Visible-Light Organophotoredox-Catalyzed Phosphonoalkylation of Alkenes via Deaminative Three-Component Radical–Radical Coupling

On the occasion of 75^th birthday of Prof. B. C. Ranu

Abstract

Here, we have reported a visible-light-mediated organic photoredox-catalyzed difunctionalization of vinyl arenes via radical–radical coupling. A stabilized benzylic radical at the α-position is generated via the regioselective addition of phosphonyl radical at the β-position of the styrene. Subsequently, benzyl or allyl radical, generated via the deaminative pathway from the Katritzky salt, combines with the α-radical of the styrene to furnish the functionalised C–P and C–C bonds in a single reaction.

Visible-light photoredox-catalyzed multicomponent coupling reactions for alkene difunctionalization is a powerful strategy to achieve molecular complexity and modularity from renewable feedstocks. In the past decades, transition-metal-catalyzed regioselective 1,1- or 1,2 difunctionalization of alkenes has been achieved via Heck reaction stabilizing transient metal–alkyl species, which is formed after the migratory alkene insertion.[1] [2] As an alternative, visible-light-mediated alkene difunctionalization under photoredox conditions via radical pathway and transition-metal catalysis is emerging. Indeed, photoredox/nickel dual catalysis has been shown by Molander and others to be an excellent tool for alkene 1,2-difunctionalizations, the electrophilic coupling partner was primarily restricted to sp²-halide complexes.[3] However, expensive organometallic photoredox catalysts and transition-metal catalysts and their contamination in the active pharmaceutical ingredients (APIs) are a serious concern. Another effective method for achieving olefin difunctionalization via ionic intermediates is the radical–polar crossover paradigm, which has recently been discovered by photocatalysis where carbon-centered electrophiles or nucleophiles have been utilized predominantly.[4] As a result, adopting a single, nondirectional strategy to include more straightforward, more universal carbon-based coupling partners across the olefinic double bond is still somewhat modest.

Alternatively, the radical–radical three-component cross-coupling approach to achieve alkene difunctionalization may have wider applications, but it is not frequently documented in the literature. This may be due to the deleterious two component radical–radical homo- and heterocoupling in all combinations with the reactants, and the regiochemical scrambling.[5] In this vein, olefin difunctionalization has been documented to occur via cross radical–radical interaction involving pyridinyl and NHC (N-heterocyclic carbene) attached ketyl radicals.[6] Hence, the judicious selection of the radical precursor is a key to overcome this problem to achieve alkene 1,2-difunctionalization product in a regiospecific manner. Phosphonate esters are particularly useful in sensors, antiviral, and antibacterial medications, herbicides, and organic phosphine ligands, among other applications.[7] As a result, there is a keen interest in the formation of C–P bonds under mild conditions.[8] Earlier, we reported that a phosphonyl radical is generated via aerial oxidation.[9] However, this process also can be accomplished by photoredox catalyst under anaerobic conditions. Furthermore, we have also demonstrated that a benzyl radical can be efficiently generated via a deamination pathway from the corresponding single-electron reduction under visible-light-mediated organophotoredox conditions for multicomponent reactions.[10] [11] From these experiences, we hypothesize a three-component alkene difunctionalization with phosphinyl and alkyl radicals. We also hypothesized that the phosphonyl radical would react with the styrenyl double bond at the β-styrene’s position, generating a stabilized α-benzylic radical. Subsequently, the benzyl or allyl radical, which will be generated via the deamination pathway from the Katritzky salt, will combine with the α-radical to furnish the functionalized C–P and C–C bonds in a single operation (Scheme [1]). The foundation of this hypothesis is based on the alkene functionalization processes, Heck- and Giese-type reactions, and photo- or metal-catalyzed cross-coupling reactions, which have been explored via deamination methods from pyridinium salts.[12] Interestingly, this method will enable access to an all-carbon quaternary center through phosphonic benzylation of commodity chemicals with regio- and chemoselectivity. Furthermore, contrary to the earlier reports, the alkyl radicals generated from the Katritzky salt are added at the α-position of the olefin instead of the β-position.[11] In addition, this protocol is redox-neutral and proceeds smoothly without additional oxidants or reductants.

Table 1 Optimization of the Reaction Conditions^a

Entry	Deviation from the optimized conditions	Yield of 4v (%)b
1	no variation	74
2	no PC	0
3	no light	0
4	Ir(ppy)2(dtbpy)PF6 instead of 4-CzIPN	66
5	Ru(bpy)3(PF6)2 instead of 4-CzIPN	43
6	Ir(ppy)3 instead of 4-CzIPN	12
7	0.5 mol% PC instead of 2.0 mol%	47
8	5.0 mol% PC instead of 2.0 mol%	72
9	without Cs2CO3	0
10	K2CO3 instead of Cs2CO3	31
11	CsF instead of Cs2CO3	71
12c	in aerial atmosphere	55

^a All reactions were carried out on a 0.2 mmol scale.

^b Yields referred to here are overall isolated yields.

^c Formation of 1,2-diphenylethan-1-one was observed.

We began our investigation using 1,1-diphenylethylene (1a), benzylamine-derived Katritzky salt (2a), and diphenylphosphine oxide (3a). Gratifyingly, using 1,2,3,5-tetrakis(carbazol-9-yl)-4,6-dicyanobenzene (4-CzIPN) as photocatalyst (PC), the expected phosphonic benzylation product 4v was obtained in 74% yield in acetonitrile as solvent and Cs₂CO₃ as the base under inert atmosphere (Table [1]). The cross-coupling product between the phosphonyl and benzyl radical was obtained when the reaction was carried out in an aerobic environment along with 4v in 55% yield (entry 12). No product was formed without the light source, or PC confirms the requirement for both components in the sequential creation of C–C and C–P bonds (entries 2 and 3). Among the photocatalysts analyzed, oxidizing catalysts such as Ir(ppy)₂(dtbpy)PF₆ or Ru(bpy)₃(PF₆)₂ showed moderate to lower performance, while reducing catalyst like Ir(ppy)₃ furnished trace product with almost complete recovery of styrene 1a (entries 4–6). Reducing the photocatalyst loading affects the reaction yield (47%), but raising the loading to 5% also fails to improve the yield (Table [1], entries 7, 8). No reaction took place in the absence of a base presumably responsible for the phosphoryl anion to initiate the reaction (entry 9). After identifying the appropriate reaction conditions, we investigated the newly established three-component photochemical paradigm’s substrate scope.

To create the densely functionalized products 4a–x (Scheme [2]), a variety of substituted vinyl arenes with diverse electrical natures are screened under optimized reaction conditions. Electron-donating or electron-neutral groups at the para position of 1,1-diarylethylenes performed very well under the reaction conditions (4h–l). The three-component reaction was substantially facilitated by electron-donating OMe substitution of diarylethylene at both para positions, yielding a 72% yield of the target product 4h.

Diarylethylene substituted with a chloro group also underwent well providing a chance for additional modifications via cross-coupling reactions (4i). Because the double phenyl group stabilized the α-radical, 1,1-diarylethylene behaved effectively under the reaction conditions. However, there is no such stability for monosubstituted styrene. But remarkably, simple styrenes also replaced 1,1-diphenylstyrenes, albeit in moderate yields (4m–p). Unfortunately, the presence of Br, CF₃, or heteroaryl styrenes (4z, 4aa, 4ab) prevents the formation of the product. Instead of the functionalization result, we obtained the hydrophosphorylation product in the presence of unactivated alkenes such as 4-phenyl-1-butene (4y).

The aryl moiety in the phosphine oxide counterpart can also be varied with a different functional group under mild reaction conditions, resulting in a good to excellent yield (65–90%) of the desired product (4a–g). For example, the electron-withdrawing fluoro substitution (4e) and the methoxy substitution at the para and meta positions (4c, 4d) were executed expertly. Notably, other phosphites (4f, 4g) that are typically nonreactive under photoredox conditions are also well-suited to go through the transformation.

We then shifted our focus to examine the range of amines. To accomplish effective functionalization, a variety of Katritzky salts were synthesized using variously substituted benzyl amines, following the standard procedure and then exposed to specific reaction conditions.[12d] The current approach allowed access to a wide range of para- and meta-substituted compounds, including halogen and methoxy substitution (4q–x). As a clear rationale, X-ray crystallographic characterization of 4q was performed (CCDC 2380247). The Katritzky salt synthesized from para-chloro benzyl amine produced the intended result in a reasonable yield (4t). To further extend the efficiency of our present technology, the allyl group containing the Katritzky salt was incorporated and a valuable all-C quaternary center was generated after successful difunctionalization in good yield (4s). To demonstrate the practical application of this present protocol, the reaction was carried out under direct renewable sunlight irradiation instead of blue LED irradiation. Enthusiastically, we found that a reaction powered by sunlight produced the desired transformation with an equivalent or occasionally higher yield than a reaction powered by a blue LED, making the reaction more sustainable and energy efficient. Further evidence of the viability of this green approach for upcoming industrial uses comes from a 1.0 mmol scale reaction carried out under direct sunlight irradiation that produced the required product 4v with a 67% (350 mg) yield (Scheme [3]).

To shed light on the mechanism, we have conducted some preliminary mechanistic experiments. A radical inhibition experiment using 2.0 equiv of the radical scavenger 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) allowed for complete recovery of the starting olefin and halted the process, indicating a likely radical mechanism (Scheme [4a]). In substitute of Katritzky salt, methanol-d ₄ was used in a control experiment to see whether a carbanion or carbocation intermediate was involved in the reaction. An ionic pathway was ruled out by the absence of deuterium or deuterated methanol in the product (Scheme [4b]). The product 7 was not formed when the standard reaction was carried out using the Katritzky salt 2ab, which is produced from cyclohexyl amines. This confirmed the necessity of the stabilized benzylic or allylic radical in the difunctionalization reaction (Scheme [4c]).

While it is possible that short-lived radical chains could arise, the light ON-OFF experiment’s results show that continuous light irradiation is required for the sequential synthesis of two subsequent C–C and C–P bonds (Scheme [4d]).

Stern–Volmer quenching studies were carried out to explore the reaction mechanism further. The results showed that the 4-CzIPN photocatalyst underwent fluorescence quenching when styrene 1 and phosphine 3 were present (Scheme [4e]) indicating the involvement of a single-electron-transfer pathway in this three-component coupling. From these control experiments and prior studies and precedence a plausible mechanism has been depicted in Scheme [4f]. The phosphonium anion reduces the excited state of the photocatalyst [4-CzIPN]* to create the phosphonium radical. A stable benzylic radical intermediate A, which is persistent in nature, is formed after the addition of reactive phosphonyl radical at the β-position of the styrenyl double bond. On the other hand, the reduced 4-CzIPN transfer one electron to the benzylic Katritzky salt (E _1/2red = –0.92 V vs SCE) by SET, which catalytically produces the benzyl radical intermediate regenerating the photocatalyst.[13] Finally, the benzyl radical and A undergo cross-radical coupling to furnish the desired three-component coupling product P regioselectively.

Here we disclose a novel visible-light-mediated photoredox-catalyzed, redox-neutral, three-component phosphonic benzylation of vinyl arenes. A phosphono radical is generated which adds to the styrene at the β-position to furnish a stabilized benzyl radical. Another benzyl radical is generated via deamination of a Katritzky salt and combines with styrenyl benzyl radical to furnish a dicarbofunctionalization product. Contrary to the earlier reports, the benzyl radical generated from the Katritzky salt combines with this incipient benzylic radical at the α-position, constructing C–P and C–C bonds simultaneously. Remarkably, all-carbon quaternary centers were generated from the corresponding 1,1-diaryl styrenes suppressing side products. We anticipate that this new radical–radical coupling protocol will allow chemists to perform previously challenging alkene difunctionalization reactions in a mild and sustainable way.

Conflict of Interest

The authors declare no conflict of interest.

Acknowledgement

We thank S. Laha and S. Kundu for the HRMS and NMR analysis. We also thank Bhaswati Paul from CSIR-Indian Institute of Chemical Biology for analyzing crystallography data. The authors thank Payel Khanra for helping in the Stern–Volmer experiment.

• References and Notes

• 1a Dorn SK, Brown MK. ACS Catal. 2022; 12: 2058
  CrossrefPubMedSearch in Google Scholar
• 1b Chen X, Xiao F, He W.-M. Org. Chem. Front. 2021; 8: 5206
  CrossrefPubMedSearch in Google Scholar
• 1c Barve BD, Kuo Y.-H, Li W.-T. Chem. Commun. 2021; 57: 12045
  CrossrefPubMedSearch in Google Scholar
• 2a Pan Q, Ping Y, Wang Y, Guo Y, Kong W. J. Am. Chem. Soc. 2021; 143: 10282
  CrossrefPubMedSearch in Google Scholar
• 2b Wickham LM, Dhungana RK, Giri R. ACS Omega 2022; 8: 1060
  CrossrefPubMedSearch in Google Scholar
• 2c Li Z.-L, Fang G.-C, Gu Q.-S, Liu X.-Y. Chem. Soc. Rev. 2020; 49: 32
  CrossrefPubMedSearch in Google Scholar
• 3a Xu S, Chen H, Zhou Z, Kong W. Angew. Chem. Int. Ed. 2021; 60: 7405
  CrossrefPubMedSearch in Google Scholar
• 3b Campbell MW, Yuan M, Polites VC, Gutierrez O, Molander GA. J. Am. Chem. Soc. 2021; 143: 3901
  CrossrefPubMedSearch in Google Scholar
• 3c Qi X, Diao T. ACS Catal. 2020; 10: 8542
  CrossrefPubMedSearch in Google Scholar
• 3d Badir SO, Molander GA. Chem. 2020; 6: 1327
  CrossrefPubMedSearch in Google Scholar
• 4a Cabrera-Afonso MJ, Sookezian A, Badir SO, Khatib ME, Molander GA. Chem Sci. 2021; 12: 9189
  CrossrefPubMedSearch in Google Scholar
• 4b Tang H.-J, Zhang B, Xue F, Feng C. Org. Lett. 2021; 23: 4040
  CrossrefPubMedSearch in Google Scholar
• 4c Wang S, Cheng B.-Y, Sršen M, König B. J. Am. Chem. Soc. 2020; 142: 7524
  CrossrefPubMedSearch in Google Scholar
• 5 Yip BR. P, Pal KB, Lin JD, Xu Y, Das M, Lee J, Liu X.-W. Chem. Commun. 2021; 57: 10783
  CrossrefPubMedSearch in Google Scholar
• 6a Mathi GR, Jeong Y, Moon Y, Hong S. Angew. Chem. Int. Ed. 2019; 59: 2049
  CrossrefPubMedSearch in Google Scholar
• 6b Ishii T, Ota K, Nagao K, Ohmiya H. J. Am. Chem. Soc. 2019; 141: 14073
  CrossrefPubMedSearch in Google Scholar
• 6c Chen D, Xu L, Long T, Zhu S, Yang J, Chu L. Chem Sci. 2018; 9: 9012
  CrossrefPubMedSearch in Google Scholar
• 7 Krečmerová M, Majer P, Rais R, Slusher BS. Front. Chem. 2022; 10: 889737
  CrossrefPubMedSearch in Google Scholar
• 8 Fu Q, Bo ZY, Ye JH, Ju T, Huang H, Liao LL, Yu DG. Nat. Commun. 2019;  10: 3592
  CrossrefPubMedSearch in Google Scholar
• 9 Bhunia SK, Das P, Jana R. Org. Biomol. Chem. 2018;  16: 9243
  CrossrefPubMedSearch in Google Scholar
• 10 Das S, Rahaman SA, Pradhan K, Jana R. Org. Lett. 2024; 26: 6955
  CrossrefPubMedSearch in Google Scholar
• 11 Nandi S, Das P, Das S, Mondal S, Jana R. Green Chem. 2023; 25: 3633
  CrossrefSearch in Google Scholar
• 12a Baker KM, Tallon A, Loach RP, Bercher OP, Perry MA, Watson MP. Org. Lett. 2021; 23: 7735
  CrossrefPubMedSearch in Google Scholar
• 12b Wang J, Hoerrner ME, Watson MP, Weix D. J. Angew. Chem. Int. Ed. 2020; 59: 13484
  CrossrefPubMedSearch in Google Scholar
• 12c Wu J, Grant PS, Li X, Noble A, Aggarwal VK. Angew. Chem. Int. Ed. 2019; 58: 5697
  CrossrefPubMedSearch in Google Scholar
• 12d Basch CH, Liao J, Xu J, Piane JJ, Watson MP. J. Am. Chem. Soc. 2017;  139: 5313
  CrossrefPubMedSearch in Google Scholar
• 13 Correia JT. M, Fernandes VA, Matsuo BT, Delgado JA. C, De Souza WC, Weber PaixãoM. Chem. Commun. 2020; 56: 503
  CrossrefPubMedSearch in Google Scholar

Corresponding Author

Publication History

Received: 30 August 2024

Accepted after revision: 30 September 2024

Accepted Manuscript online:
30 September 2024

Article published online:
22 October 2024

© 2024. Thieme. All rights reserved

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany

• References and Notes

• 1a Dorn SK, Brown MK. ACS Catal. 2022; 12: 2058
  CrossrefPubMedSearch in Google Scholar
• 1b Chen X, Xiao F, He W.-M. Org. Chem. Front. 2021; 8: 5206
  CrossrefPubMedSearch in Google Scholar
• 1c Barve BD, Kuo Y.-H, Li W.-T. Chem. Commun. 2021; 57: 12045
  CrossrefPubMedSearch in Google Scholar
• 2a Pan Q, Ping Y, Wang Y, Guo Y, Kong W. J. Am. Chem. Soc. 2021; 143: 10282
  CrossrefPubMedSearch in Google Scholar
• 2b Wickham LM, Dhungana RK, Giri R. ACS Omega 2022; 8: 1060
  CrossrefPubMedSearch in Google Scholar
• 2c Li Z.-L, Fang G.-C, Gu Q.-S, Liu X.-Y. Chem. Soc. Rev. 2020; 49: 32
  CrossrefPubMedSearch in Google Scholar
• 3a Xu S, Chen H, Zhou Z, Kong W. Angew. Chem. Int. Ed. 2021; 60: 7405
  CrossrefPubMedSearch in Google Scholar
• 3b Campbell MW, Yuan M, Polites VC, Gutierrez O, Molander GA. J. Am. Chem. Soc. 2021; 143: 3901
  CrossrefPubMedSearch in Google Scholar
• 3c Qi X, Diao T. ACS Catal. 2020; 10: 8542
  CrossrefPubMedSearch in Google Scholar
• 3d Badir SO, Molander GA. Chem. 2020; 6: 1327
  CrossrefPubMedSearch in Google Scholar
• 4a Cabrera-Afonso MJ, Sookezian A, Badir SO, Khatib ME, Molander GA. Chem Sci. 2021; 12: 9189
  CrossrefPubMedSearch in Google Scholar
• 4b Tang H.-J, Zhang B, Xue F, Feng C. Org. Lett. 2021; 23: 4040
  CrossrefPubMedSearch in Google Scholar
• 4c Wang S, Cheng B.-Y, Sršen M, König B. J. Am. Chem. Soc. 2020; 142: 7524
  CrossrefPubMedSearch in Google Scholar
• 5 Yip BR. P, Pal KB, Lin JD, Xu Y, Das M, Lee J, Liu X.-W. Chem. Commun. 2021; 57: 10783
  CrossrefPubMedSearch in Google Scholar
• 6a Mathi GR, Jeong Y, Moon Y, Hong S. Angew. Chem. Int. Ed. 2019; 59: 2049
  CrossrefPubMedSearch in Google Scholar
• 6b Ishii T, Ota K, Nagao K, Ohmiya H. J. Am. Chem. Soc. 2019; 141: 14073
  CrossrefPubMedSearch in Google Scholar
• 6c Chen D, Xu L, Long T, Zhu S, Yang J, Chu L. Chem Sci. 2018; 9: 9012
  CrossrefPubMedSearch in Google Scholar
• 7 Krečmerová M, Majer P, Rais R, Slusher BS. Front. Chem. 2022; 10: 889737
  CrossrefPubMedSearch in Google Scholar
• 8 Fu Q, Bo ZY, Ye JH, Ju T, Huang H, Liao LL, Yu DG. Nat. Commun. 2019;  10: 3592
  CrossrefPubMedSearch in Google Scholar
• 9 Bhunia SK, Das P, Jana R. Org. Biomol. Chem. 2018;  16: 9243
  CrossrefPubMedSearch in Google Scholar
• 10 Das S, Rahaman SA, Pradhan K, Jana R. Org. Lett. 2024; 26: 6955
  CrossrefPubMedSearch in Google Scholar
• 11 Nandi S, Das P, Das S, Mondal S, Jana R. Green Chem. 2023; 25: 3633
  CrossrefSearch in Google Scholar
• 12a Baker KM, Tallon A, Loach RP, Bercher OP, Perry MA, Watson MP. Org. Lett. 2021; 23: 7735
  CrossrefPubMedSearch in Google Scholar
• 12b Wang J, Hoerrner ME, Watson MP, Weix D. J. Angew. Chem. Int. Ed. 2020; 59: 13484
  CrossrefPubMedSearch in Google Scholar
• 12c Wu J, Grant PS, Li X, Noble A, Aggarwal VK. Angew. Chem. Int. Ed. 2019; 58: 5697
  CrossrefPubMedSearch in Google Scholar
• 12d Basch CH, Liao J, Xu J, Piane JJ, Watson MP. J. Am. Chem. Soc. 2017;  139: 5313
  CrossrefPubMedSearch in Google Scholar
• 13 Correia JT. M, Fernandes VA, Matsuo BT, Delgado JA. C, De Souza WC, Weber PaixãoM. Chem. Commun. 2020; 56: 503
  CrossrefPubMedSearch in Google Scholar

• Supplementary Material