography (silica gel) to give a yellow oil; yield: 22.9 mg (98%). IR (FTIR): 3056, 2907, 1647, 1466, 1361, 1014, 890 cm^–1. ¹H NMR (400 MHz, CDCl₃): δ = 7.64 (d, J = 7.8 Hz, 1 H), 7.53–7.47 (m, 2 H), 7.47–7.40 (m, 3 H), 7.37 (d, J = 8.2 Hz, 1 H), 7.30–7.24 (m, 1 H), 7.15 (ddd, J = 7.9, 6.9, 1.0 Hz, 1 H), 4.78 (s, 1 H), 4.70 (s, 1 H), 3.64 (s, 3 H), 3.40 (s, 2 H), 1.73 (s, 3 H). ¹³C NMR (100 MHz, CDCl₃): δ = 145.5, 138.7, 137.4, 132.1, 130.6, 128.4, 128.2, 128.1, 121.7, 119.7, 119.3, 110.8, 110.8, 109.4, 33.4, 31.1, 22.9. HRMS (ESI): m/z [M + Na]⁺ calcd for C₁₉H₁₉NNa: 284.1410; found: 284.1409.

Corresponding Author

Publication History

Received: 21 November 2022

Accepted after revision: 13 January 2023

Accepted Manuscript online:
13 January 2023

Article published online:
09 February 2023

© 2023. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References and Notes

• 1a Makarieva TN, Ilyin SG, Stonik VA, Lyssenko KA, Denisenko VA. Tetrahedron Lett. 1999; 40: 1591
  CrossrefSearch in Google Scholar

For a review on recent developments of indole-containing antiviral agents, see:
• 1b Zhang M.-Z, Chen Q, Yang G.-F. Eur. J. Med. Chem. 2015; 89: 421
  CrossrefPubMedSearch in Google Scholar
• 1c Hamid H A, Ramli AN. M, Yusoff MM. Front. Pharmacol. 2017; 8: 96
  PubMedSearch in Google Scholar
• 1d Zhang Y, Li X, Xu T. Org. Biomol. Chem. 2021; 19: 9138
  CrossrefPubMedSearch in Google Scholar
• 1e Yu H, Li X, Jia Y, Zhang D, Xu T. Tetrahedron 2021; 91:  132240
  CrossrefSearch in Google Scholar
• 2a Cacchi S, Fabrizi G, Pace P. J. Org. Chem. 1998; 63: 1001
  CrossrefSearch in Google Scholar
• 2b Cacchi S. J. Organomet. Chem. 1999; 576: 42
  CrossrefSearch in Google Scholar
• 3a Kamijo S, Yamamoto Y. J. Am. Chem. Soc. 2002; 124:  11940
  CrossrefPubMedSearch in Google Scholar
• 3b Shimada T, Nakamura I, Yamamoto Y. J. Am. Chem. Soc. 2004; 126: 10546 ; and references cited therein
  CrossrefPubMedSearch in Google Scholar
• 4 For a stoichiometric carboamination using a Cr complex, see: Rudler H, Parlier A, Bezennine-Lafollée S, Vaissermann J. Eur. J. Org. Chem. 1999; 2825
  CrossrefPubMed

For representative reports on carboaminations to access 2,3-difunctionalized indoles through a N to C3 translocation, see (Pd):
• 5a Zhao F, Zhang D, Nian Y, Zhang L, Yang W, Liu H. Org. Lett. 2014; 16: 5124
  CrossrefPubMedSearch in Google Scholar

(Au):
• 5b Zeng X, Kinjo R, Donnadieu B, Bertrand G. Angew. Chem. Int. Ed. 2010; 49: 942
  CrossrefPubMedSearch in Google Scholar
• 5c Nakamura I, Yamagishi U, Song D, Konta S, Yamamoto T. Angew. Chem. Int. Ed. 2007; 46: 2284
  CrossrefPubMedSearch in Google Scholar
• 5d Chong E, Blum SA. J. Am. Chem. Soc. 2015; 137:  10144
  CrossrefPubMedSearch in Google Scholar

(Pt):
• 5e Nakamura I, Sato Y, Konta S, Terada M. Tetrahedron Lett. 2009; 50: 2075
  CrossrefSearch in Google Scholar
• 5f Patra SR, Sangma SW, Padhy AK, Bhunia S. J. Org. Chem. 2022; 87: 9714
  CrossrefPubMedSearch in Google Scholar

(Ru):
• 5g Chiang P.-Y, Lin Y.-C, Wang Y, Liu Y.-H. Organometallics 2010; 29:  5776
  CrossrefSearch in Google Scholar
• 5h Wu C.-Y, Hu M, Liu Y, Song R.-J, Lei Y, Tang B.-X, Li R.-J, Li J.-H. Chem. Commun. 2012; 48: 3197
  CrossrefPubMedSearch in Google Scholar

(Re):
• 5i Rong M.-G, Qin T.-Z, Zi W. Org. Lett. 2019; 21: 5421
  CrossrefPubMedSearch in Google Scholar

(Ir):
• 5j Badart MP, Hawkins BC. Synthesis 2021; 53: 1683
  Thieme ConnectSearch in Google Scholar
• 6 For a benzofuran synthesis by a Rh-mediated enyne cycloisomerization, see: Sakiyama N, Noguchi K, Tanaka K. Angew. Chem. Int. Ed. 2012; 51: 5976
  CrossrefPubMedSearch in Google Scholar
• 7 For a representative review, see: Escoubet S, Gastaldi S, Bertrand M. Eur. J. Org. Chem. 2005; 2005: 3855
  CrossrefPubMedSearch in Google Scholar
• 8 CCDC 2215689 contains the supplementary crystallographic data for compound 7. The data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/structures
• 9 Zhang J, Wang X, Xu T. Nat. Commun. 2021; 12: 3022
  CrossrefPubMedSearch in Google Scholar
• 10 Wang Y, Qiu B, Hu L, Lu G, Xu T. ACS Catal. 2021; 11: 9136
  CrossrefSearch in Google Scholar
• 11 Qiu B, Li X.-T, Zhang J.-Y, Zhan J.-L, Huang S.-P, Xu T. Org. Lett. 2018; 20: 7689
  CrossrefPubMedSearch in Google Scholar
• 12 Wang X, Liu F, Xu T. Chin. Chem. Lett. 2023; 34: 107624
  CrossrefPubMedSearch in Google Scholar
• 13a Zhang Y, Shen S, Fang H, Xu T. Org. Lett. 2020; 22: 1244
  CrossrefPubMedSearch in Google Scholar
• 13b Sun T, Zhang Y, Qiu B, Wang Y, Qin Y, Dong G, Xu T. Angew. Chem. Int. Ed. 2018; 57: 2859
  CrossrefPubMedSearch in Google Scholar
• 14 1-Methyl-3-(2-methylprop-2-en-1-yl)-2-phenyl-1H-indole (2a); Typical Procedure In a nitrogen-filled glove box, an oven-dried 4 mL vial was charged with amine 1a (23.7 mg, 0.091 mmol) and RhI₃ (2.3 mg, 0.005 mmol). Toluene (1 mL) was added, and the mixture was stirred at rt for 5 minutes until the solids fully dissolved. The vial was then sealed with a PTFE-lined cap and the mixture was stirred for 12 h on a pie-block preheated to 120 °C. The mixture was then directly filtered through Celite and silica gel, which were washed with EtOAc (20 mL). The solvent was removed under reduced pressure, and the crude residue was purified by flash column chromatography (silica gel) to give a yellow oil; yield: 22.9 mg (98%). IR (FTIR): 3056, 2907, 1647, 1466, 1361, 1014, 890 cm^–1. ¹H NMR (400 MHz, CDCl₃): δ = 7.64 (d, J = 7.8 Hz, 1 H), 7.53–7.47 (m, 2 H), 7.47–7.40 (m, 3 H), 7.37 (d, J = 8.2 Hz, 1 H), 7.30–7.24 (m, 1 H), 7.15 (ddd, J = 7.9, 6.9, 1.0 Hz, 1 H), 4.78 (s, 1 H), 4.70 (s, 1 H), 3.64 (s, 3 H), 3.40 (s, 2 H), 1.73 (s, 3 H). ¹³C NMR (100 MHz, CDCl₃): δ = 145.5, 138.7, 137.4, 132.1, 130.6, 128.4, 128.2, 128.1, 121.7, 119.7, 119.3, 110.8, 110.8, 109.4, 33.4, 31.1, 22.9. HRMS (ESI): m/z [M + Na]⁺ calcd for C₁₉H₁₉NNa: 284.1410; found: 284.1409.

• Supplementary Material