A chemoselective radical cascade polarity-mismatched silylarylation of unactivated alkenes

Abstract

Here, we report a practical method to selectively access silylated pyrrolo[1,2-a]indoles using alkenyl indoles and readily available silanes through a radical cascade cyclization. The reaction proceeds through an unusual intermolecular polarity-mismatched addition of a nucleophilic silyl radical to e-rich alkene SOMOphiles via a coordination-assisted interaction. The scope and functional group compatibility of the protocol as well as a detailed mechanistic investigation are presented.

---

Polarity plays an important role in the overall rate of free radical addition reactions. Waters and co-workers developed pioneering work on the research of the polarity effect on the rate and selectivity of addition of radicals to alkenes.¹ Later, excellent studies of the importance of polarity in determining the rate and orientation of free radical addition reactions were developed by Walling, Russell, and Tedder et al.² It can be stated that the nature of the polarity-matched radical addition is favorable to proceed, while the radical reactions with mismatched polarity are unfavorable (Scheme 1a). Owing to the mismatched polarity,³ the addition of nucleophilic radicals to e-rich SOMOphiles in a radical cascade transformation is still a long-standing challenge. So far, examples of polarity-mismatched radical additions are exceedingly rare. For example, an efficient intermolecular polarity-mismatched addition of a nucleophilic alkyl radical to e-rich alkene SOMOphiles via a palladium-catalyzed two-component dicarbofunctionalization reaction was developed by Loh and co-authors (Scheme 1b).⁴ Later, Yu has demonstrated a polarity-mismatched photoredox-catalyzed iminoalkenylation of e-rich alkene SOMOphiles based on a sequential single electron transfer (SET)/energy transfer/hydrogen atom transfer (HAT) process, where radical intermediates are intercepted by organoboron reagents.⁵ Additionally, polarity transduction is an efficient strategy to realize formal electronically mismatched radical addition to e-rich alkene SOMOphiles, as reported by Silvi and co-workers (Scheme 1c).⁶ Vinyl sulfonium salts were used as vinyl halide equivalents, allowing the preparation of products that display matched polarity. Recently, a photoredox-catalyzed C(sp³)–C(sp²) coupling reaction has been applied for realizing an unusual polarity-mismatched radical addition by synergistic activation of the radical precursor and organoboron (Scheme 1d).⁷

Scheme 1 Radical additions to unactivated alkenes via the polarity-mismatched pathway. (a) General description of polarity additions. (b) Pd-catalyzed polarity-mismatched addition via coordination-assisted interaction. (c) Polarity transduction strategy for the polarity-mismatched Giese-type reaction. (d) Polarity-mismatched radical addition via synergistic boron activation. (e) This work: Cu-promoted polarity-mismatched addition via coordination-assisted interaction.

Organosilane-based compounds play an indispensable role in various aspects of modern chemistry. Due to their high chemical and thermal stability as well as their low toxicity, organosilicons are widely employed in natural products,⁸ paints,⁹ polymer materials¹⁰ and other medicinal applications.¹¹ Thus, a significant effort has been made to establish new and efficient approaches for accessing a wide variety of organosilicon compounds.¹²

Liu and co-authors have developed an efficient route to silylated oxindoles by the radical cascade silylarylation of activated alkenes through selective activation of the Si–H/C–H bonds.¹³ A recent approach by Qin and co-workers developed a practical dearomatization method for the hydrosilylation and reduction of bicyclic heterocycles to selectively access hydrosilylated and reduced heterocycles via a stepwise HAT/radical recombination process.¹⁴ Despite many elegant works that have been made, there is still an unmet need for general and versatile methods to prepare organosilane-based compounds, especially for the synthesis of silylated heterocycles. Herein, we report on a direct polarity-mismatched radical cascade addition of a nucleophilic radical to an e-rich SOMOphile, enabling access to a series of silylated pyrrolo[1,2-a]indoles (Scheme 1e).

Inspired by these advances and based on our previous work,¹⁵ we envisioned an alternative and potentially general method for the direct polarity-mismatched radical cascade addition by introducing a coordination group to unactivated alkenes, giving rise to polarity-reversed nucleophilic silyl radicals. We selected 1-(but-3-en-1-yl)-1H-indole 1a as the model substrate because the corresponding fused tricyclic products with pyrrolo[1,2-a]indole core skeletons are key building blocks widely employed in the practice of natural products and medicinal chemistry,¹⁶ and, to the best of our knowledge, the direct polarity-mismatched radical cascade silylarylation of 1-(but-3-en-1-yl)-1H-indole 1a has not been reported to date. Selected optimization studies are depicted in Table 1. Not surprisingly, the yield of the desired product 3aa was unsatisfactory with the recovery of the starting material (Table 1, entry 1). To improve the yield of 3aa, we screened various parameters including the initiator, peroxide initiator, ligand, and solvent. Gratifyingly, when CuCN and DTBP were used as radical initiators with the 2,2′-bipyridine ligand and t-BuOH as a reaction solvent, the desired cascade cyclization product 3aa was obtained in 50% yield along with a trace of an alkenylsilane side product, which was routinely obtained in polarity-mismatched radical addition transformations (Table 1, entry 5). After a series of radical initiators were investigated, CuCN was proved to be the most effective radical initiator (Table 1, entries 1–14). Further optimization of the effect of the ligand indicated that 2, 4′-bipyridine was the best, affording 3aa in 66% yield (Table 1, entries 15–20). Subsequent optimization of the peroxide radical initiators indicated that DTBP was the best, giving 3aa in the best yield (Table 1, entries 21 and 22). The reaction was found to be sensitive to the solvent and t-BuOH provided an optimal cyclization product (Table 1, entries 23–28). Other solvents led to lower yields or no corresponding product formation. Besides, the transformation failed to work with the recovery of the starting material when Fe salts were used as radical initiators (Table 1, entries 29 and 30). It is noteworthy that the initiator and peroxide radical initiator are essential for this transformation (Table 1, entries 32 and 33).

Table 1 Optimization of the reaction conditions^a

Entry	Initiator	Peroxide initiator	Ligand	Solvent	Yield /3aa ^b
a A mixture of 1aa (0.2 mmol, 1.0 equiv.), 2aa (2.0 mmol, 10.0 equiv.), initiator (0.04 mmol, 20.0 mol%), peroxide initiator (1.0 mmol, 5.0 equiv.), ligand (0.04 mmol, 20.0 mol%), and solvent (3 mL) was sealed in a 25 ml Schlenk tube under a nitrogen atmosphere at 130 °C for 15 h. b Yields of the isolated product 3aa. c 2,2′-Bipyridine. d C27H36ClCuN2 = Chloro[1,3-bis(2,6-di-i-propylphenyl)imidazol-2-ylidene]copper(I). e C8H12CuF6N4P = Tetrakis(acetonitrile)copper(I) hexafluorophosphate. f 2,4′-Bipyridine. g TBHP = t-Butyl hydroperoxide (in H2O). h DTBP = di-tert-Butyl peroxide. i TFE = Trifluoroethanol. j Under an air atmosphere. k Without an initiator. l Without a peroxide initiator.
1	CuCl	DTBP	2,2′-bpy^c	t-BuOH	Trace
2	CuBr	DTBP	2,2′-bpy	t-BuOH	Trace
3	CuI	DTBP	2,2′-bpy	t-BuOH	24
4	Cu2O	DTBP	2,2′-bpy	t-BuOH	16
5	CuCN	DTBP	2,2′-bpy	t-BuOH	50
6	CuSCN	DTBP	2,2′-bpy	t-BuOH	33
7	CuOTf	DTBP	2,2′-bpy	t-BuOH	Trace
8	Cu(TC)	DTBP	2,2′-bpy	t-BuOH	15
9	Cu(acac)2	DTBP	2,2′-bpy	t-BuOH	34
10	CuCl2	DTBP	2,2′-bpy	t-BuOH	17
11	CuBr2	DTBP	2,2′-bpy	t-BuOH	16
12	Cu(OAc)2	DTBP	2,2′-bpy	t-BuOH	20
13^d	C27H36ClCuN2	DTBP	2,2′-bpy	t-BuOH	Trace
14^e	C8H12CuF6N4P	DTBP	2,2′-bpy	t-BuOH	17
15	CuCN	DTBP	2,4′-bpy	t-BuOH	66
16	CuCN	DTBP	4,4′-bpy	t-BuOH	19
17	CuCN	DTBP	dtbbpy	t-BuOH	24
18	CuCN	DTBP	1.10-Phen	t-BuOH	15
19	CuCN	DTBP	Isoquinoline	t-BuOH	18
20	CuCN	DTBP	Indole	t-BuOH	34
21^g	CuCN	TBHP	2,4′-bpy	t-BuOH	30
22	CuCN	K2S2O8	2,4′-bpy	t-BuOH	n.d.
23^h	CuCN	DTBP	2,4′-bpy	DMF	Trace
24	CuCN	DTBP	2,4′-bpy	Toluene	32%
25	CuCN	DTBP	2,4′-bpy	THF	Trace
26	CuCN	DTBP	2,4′-bpy	Dioxane	10%
27^i	CuCN	DTBP	2,4′-bpy	TFE	12%
28	CuCN	DTBP	2,4′-bpy	H2O	Trace
29	FeI2	DTBP	2,2′-bpy	t-BuOH	Trace
30	FeCl2	DTBP	2,2′-bpy	t-BuOH	Trace
31^j	CuCN	DTBP	2,4′-bpy	t-BuOH	42%
32^k	—	DTBP	2,4′-bpy	t-BuOH	26%
33^l	CuCN	—	2,4′-bpy	t-BuOH	n.d.

---

Having established the optimized reaction conditions, the scope and functional group compatibility of the introduced system were evaluated (Table 2). A broad variety of different substituted alkenyl indoles containing 4-, 5-, 6-, and 7-position substituents were compatible with the desired radical cascade cyclization products (the structure of 3aa was confirmed by X-ray crystallography).¹⁷ Indole alkene derivatives with strong electron-donating groups (3ac, 3ah, and 3am) or electron-withdrawing groups (3af and 3ak) can be employed to generate the anticipated cyclized products in medium yields. Functional groups such as halides and ether were also well tolerated. C3-Alkyl-substituted indole was a viable substrate and delivered the corresponding polarity-mismatched silylated product 3av in 72% yield. Notably, a range of six-membered heteropolycycles could also be accessed using this protocol (3aw–3ay). Notably, this transformation was also very effective towards biologically active alkenyl indoles from melatonin derivatives to afford the desired product 3az.

Table 2 Substrate scope of alkenyl indoles^a,b

a A mixture of 1 (0.2 mmol, 1.0 equiv.), 2 (2.0 mmol, 10.0 equiv.), CuCN (0.04 mmol, 20.0 mol%), 2,4′-bipyridine (0.04 mmol, 20.0 mol%), DTBP (0.6 mmol, 5.0 equiv.), and t-BuOH (3.0 mL) was sealed in a 25.0 mL Schlenk tube under a nitrogen atmosphere at 130 °C for 15 h. b Isolated yields based on 1.

---

To further diversify the protocol, a wide array of different silanes was examined (Table 3). Besides dimethylphenylsilane, other commercially available silanes, such as (3,5-bis(trifluoro-methyl)phenyl)dimethylsilane, methyldiphenylsilane and triphenylsilane, could uniformly afford the corresponding polarity-mismatched radical addition products (3ba–3bc, 3bf, 3bg). Interestingly, when diphenylsilane was employed, the radical cascade cyclization exhibited good performance for this transformation with a yield of up to 65% (3bd, 3bh), whereas most of the reported approaches are unable to work.^12a,18 However, no polarity-mismatched addition silylated product could be detected upon substrate scope investigation of phenylsilane (3bi).

Table 3 Cascade cyclization of alkenyl indoles with silanes^a,b

a A mixture of 1 (0.2 mmol, 1.0 equiv.), 2 (2.0 mmol, 10.0 equiv.), CuCN (0.04 mmol, 20.0 mol%), 2,4′-bipyridine (0.04 mmol, 20.0 mol%), DTBP (0.6 mmol, 5.0 equiv.), and t-BuOH (3.0 mL) was sealed in a 25.0 mL Schlenk tube under a nitrogen atmosphere at 130 °C for 15 h. b Isolated yields based on 1.

---

As a demonstration of the scalability of this transformation, a preparative gram-scale experiment of radical cascade polarity-mismatched silylarylation of unactivated alkenes was evaluated and delivered the desired product 3ap smoothly to give 1.240 g with 72% yield (Scheme 2).

Scheme 2 Gram-scale radical cascade polarity-mismatched silylarylation of unactivated alkenes.

To gain some insight into the mechanism of this transformation, radical trapping experiments were performed by adding radical scavengers such as 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) and butylated hydroxytoluene (BHT), which completely inhibited the reaction and produced the silyl-TEMPO/BHT adducts 4/5, supporting the formation of a silyl radical (Scheme 3, eqn (1) and (2)). Moreover, when 1,1-diphenylethylene was used as a radical inhibitor, the yield of product 3bc dramatically dropped to 35%, with the formation of 6 in 30% yield determined by GC-MS (Scheme 3, eqn (3)).

Scheme 3 Radical trapping experiments.

Based on the experimental results and relevant studies on silyl radicals,^13,19 a plausible mechanism was proposed (Scheme 4). The process would begin with an electron transfer from Cu(I) to DTBP with the concomitant generation of Cu(II) species, a tert-butyl anion, and a tert-butyl radical. Subsequent formation of silyl radical A via hydrogen atom abstraction by the tert-butyl radical or methyl radical, which underwent by β-cleavage of the tert-butyl radical. Further radical addition of silyl radical A to unactivated alkenes via a coordination-assisted interaction would generate the corresponding radical intermediate B. Subsequent radical cascade cyclization to the aromatic core would give radical intermediate C. The resulting intermediate C deprotonated by the tert-butyl anion would form radical anion D and finally release an electron to the next catalytic cycle to afford the desired products 3.

Scheme 4 Possible mechanism.

Conclusions

In conclusion, we report on the copper-promoted radical cascade polarity-mismatched silylarylation of unactivated alkenes, which provides straightforward access to a variety of silylated heteropolycycles. The reaction features broad functional group tolerance and a wide substrate scope with moderate to good yields. Remarkably, the coordination-assisted interaction between the silyl radical and indole alkene allows for bond formation between polarity-mismatched sites to form radical cascade cyclization adducts. Further efforts directed at the development of mechanistic details and explorations of medical applications of the products are currently in progress and will be reported in due course.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (22161004), the Fundamental Research Funds for Gannan Medical University (QD202019, QD202106, and TD2021YX05) and Shaoguan University (408/9900064703).

References

1. (a) W. A. Waters, Thiol catalysis of the homolytic decomposition of aldehydes, Nature, 1952, 170, 212–213; (b) K. E. J. Barrett and W. A. Waters, Liquid phase reactions between free radicals and aldehydes. Catalysis of decarbonylation by means of thiols, Discuss. Faraday Soc., 1953, 14, 221–227.
2. (a) C. Walling, Free Radicals in Solution, John Wiley & Sons, New York, 1957; (b) C. Walling, Some aspects of the chemistry of alkoxy radicals, Pure Appl. Chem., 1967, 15, 69–80; (c) G. Russell, Free Radicals, Wiley, New York, 1st edn, 1973; (d) J. M. Tedder, Which factors determine the reactivity and regioselectivity of free radical substitution and addition reactions?, Angew. Chem., Int. Ed. Engl., 1982, 21, 401–410; (e) B. Giese, Syntheses with radicals-C, C bond formation via organotin and organomercury compounds, Angew. Chem., Int. Ed. Engl., 1985, 24, 553–565; (f) F. D. Vleeschouwer, V. V. Speybroeck, M. Waroquier, P. Geerlings and F. D. Proft, Electrophilicity and nucleophilicity index for radicals, Org. Lett., 2007, 9, 2721–2724.
3. B. P. Roberts, Polarity-reversal catalysis of hydrogen-atom abstraction reactions, Chem. Soc. Rev., 1999, 28, 25–35.
4. X. Wu, G. Xiao, Y. Ding, Y. Zhan, Y. Zhao, R. Chen and T.-P. Loh, Palladium-catalyzed intermolecular polarity-mismatched addition of unactivated alkyl radicals to unactivated alkenes, ACS Catal., 2020, 10, 14107–14116.
5. (a) X. Shen, C. Huang, X.-A. Yuan and S. Yu, Diastereoselective and stereodivergent synthesis of 2-cinnamylpyrrolines enabled by photoredox-catalyzed iminoalkenylation of alkenes, Angew. Chem., Int. Ed., 2021, 60, 9672–9679; (b) H. Chen, L. Guo and S. Yu, Primary, secondary, and tertiary γ-C (sp³)-H vinylation of amides via organic photoredox-catalyzed hydrogen atom transfer, Org. Lett., 2018, 20, 6255–6259.
6. S. Paul, D. Filippini and M. Silvi, Polarity transduction enables the formal electronically mismatched radical addition to alkenes, J. Am. Chem. Soc., 2023, 145, 2773–2778.
7. J. Brals, T. M. McGuire and A. J. B. Watson, A chemoselective polarity-mismatched photocatalytic C(sp³)-C(sp²) cross-coupling enabled by synergistic boron activation, Angew. Chem., 2023, 135, e202310462.
8. D. Troegel and J. Stohrer, Recent advances and actual challenges in late transition metal catalyzed hydrosilylation of olefins from an industrial point of view, Coord. Chem. Rev., 2011, 255, 1440–1459.
9. I. Fleming, A. Barbero and D. Walter, Stereochemical control in organic synthesis using silicon-containing compounds, Chem. Rev., 1997, 97, 2063–2192.
10. M. A. Brook, Silicon in Organic, Organometallic, and Polymer Chemistry, Wiley–Interscience, New York, 2000.
11. G. A. Showell and J. S. Mills, Chemistry challenges in lead optimization: silicon isosteres in drug discovery, Drug Discovery Today, 2003, 8, 551–516.
12. (a) S. Weber, M. Glavic, B. Stöger, E. Pittenauer, M. Podewitz, L.-F. Veiros and K. Kirchner, Manganese-catalyzed dehydrogenative silylation of alkenes following two parallel inner-sphere pathways, J. Am. Chem. Soc., 2021, 143, 17825–17832; (b) A. Banerjee, S. Sarkar, J. A. Shah, N. C. Frederiks, E. A. Bazan-Bergamino, C. J. Johnson and M.-Y. Ngai, Excited-state copper catalysis for the synthesis of heterocycles, Angew. Chem., Int. Ed., 2022, 61, e202113841; (c) L. Lu, J. C. Siu, Y. Lai and S. Lin, An electroreductive approach to radical silylation via the activation of strong Si-Cl bond, J. Am. Chem. Soc., 2020, 51, 21272–21278; (d) Q.-Q. Pan, L. Qi, X. Pang and X.-Z. Shu, Nickel-catalyzed cross-electrophile 1, 2-silyl-arylation of 1, 3-dienes with chlorosilanes and aryl bromides, Angew. Chem., Int. Ed., 2023, 62, e202215703; (e) S. Ye, K. Zhou, P. Rojsitthisak and J. Wu, Metal-free insertion of sulfur dioxide with aryl iodides under ultraviolet irradiation: direct access to sulfonated cyclic compounds, Org. Chem. Front., 2020, 7, 14–18; (f) G. Zhang, Y. Li, Y. Wang, Q. Zhang, T. Xiong and Q. Zhang, Asymmetric synthesis of silicon-stereogenic silanes by copper-catalyzed desymmetrizing protoboration of vinylsilanes, Angew. Chem., Int. Ed., 2020, 59, 11927–11931; (g) L. Zhang, Z. Hang and Z.-Q. Liu, A free-radical-promoted stereospecific decarboxylative silylation of α,β-unsaturated acids with silanes, Angew. Chem., Int. Ed., 2016, 55, 236–239; (h) X. Shang and Z.-Q. Liu, Recent developments in free-radical-promoted C–Si formation via selective C–H/Si–H functionalization, Org. Biomol. Chem., 2016, 14, 7829–7831.
13. L. Zhang, D. Liu and Z.-Q. Liu, A free radical cascade silylarylation of activated alkenes: highly selective activation of the Si-H/C-H bonds, Org. Lett., 2015, 17, 2534–2537.
14. C. Hu, C. Vo, R. R. Merchant, S.-J. Chen, J. M. E. Hughes, B. K. Peters and T. Qin, Uncanonical semireduction of quinolines and isoquinolines via regioselective HAT-promoted hydrosilylation, J. Am. Chem. Soc., 2023, 145, 25–31.
15. (a) X. Yang, Y. Xia, J. Tong, L. Ouyang, Y. Lai, R. Luo and J. Liao, Photoinduced radical cascade cyclization of acetylenic acid esters with oxime esters to access cyanalkylated coumarins, Org. Biomol. Chem., 2022, 20, 5239–5244; (b) J. Liao, X. Yang, L. Ouyang, Y. Lai, J. Huang and R. Luo, Recent advances in cascade radical cyclization of radical acceptors for the synthesis of carbo-and heterocycles, Org. Chem. Front., 2021, 8, 1345–1363; (c) J. Liao, L. Ouyang, Y. Lai and R. Luo, Photoredox-catalyzed oxy-/aminofluoroalkylative cyclization of alkenes, J. Org. Chem., 2020, 85, 5590–5597.
16. (a) J. Bonjoch and D. Solé, Synthesis of strychnine, Chem. Rev., 2000, 100, 3455–3482; (b) R. A. Bit, P. D. Davis, L. H. Elliott, W. Harris, C. H. Hill, E. Keech, H. Kumar, G. Lawton, A. Maw and J. S. Nixon, Inhibitors of protein kinase C. 3. Potent and highly selective bisindolylmaleimides by conformational restriction, J. Med. Chem., 1993, 36, 21–29; (c) G. A. Elmegeed, A. R. Baiuomy and O. M. Abdel-Salam, Evaluation of the anti-inflammatory and anti-nociceptive activities of novel synthesized melatonin analogues, Eur. J. Med. Chem., 2007, 42, 1285–1292; (d) Z. Ding and N. Yoshikai, Cobalt-catalyzed intramolecular olefin hydroarylation leading to dihydropyrroloindoles and tetrahydropyridoindoles, Angew. Chem., Int. Ed., 2013, 52, 8574–8578; (e) Y. Gao, J. Li, S. Bai, D. Tu, C. Yang, Z. Ye, B. Hu, X. Qi and C. Jiang, Direct synthesis of annulated indoles through palladium-catalyzed double alkylations, Org. Chem. Front., 2020, 7, 1149–1157.
17. CCDC 2264465 (3aa) contains the supplementary crystallographic data.†.
18. Y. Ge, Y. Tian, J. Wu, Q. Yan, L. Zheng, Y. Ren, J. Zhao and Z. Li, Iron-promoted free radical cascade difunctionalization of unsaturated benzamides with silanes, Chem. Commun., 2020, 56, 12656–12659.
19. A. Studer and D.-P. Curran, The electron is a catalyst, Nat. Chem., 2014, 6, 765–773.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2264465. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4qo01461h

---