Visible-light-mediated β-acylative divergent alkene difunctionalization with Katritzky salt/CO₂

Abstract

Multicomponent cross-coupling reactions involving alkenes under visible-light photoredox catalysis have tremendous potential to achieve molecular complexity and modularity from renewable feedstocks in a sustainable manner. Here we disclose two similar yet mechanistically distinct visible-light mediated photoredox-catalyzed, redox-neutral, regioselective dicarbofunctionalization reactions of vinyl arenes through three-component coupling. Acylative carbobenzylation and carbocarboxylation have been achieved successfully where the acyl radical is generated via decarboxylation from the ketocarboxylic acid and undergoes addition regioselectively at the β-position of the vinyl arene to generate a stabilized benzylic radical as a common intermediate. Subsequently, this incipient benzyl radical undergoes two distinct pathways: (a) radical–radical cross-coupling with the benzyl radical which is generated from a Katritzky salt to realize acylative benzylation; (b) reductive radical–polar crossover to react with CO₂ (1.0 atm) for acylative carboxylation. In contrast to the previous reports, the acyl radical undergoes addition regioselectively at the β-position of the vinyl arene and deaminative benzylation takes place at the α-position. Remarkably, an all carbon quaternary center is generated from the corresponding 1,1-diaryl styrenes bypassing all other undesired pathways. This unprecedented acylative carboxylation of alkenes afforded a γ-keto carboxylic acid which is a precursor of γ-amino acids. The reactions are reproducible under sunlight irradiation on a gram scale, which is a step forward in sustainable development.

---

Introduction

Visible light mediated multicomponent reactions involving alkenes are an upcoming research field for pot, atom, and step economic (PASE) organic synthesis.¹ The vicinal difunctionalization of alkenes has emerged as a resourceful synthetic approach to access densely functionalized molecular frameworks.² In the past decade, extensive research has been successfully carried out categorically on hydrofunctionalization or difunctionalization including minimum one C–X (X = heteroatom) bond formation using transition metal or transition metal–photoredox dual catalytic systems.³ Installation of two carbon subunits through the C–C double bond could rapidly increase the molecular complexity. In this line, regioselective dicarbofunctionalization has been an attractive synthetic challenge to organic chemists in recent years.⁴ In the established methods of dicarbofunctionalization, inclusion of transition metals generates the inherent problems of β-hydride elimination, homocoupling, isomerization, or proto-demetalation. Though Molander and other groups have established photoredox/nickel dual-catalysis as a remarkable tool for alkene 1,2-difunctionalizations, the electrophilic coupling partner is mainly limited to sp²-halide systems.^3a,e,f,5 Recently, a radical–polar crossover paradigm solely by photocatalysis has emerged as another potent tool to accomplish olefin difunctionalization via ionic intermediates.^3c,6 In this approach, generally a radical species is added to one end of the C–C double bond of olefins that gives rise to the formation of a stabilized radical at the other end which reacts in the subsequent steps via ionic intermediates. However, radical–radical cross-coupling to realize alkene dicarbofunctionalization may potentially show broader application but is rarely reported in the literature. We aimed to attain β-acylative carbofunctionalization of alkenes using a modular approach to achieve different functionalizations at the α-position in both ways. The challenges of selective radical–radical cross-coupling lie in the prevention of plentiful chances of two-component radical–radical homo and hetero-coupling as competitive facile reactions and the low possibility of regioisomeric product formation.⁷ For these reasons, so far, only pyridinyl radicals and NHC (N-heterocyclic carbene)-attached ketyl radicals have been reported to undergo cross radical–radical coupling to accomplish olefin difunctionalization with alkylation at the β-position (Scheme 1a).⁸ To overcome these issues, the decisive task is the identification of appropriate radical precursors which will add across the olefinic double bond in a regiospecific manner. With this aim, we targeted the acyl radical to be the primary radical component to attach to the C–C double bond. The acyl radical is a reactive nucleophilic radical with an established propensity to add to the olefinic double bond and it can be accessed from an α-ketocarboxylic acid via single-electron transfer of the corresponding carboxylate by photocatalytic oxidation and subsequent decarboxylation.⁹ However, most of the reactions earlier provided α-acylation of alkenes with acyl radicals generated from acyl chlorides or aldehydes (Scheme 1b).¹⁰ We anticipated that the addition of the acyl radical to the styrenyl double bond at the β-position would generate the key radical intermediate which could attach to another carbon subunit either via radical–radical coupling or radical–polar crossover under appropriate conditions. We envisioned that in the presence of a shorter-lived transient radical in an excess amount in the reaction medium, the cross radical–radical coupling would dominate due to the persistent radical effect and the likelihood of competitive two-component coupling by-products would therefore be low.¹¹ Recently, an attractive strategy to generate alkyl radicals has been developed by Watson and other groups by activating the C–N bonds of abundant amines through Katritzky salt formation.¹² The deaminative protocols from pyridinium salts have been explored by photo- or metal-catalysed cross coupling reactions, Heck and Giese type reactions and alkene difunctionalization reaction.¹³ We anticipated that pyridinium salt derived from benzyl amines could serve as the source of benzyl radicals (transient but stabilized), which may suitably couple with our radical intermediate to accomplish the goal.¹⁴ As the reactivity of the acyl radical is much higher than that of the benzyl radical, the acyl radical is shorter-lived. As a result, the acyl radical should attack the double bond as soon as it is generated^11,15 and the stability of the benzyl radical is optimum to undergo a subsequent radical–radical coupling.¹¹ Notably, in all the literature reports, the alkyl radicals generated from the Katritzky salt typically add to the β-position of the olefinic double bond to undergo the desired transformations.¹⁶ By judiciously choosing the acyl radical as the reactive partner, we can perform simple benzylation at the α-position of styrenes, which was not known earlier. In contrast, we were also intrigued to explore the possibility of acylative dicarbofunctionalization using the radical–polar crossover protocol. We hypothesized that in the absence of any suitable radical partner, the key radical intermediate could undergo SET reduction under photocatalytic conditions and then react with an electrophile.

Scheme 1 Photochemical dicarbofunctionalization of alkenes.

Lately, owing to the increasing concentration of CO₂ in the atmosphere, CO₂ capture and utilization (CCU) has become an emergent and urgent research area as designated by the environment protection agencies (EPA) and G20 summit.¹⁷ Despite being feebly electrophilic, due to the sustainable, abundant, low-cost, and nontoxic nature, profound research has been directed towards the photocatalytic activation/utilization of CO₂ as an ideal one-carbon (C1) building block in fine and bulk chemical synthesis. Mimicking the nature's protocol of photosynthesis using sunlight and CO₂, photocatalytic carboxylation reactions using CO₂ have tremendous potential for “waste-to-wealth” generation.¹⁸ Typically, CO₂ and alkenes are petroleum-cracking products produced in tonnes from the petroleum industry. Before releasing to the environment, converting these so-called waste materials into value-added products is an urgent environmental, social, and economic need. Consequently, visible-light-induced regioselective silacarboxyaltion, phosphonocarboxylation and thiocarboxylation of alkenes have been developed with CO₂ and various radical precursors by the groups of Martin, Yu, Wu, and Xi.¹⁹ Still, the scope of carbocarboxylation is small as per the literature (Scheme 1c).²⁰ To achieve this, one interesting strategy would be decarboxylation of carboxylic acids followed by addition to an alkene and subsequent carboxylation with CO₂. Recently, in this manner, the Yu group and Sun group separately utilized amino acids as a precursor of the alkyl-radical for alkylative carboxylation of alkenes with or without external carbon dioxide using metal photocatalysts.²¹ Later, the Sun group achieved acylative carboxylation using acyclic ketone oxime esters as acyl radical precursors by Ir photocatalysis (Scheme 1c).²² However, simultaneous decarboxylative acylation and carboxylation of alkenes with carbon dioxide has never been achieved. While Yu^21a efficiently utilized tert-alkyl ketocarboxylic acids as the tert-alkyl radical source via successive decarboxylation and decarbonylation to give alkylative carboxylation, no more such effort has been devoted to realize acylative carboxylation using aryl/1° alkyl ketocarboxylic acids as the acyl radical source. Continuing with our constant effort to develop carboxylation reactions and using carboxylic acids as synthetic intermediates,²³ we were intrigued to take the challenge. If successful, we could switch between two apparently similar but characteristically distinct mechanistic pathways and thus make a move from carbobenzylation to carbocarboxylation/carboxyacylation from commodity chemicals with regio- and chemoselective control to generate a valuable all-carbon quaternary centre (Scheme 1d). In both cases, despite previous developments of alkene difunctionalization, the challenges remaining are: (1) achieving β-acylation unanimously, (2) achieving α-alkylation using a Kartritzky salt, (3) organophotocatalysis to achieve decarboxylation and carboxylation in the same reaction, and (4) acylative carboxylation of alkenes. Notably, under a CO₂ atmosphere, to achieve decarboxylation satisfactorily and finally to achieve carboxylation, the system would have to behave against “Le Chatelier's principle”,²⁴ which has been achieved here under transition-metal free mild photocatalytic conditions. Remarkably, the substrates and intermediate of these protocols are redox active and are both oxidized or reduced by the SET process during the redox-neutral reaction mechanism avoiding the necessity for stoichiometric external oxidants or reductants, the high-value byproduct is easily recovered, sunlight irradiation serves the satisfactory role – which is a step forward in sustainable development, solvents are permissible, and the reactions are scalable.

Results and discussion

We optimized our reaction conditions taking 1,1-diphenylethylene 1a, 4-methoxyphenylglyoxalic acid 2a as model substrates (Table 1). To check the viability of the hypothesis of radical–radical coupling leading to carbobenzylation, Katritzky salt 3a, derived from benzyl amine, was taken. We observed the formation of the anticipated acylative benzylation product 4a regiospecifically only in 12% yield in acetonitrile solvent using Ir(ppy)₃ (E_1/2(PC*/PC˙⁻) = −0.31 V vs. SCE)²⁵ as the photocatalyst (PC) and Cs₂CO₃ as the base in MeCN solvent (entry 1). As the glyoxylic acid 2a has E^red_1/2(acid/acyl radical) = −1.00 V vs. SCE, and 3a has E^red_1/2(3a/Ph˙) = −0.92 V vs. SCE, we anticipated that the use of a photocatalyst (PC) having slightly lower negative E^red_1/2(PC*/PC˙⁻) value than −1.00 V (vs. SCE) as well as having higher negative E^red_1/2(PC˙⁻/PC) value than −0.92 V (vs. SCE) would serve the purpose. Similarly, when we varied the photocatalysts, 4-CzIPN, Ir(ppy)₂(dtbpy)PF₆, Ru(bpy)₃Cl₂·6H₂O improved the yield of 4a (entries 2–4). After thorough screening of other different reaction parameters along with the photocatalyst, we found that the desired dicarbofunctionalization was best to provide the product 4a exclusively in 82% yield within 2 hours upon irradiation with a 5 W blue LED (λ_max = 455 nm) in the presence of only 1 mol% of Ru(bpy)₃Cl₂·6H₂O as the photocatalyst and Cs₂CO₃ as the base in acetonitrile solvent under inert atmosphere (entry 4), which is denoted as “Condition A”. During further optimization, when the reaction was executed under aerobic conditions, the coupling product between benzyl and benzoyl radicals was observed and the yield of the product decreased to 55% (entry 6). Control experiments conducted by omitting the light source or photocatalyst resulted in no product formation which proves the necessity of both light and photocatalyst in the sequential C–C bond forming transformation (entries 8 and 9). The product formation was completely suppressed in the absence of the base, apparently because of the fact that decarboxylative acyl radical formation gets hampered without a base (entry 7). Due to the larger size of cesium, it may form loosely bound ion pairs with 3a promoting the decarboxylation step. Therefore, cesium-based bases offered optimal reactivity towards product yield compared to Na or K bases (entry 5 and Table S3†). Use of Lewis acids like Cu(OTf)₂ or In(OTf)₂ to stabilise the radical coupling partners actually ended up in a diminished yield of the difunctionalized product (please see the ESI† for details).

Table 1 Optimization of the reaction conditions^a,b

Entry	PC (photocatalyst)	Base	Yield^b (%) (4a or 5a)
a All reactions were carried out in 0.2 mmol scale. b Yields refer to here are overall isolated yields. c Under air atmosphere, formation of 1,2-diphenylethan-1-one was observed. d At dark. e MeCN as solvent. f Under an Ar atmosphere. g With 6.0 equiv. of 2a, under an Ar atmosphere. h Under O2 atmosphere.
X = 3a
1	Ir(ppy)3	Cs2CO3	12
2	Ir(ppy)2(dtbpy)PF6	Cs2CO3	66
3	4-CzIPN	Cs2CO3	43
4	Ru(bpy) 3 Cl 2 ·6H 2 O	Cs 2 CO 3	82
5	Ru(bpy)3Cl2·6H2O	Cs2CO3	31
6^c	Ru(bpy)3Cl2·6H2O	Cs2CO3	55
7	Ru(bpy)3Cl2·6H2O	Cs2CO3	0
8	—	Cs2CO3	0
9^d	Ru(bpy)3Cl2·6H2O	Cs2CO3	0
X = CO 2 (1 atm)
10^e	Ru(bpy)3Cl2·6H2O	Cs2CO3	16
11	Ir(ppy)2(dtbpy)PF6	Cs2CO3	70
12	4-CzIPN	Cs 2 CO 3	81
13	4-CzIPN	LiCl	74
14^d	4-CzIPN	Cs2CO3	0
15	—	Cs2CO3	0
16^f	4-CzIPN	Cs2CO3	30
17^g	4-CzIPN	Cs2CO3	58
18^h	4-CzIPN	Cs2CO3	0

---

After successfully accomplishing the radical–radical coupling for the acylative benzylation of olefin, we turned our attention to satisfy our second hypothesis for obtaining acylative carboxylation via an ionic intermediate. According to our assumption, we omitted the addition of Kartizky salt 3a for the elimination of any potential radical coupling agent in the reaction medium. Therefore, when a mixture of 1a and 2a was subjected to the same reaction conditions with a CO₂ balloon, we were delighted to observe the formation of the desired product 5a in 16% yield (entry 10).

After screening several solvents, polar solvent DMSO provided the best result (Table S5†). Here, during carbocarboxylation, after accomplishing decarboxylation, a photocatalyst would be required to reduce the incipient benzyl radical to produce a benzylic carbanion for reductive coupling with slightly electrophilic CO₂. On varying the photocatalysts, Ir(ppy)₂(dtbpy)PF₆ having E_1/2(Ir^III/Ir^II) = −0.31 V vs. SCE, 4-CzIPN having E_1/2(PC˙⁻/PC) = −0.31 V vs. SCE^25a provided improved yield (entries 11 and 12, see the ESI† for detailed optimization). On further screening of several photocatalysts, bases and solvents, it was found that irradiation with 455 nm blue LEDs for 12 h under CO₂ balloon pressure with 1.0 equivalent of 2a in the presence of 1 mol% of 4-CzIPN in DMSO solvent furnished the desired carboxylated product 5a in 81% yield (entry 12). Remarkably, in place of expensive Cs₂CO₃, the non-carbonated base LiCl furnished a comparable yield (entry 13). No product was detected in the absence of light and photocatalyst (entries 14 and 15). With an ambition to reutilize the released CO₂ from the α-ketocarboxylic acid, the reaction was performed under argon without external CO₂ furnishing 30% yield (entry 16). The partial pressure of P_CO2 was sufficiently increased to afford 58% of the desired product using 6.0 equiv. of 2a (entry 17) without the addition of external CO₂. However, use of super-stoichiometric amount of 2a was avoided in the present investigation to minimize waste production. Reaction under an O₂ atmosphere completely prevented the formation of 5a and the phenylglyoxylic acid converted to deleterious benzoic acid rapidly (entry 18). As entry 12 provided the best result, we decided to proceed with it as the optimized conditions denoted as “Condition B”. As the final reaction mixture contains remaining 2a, product 5a and side product benzoic acid, for better isolation, in situ esterification was performed in most of the cases by the addition of MeI into the same reaction mixture at the later stage.

Having identified the optimized reaction conditions, we performed the substrate scope of the newly developed three-component photochemical paradigm to achieve carbobenzylation (Table 2). An array of differently substituted 1,1-diarylethylenes with diverse electronic nature is susceptible to the mild reaction conditions of “Condition A” to form the densely functionalized product (4a–4n). Electron-donating or electron-neutral groups at the para position of 1,1-diarylethylenes performed very well under the reaction conditions (4a–4g). The representative structure of 4d was unambiguously characterized by X-ray crystallography (CCDC 2160747†). It is noteworthy that para-O-propargyl substituted diphenylethylene underwent chemoselective transformation smoothly leaving the triple bond intact (4f). Electron withdrawing ester substitution at the para position of the styrene substantially hampered the three-component reaction providing 36% yield of the desired product (4h). Chloro substituted diarylethylene also survived well providing an opportunity for further manipulations through cross-coupling reactions (4i). A Boc-protected amine group afforded the desired product in acceptable yield (4j). An adamantyl carboxylic acid ester protected styrene rendered the product in 67% yield (4k). Remarkably, simple styrenes instead of 1,1-diphenylethylenes were also capable of the transformation albeit in moderate yields (4l–4n). The reaction is amenable with a wide range of α-keto acids containing electron-neutral, electron donating and halogen substitution, furnishing good yields of the resulting polyaromatic carbon frameworks (4o–4z). 2-Chloro and 2,4-dimethyl substituted phenyl glyoxylic acids (4u and 4v) along with 2-naphthyl and 2-thiophenyl ketocarboxylic acid afforded the desired acylative benzylation in very good yield (4w and 4x). Importantly, 2-(methyl(phenyl)amino)-2-oxoacetic acid was capable of undergoing the transformation providing 60% yield of the desired amide-incorporated difunctionalised product (4y). Aliphatic pyruvic acid was also well-suited to undergo the decarboxylative transformation providing 56% product yield without further decarbonylation of the acyl radical (4z).

Table 2 Substrate scope of carbobenzylation^a,b

a All reactions were carried out in 0.2 mmol scale. b Yields refer to the overall isolated yields.

---

Next, a range of Katritzky salts was prepared from diversely substituted benzyl amines and subjected to the reaction conditions to achieve an efficient dicarbofunctionalization. A variety of para and meta substituted products were accessed by the present protocol including methoxy and halogen substitution (4aa–4ai). Ortho-Trifluoromethoxy benzyl amine derived Katrizky salts afforded the desired product in a moderate yield (4ag). Pyridine and allyl containing valuable an all-C quaternary centre was successfully generated by the present methodology although in lower yields (4ah and 4ai). Notably, in all these examples, the by-product 2,4,6-triphenylpyridine was recovered in quantitative amounts, which is a valuable compound for its limited synthetic methods and an important substrate for other organic transformations (Table 2).²⁶

Thereafter, with the acceptable reaction conditions for carboxyacylation/carbocarboxylation in hand i.e., under “Condition B”, generality of the substrates was evaluated (Table 3). Substituted aryl α-keto carboxylic acids containing electron-donating groups like Me, OMe, ⁱBu, Ph (5a–5d, 5h, 5o, 5q, 5u–5y) and withdrawing groups like F, Cl, Br (5f, 5g) provided good to moderate yields under the optimized reaction conditions. The formation of the said product was confirmed by X-ray structure of 5b′ (CCDC 2161689†) derived from 5b on hydrolysis. Similarly, while the styrene part was varied, α-substituted styrene having electron donating substitution such as OBn, Me, OMe (5j–5k, 5n–5q) and withdrawing substitution as Cl, CO₂Me (5m, 5s–5t) underwent the reaction smoothly. Interestingly, heterocycle 2-oxo-2-(thiophen-2-yl)acetic acid furnished the desired product in high yield (5r, 5t). To our delight, the styrenes without any α-substitution provided fair yields where both donating groups as Me, ^tBu (5w, 5x) and withdrawing group CN (5z) survived. Naphthyl styrene also provided satisfactory results (5aa, 5ab) under the reaction conditions. Interestingly, easily hydrolysable allyloxy and propargyloxy groups were well-tolerated (5p, 5u) in this mild condition. To expand the scope beyond substituted phenylglyoxylic acids, the conditions were applied to 2-(methyl(phenyl)amino)-2-oxoacetic acid (5ac). And notably, the aliphatic ketocarboxylic acids provided only the expected products (5ad–5af) via decarboxylation. Further decarbonylation to furnish alkyl radicals and subsequent reaction manifold was not observed under these reaction conditions.

Table 3 Substrate scope of decarboxylative acylcarboxylation of alkene^a,b

a All reactions were carried out in 0.2 mmol scale. b Yields refer to the overall isolated yields.

---

In order to demonstrate the practical utility of the method, we tested the reaction under sunlight irradiation. To our delight, when blue LED was replaced in “Condition B” by sunlight irradiation (22.57° N, 88.36° E at 1100 h), the standard reaction between 1a, 2a and CO₂ provided the expected product in 60% yield within 5 h. To check the generality of this method, several substrates were made to react under this condition. As shown in the Table 3, both the donating and withdrawing groups survived the reaction conditions yielding the expected product satisfactorily (5a–d, 5p, 5q). Heterocyclic moiety was also well-tolerated (5r, 5ag). Allyloxy and adamantyl carboxylate substituted styrenes provided high yields under sunlight irradiation (5p, 5ah–5ai).

Similarly, instead of blue LED, the carbobenzylation reaction was also performed under solar irradiation. Delightfully, we observed that the sunlight-driven reaction afforded the desired transformation with a similar yield compared to blue LED which makes the reaction energy-efficient and more sustainable. Furthermore, a 3.16 mmol scale reaction, performed under direct sunlight irradiation furnishing the desired product 4d with 61% (840 mg) yield validates the potential of this green protocol for future industrial applications (Scheme 2a). To demonstrate the practical utility, further, the acyl keto product 5b was reduced by NaBH₄in situ to afford a 2° alcohol, which spontaneously reacts with the carboxylate group to furnish the hindered lactone 6b in 61% yield (Scheme 2b).

Scheme 2 Practical utility.

To shed light on the mechanism, we conducted several control experiments (Scheme 3). Radical inhibition experiment with 2.0 equiv. of radical scavenger (2,2,6,6-tetramethylpiperidin-1-yl)oxyl (TEMPO) completely shut down both reactions with full recovery of the starting olefin suggesting that a probable radical mechanism is operating in each case (Scheme 3a). Radical addition followed by ring opening occurs with olefin 1aa in the radical-clock experiment which indicates the generation of benzylic radical intermediate from styrene for both carbobenzylation and carbocarboxylation to give 7 (Scheme 3b). Further no radical–radical coupling^18e between benzylic radical and CO₂ took place indicating a radical–polar crossover in the carboxylation process. During carbobenzylation, chalcone Michael acceptor 8 may be generated as an intermediate via a base-promoted acylation-elimination pathway and 1,4 addition of benzyl radical may also furnish the desired product 4. Hence, chalcone 8 was prepared independently and subjected to the reaction with the Katritzky salt 3a. But no desired product (4n) was observed under standard reaction conditions with or without ketoacid 2a, ruling out the possibility of the Michael addition of the benzyl radical (Scheme 3c). Next, the possibility of carboxylation by the carbonate base²⁷ was ruled out as LiCl instead of Cs₂CO₃ also furnished a comparable yield (entry 13, Table 1). To clearly understand the carboxylating agent, we performed the isopope-labeling experiment. We performed a reaction with ¹³CO₂ taking 2-vinylnaphthalene 1ab and 4-methylphenylglyoxylic acid 2d as formal substrates which ended up in the formation of [¹³C]-5ab with 80% ¹³C incorporation (Scheme 3d). This provides strong evidence that the carboxyl group in the desired product originated from the external CO₂, although a portion might be contributed by the glyoxylic acid as per the observation in the absence of external CO₂ (entries 16 and 17, Table 1; Scheme 3e). Light on–off experiment for both of the reactions suggests that continuous light irradiation is required for the sequential C–C bond formation although the possibility of short-lived radical chains cannot be excluded (Scheme 3f and see Section 8.6 in ESI†).

Scheme 3 Control experiments.

Based on the control experiments and previous literature precedence^25,28 a probable mechanism has been portrayed in Scheme 4. The excited state of the photocatalyst (under Condition A, [Ru(II)]*: E^red_1/2 = 0.77 V vs. SCE or under Condition B, 4-CzIPN*: E^red_1/2 = 1.35 V vs. SCE) undergoes reduction by the carboxylate anion (E^ox_1/2 ≈ +1.0 V vs. SCE) to produce acyl radical via decarboxylation.²⁵ The reactive acyl radical selectively adds to the terminal position of the styrenyl double bond to form the more stable benzylic radical intermediate A which is persistent in nature. Now, under Condition A, during carbobenzylation, the benzylic Katritzky salt 3a (E^red_1/2 = −0.92 V vs. SCE) takes up one electron from the reduced Ru(I) (E_1/2 = −1.33 V vs. SCE) via SET to produce the benzyl radical intermediate catalytically with the elimination of 2,4,6-triphenylpyridine to regenerate the photocatalyst.²⁹ The benzyl radical undergoes cross radical–radical coupling with A, leading to regiospecific product 4. On the other hand, in the absence of 3a, under Condition B, SET could take place between A and reduced photocatalyst (PC˙⁻) (E^red_1/2 = −1.21 V vs. SCE) to generate carbanion B.³⁰ Subsequently, it attacks to slightly electrophilic CO₂ to deliver the carboxylated product C, which on methylation or protonation, furnishes the expected product 5.

Scheme 4 Plausible divergent catalytic cycle.

Conclusions

In conclusion, we have developed a divergent strategy for visible light mediated, regioselective acylative-benzylation and acylative-carboxylation with CO₂ of styrenes. We have achieved previously unprecedented selective benzylation at the α-position and acylation at the β-position of styrenes. Remarkably, the acylative carboxylation reaction was achieved under 1.0 atm pressure of CO₂ from a common intermediate through mechanistic divergence. Mechanistic studies suggest that while the acylative-benzylation takes place via radical–radical cross-coupling, the acylative-carboxylation proceeds via radical polar cross-over to generate a benzylic carbanion followed by nucleophilic attack on feebly electrophilic CO₂. We hope that this work will enrich the alkene difunctionalisation paradigm via the rarely reported radical–radical cross-coupling method or a mild carboxylation strategy with gaseous CO₂. Scale-up synthesis under sunlight irradiation, low catalyst loading with mild reaction conditions, and regiospecific product formation with no potential byproducts are some of the practical aspects of this reaction. Considering the stringent regulation of environmental protection agencies (EPA), G20 summit and other forums, the direct utilization of CO₂ and visible light mediated ptotoredox-mediated organic transformations under mild conditions are considered as high priority research areas for sustainable development that should interest the scientific community. Interestingly, both of these alkene difunctionalization reactions furnished an all-C quaternary center which is difficult to achieve by organometallic transformations. Surprisingly, this complexity generating reaction occurs under very simple and mild reaction conditions even under sustainable sunlight irradiation. Further exploration of this discovery on the confluence of radical–radical and radical–polar cross-over mechanisms is underway in our laboratory.

Author contributions

R. J., S. N., and P. D. designed the project and wrote the manuscript. S. N., P. D., S. D., and S. M. performed the experiments. R. J. supervised the project overall.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by DST, SERB, Govt. of India Core Research Grant No. CRG/2021/006717. S. N. thanks DST-INSPIRE and P. D., S. D., and S. M. thank CSIR for their respective fellowships.

Notes and references

1. P. A. Clarke, S. Santos and W. H. C. Martin, Green Chem., 2007, 9, 438–440.
2. (a) T. Koike and M. Akita, Acc. Chem. Res., 2016, 49, 1937–1945; (b) S. K. Dorn and M. K. Brown, ACS Catal., 2022, 12, 2058–2063; (c) X. Chen, F. Xiao and W.-M. He, Org. Chem. Front., 2021, 8, 5206–5228; (d) Z. Wu, M. Hu, J. Li, W. Wu and H. Jiang, Org. Biomol. Chem., 2021, 19, 3036–3054; (e) B. D. Barve, Y.-H. Kuo and W.-T. Li, Chem. Commun., 2021, 57, 12045–12057; (f) Y. Li, D. Wu, H.-G. Cheng and G. Yin, Angew. Chem., Int. Ed., 2020, 59, 7990–8003; (g) D. R. White, E. C. Bornowski and J. P. Wolfe, Isr. J. Chem., 2020, 60, 259–267.
3. (a) S. Xu, H. Chen, Z. Zhou and W. Kong, Angew. Chem., Int. Ed., 2021, 60, 7405–7411; (b) T. Koike and M. Akita, Chem, 2018, 4, 409–437; (c) L. Pitzer, J. L. Schwarz and F. Glorius, Chem. Sci., 2019, 10, 8285–8291; (d) J. Derosa, O. Apolinar, T. Kang, V. T. Tran and K. M. Engle, Chem. Sci., 2020, 11, 4287–4296; (e) X. Qi and T. Diao, ACS Catal., 2020, 10, 8542–8556; (f) S. O. Badir and G. A. Molander, Chem, 2020, 6, 1327–1339; (g) X.-H. Ouyang, Y. Li, R.-J. Song, M. Hu, S. Luo and J.-H. Li, Sci. Adv., 2019, 5, eaav9839; (h) R. K. Dhungana, S. Kc, P. Basnet and R. Giri, Chem. Rec., 2018, 18, 1314–1340; (i) W. Shu, A. García-Domínguez, M. T. Quirós, R. Mondal, D. J. Cárdenas and C. Nevado, J. Am. Chem. Soc., 2019, 141, 13812–13821; (j) X.-H. Ouyang, R.-J. Song, M. Hu, Y. Yang and J.-H. Li, Angew. Chem., Int. Ed., 2016, 55, 3187–3191; (k) L. Liu, W. Lee, C. R. Youshaw, M. Yuan, M. B. Geherty, P. Y. Zavalij and O. Gutierrez, Chem. Sci., 2020, 11, 8301–8305; (l) X. Ma and G. Zhang, ACS Catal., 2021, 11, 5108–5118; (m) L.-J. Zhong, Z.-Q. Xiong, X.-H. Ouyang, Y. Li, R.-J. Song, Q. Sun, X. Lu and J.-H. Li, J. Am. Chem. Soc., 2022, 144, 339–348; (n) W. Sha, L. Deng, S. Ni, H. Mei, J. Han and Y. Pan, ACS Catal., 2018, 8, 7489–7494.
4. (a) Q. Pan, Y. Ping, Y. Wang, Y. Guo and W. Kong, J. Am. Chem. Soc., 2021, 143, 10282–10291; (b) R. K. Dhungana, R. R. Sapkota, L. M. Wickham, D. Niroula and R. Giri, J. Am. Chem. Soc., 2020, 142, 20930–20936; (c) Z.-L. Li, G.-C. Fang, Q.-S. Gu and X.-Y. Liu, Chem. Soc. Rev., 2020, 49, 32–48; (d) H.-Y. Tu, S. Zhu, F.-L. Qing and L. Chu, Synthesis, 2020, 1346–1356; (e) S. Kc, R. K. Dhungana, V. Aryal and R. Giri, Org. Process Res. Dev., 2019, 23, 1686–1694.
5. (a) M. W. Campbell, M. Yuan, V. C. Polites, O. Gutierrez and G. A. Molander, J. Am. Chem. Soc., 2021, 143, 3901–3910; (b) C. Zhu, H. Yue, L. Chu and M. Rueping, Chem. Sci., 2020, 11, 4051–4064; (c) R. S. Mega, V. K. Duong, A. Noble and V. K. Aggarwal, Angew. Chem., Int. Ed., 2020, 59, 4375–4379; (d) M. W. Campbell, J. S. Compton, C. B. Kelly and G. A. Molander, J. Am. Chem. Soc., 2019, 141, 20069–20078; (e) A. García-Domínguez, R. Mondal and C. Nevado, Angew. Chem., Int. Ed., 2019, 58, 12286–12290; (f) S. Zheng, Á. Gutiérrez-Bonet and G. A. Molander, Chem, 2019, 5, 339–352.
6. (a) M. J. Cabrera-Afonso, A. Sookezian, S. O. Badir, M. El Khatib and G. A. Molander, Chem. Sci., 2021, 12, 9189–9195; (b) X.-G. Zhang, X. Li, C. Zhang and C. Feng, Org. Lett., 2021, 23, 9611–9615; (c) J. Shi, L.-Y. Guo, Q.-P. Hu, Y.-T. Liu, Q. Li and F. Pan, Org. Lett., 2021, 23, 8822–8827; (d) M. D. Kosobokov, M. O. Zubkov, V. V. Levin, V. A. Kokorekin and A. D. Dilman, Chem. Commun., 2020, 56, 9453–9456; (e) S. Wang, B.-Y. Cheng, M. Sršen and B. König, J. Am. Chem. Soc., 2020, 142, 7524–7531; (f) T. Alam, A. Rakshit, P. Begum, A. Dahiya and B. K. Patel, Org. Lett., 2020, 22, 3728–3733; (g) I. U. Hoque, S. R. Chowdhury and S. Maity, J. Org. Chem., 2019, 84, 3025–3035; (h) S. H. Oh, Y. R. Malpani, N. Ha, Y.-S. Jung and S. B. Han, Org. Lett., 2014, 16, 1310–1313; (i) T. Koike and M. Akita, Org. Chem. Front., 2016, 3, 1345–1349.
7. (a) B. R. P. Yip, K. B. Pal, J. D. Lin, Y. Xu, M. Das, J. Lee and X.-W. Liu, Chem. Commun., 2021, 57, 10783–10786; (b) W. Zhou, S. Wu and P. Melchiorre, J. Am. Chem. Soc., 2022, 144, 8914–8919.
8. (a) G. R. Mathi, Y. Jeong, Y. Moon and S. Hong, Angew. Chem., Int. Ed., 2020, 59, 2049–2054; (b) T. Ishii, K. Ota, K. Nagao and H. Ohmiya, J. Am. Chem. Soc., 2019, 141, 14073–14077; (c) D. Chen, L. Xu, T. Long, S. Zhu, J. Yang and L. Chu, Chem. Sci., 2018, 9, 9012–9017.
9. (a) G.-Z. Wang, R. Shang, W.-M. Cheng and Y. Fu, Org. Lett., 2015, 17, 4830–4833; (b) A. Banerjee, Z. Lei and M.-Y. Ngai, Synthesis, 2019, 303–333; (c) C. Raviola, S. Protti, D. Ravelli and M. Fagnoni, Green Chem., 2019, 21, 748–764; (d) M. Zhang, J. Xi, R. Ruzi, N. Li, Z. Wu, W. Li and C. Zhu, J. Org. Chem., 2017, 82, 9305–9311.
10. (a) D. Ni and M. K. Brown, ACS Catal., 2021, 11, 1858–1862; (b) D. Wang and L. Ackermann, Chem. Sci., 2022, 13, 7256–7263; (c) X. Zhao, H.-Y. Tu, L. Guo, S. Zhu, F.-L. Qing and L. Chu, Nat. Commun., 2018, 9, 3488; (d) Y. Lan and C. Wang, Commun. Chem., 2020, 3, 45; (e) K. Liu and A. Studer, J. Am. Chem. Soc., 2021, 143, 4903–4909.
11. D. Leifert and A. Studer, Angew. Chem., Int. Ed., 2020, 59, 74–108.
12. (a) J. T. M. Correia, V. A. Fernandes, B. T. Matsuo, J. A. C. Delgado, W. C. de Souza and M. W. Paixão, Chem. Commun., 2020, 56, 503–514; (b) S. L. Rössler, B. J. Jelier, E. Magnier, G. Dagousset, E. M. Carreira and A. Togni, Angew. Chem., Int. Ed., 2020, 59, 9264–9280; (c) F.-S. He, S. Ye and J. Wu, ACS Catal., 2019, 9, 8943–8960.
13. (a) K. M. Baker, A. Tallon, R. P. Loach, O. P. Bercher, M. A. Perry and M. P. Watson, Org. Lett., 2021, 23, 7735–7739; (b) J. Wang, M. E. Hoerrner, M. P. Watson and D. J. Weix, Angew. Chem., Int. Ed., 2020, 59, 13484–13489; (c) K. M. Baker, D. Lucas Baca, S. Plunkett, M. E. Daneker and M. P. Watson, Org. Lett., 2019, 21, 9738–9741; (d) J. Wu, P. S. Grant, X. Li, A. Noble and V. K. Aggarwal, Angew. Chem., Int. Ed., 2019, 58, 5697–5701; (e) J. Liao, W. Guan, B. P. Boscoe, J. W. Tucker, J. W. Tomlin, M. R. Garnsey and M. P. Watson, Org. Lett., 2018, 20, 3030–3033; (f) X. Li, W. Cui, Q. Deng, X. Song, J. Lv and D. Yang, Green Chem., 2022, 24, 1302–1307; (g) Z. Cai, R. Gu, W. Si, Y. Xiang, J. Sun, Y. Jiao and X. Zhang, Green Chem., 2022, 24, 6830–6835.
14. S. Wang, S. Tang and A. Lei, Sci. Bull., 2018, 63, 1006–1009.
15. W. Liu, Y. Li, K. Liu and Z. Li, J. Am. Chem. Soc., 2011, 133, 10756–10759.
16. M. D. Rathnayake and J. D. Weaver, Org. Lett., 2021, 23, 2036–2041.
17. D.-G. Yu and L.-N. He, Green Chem., 2021, 23, 3499–3501.
18. (a) B. Cai, H. W. Cheo, T. Liu and J. Wu, Angew. Chem., Int. Ed., 2021, 60, 18950–18980; (b) Z. Fan, Z. Zhang and C. Xi, ChemSusChem, 2020, 13, 6201–6218; (c) Z. Zhang, J.-H. Ye, T. Ju, L.-L. Liao, H. Huang, Y.-Y. Gui, W.-J. Zhou and D.-G. Yu, ACS Catal., 2020, 10, 10871–10885; (d) G. Zhang, Y. Cheng, M. Beller and F. Chen, Adv. Synth. Catal., 2021, 363, 1583–1596; (e) S. Nandi and R. Jana, Asian J. Org. Chem., 2022, 11, e202200356; (f) X. He, L.-Q. Qiu, W.-J. Wang, K.-H. Chen and L.-N. He, Green Chem., 2020, 22, 7301–7320; (g) C.-K. Ran, X.-W. Chen, Y.-Y. Gui, J. Liu, L. Song, K. Ren and D.-G. Yu, Sci. China: Chem., 2020, 63, 1336–1351; (h) Y. Jin, N. Toriumi and N. Iwasawa, ChemSusChem, 2022, 15, e202102095; (i) K. Shimomaki, T. Nakajima, J. Caner, N. Toriumi and N. Iwasawa, Org. Lett., 2019, 21, 4486–4489; (j) W. Huang, J. Lin, F. Deng and H. Zhong, Asian J. Org. Chem., 2022, 11, e202200220.
19. (a) V. R. Yatham, Y. Shen and R. Martin, Angew. Chem., Int. Ed., 2017, 56, 10915–10919; (b) J. Hou, A. Ee, H. Cao, H.-W. Ong, J.-H. Xu and J. Wu, Angew. Chem., Int. Ed., 2018, 57, 17220–17224; (c) Q. Fu, Z.-Y. Bo, J.-H. Ye, T. Ju, H. Huang, L.-L. Liao and D.-G. Yu, Nat. Commun., 2019, 10, 3592; (d) Y. Yi, Z. Fan and C. Xi, Green Chem., 2022, 24, 7894–7899.
20. (a) B. Zhang, Y. Yi, Z.-Q. Wu, C. Chen and C. Xi, Green Chem., 2020, 22, 5961–5965; (b) M. Zhang, L. Yang, C. Zhou, L. Fu and G. Li, Asian J. Org. Chem., 2022, e202200087; (c) H. Wang, Y. Gao, C. Zhou and G. Li, J. Am. Chem. Soc., 2020, 142, 8122–8129.
21. (a) L.-L. Liao, G.-M. Cao, Y.-X. Jiang, X.-H. Jin, X.-L. Hu, J. J. Chruma, G.-Q. Sun, Y.-Y. Gui and D.-G. Yu, J. Am. Chem. Soc., 2021, 143, 2812–2821; (b) C. Zhou, M. Li, J. Sun, J. Cheng and S. Sun, Org. Lett., 2021, 23, 2895–2899.
22. J. Bai, M. Li, C. Zhou, Y. Sha, J. Cheng, J. Sun and S. Sun, Org. Lett., 2021, 23, 9654–9658.
23. (a) S. K. Bhunia, P. Das, S. Nandi and R. Jana, Org. Lett., 2019, 21, 4632–4637; (b) H. M. Begam, S. Nandi and R. Jana, Chem. Sci., 2022, 13, 5726–5733; (c) S. Nandi, S. Mondal and R. Jana, iScience, 2022, 25, 104341; (d) K. Manna, H. M. Begam, K. Samanta and R. Jana, Org. Lett., 2020, 22, 7443–7449.
24. F. P. Miller, A. F. Vandome and J. McBrewster, Le Chatelier's Principle: Chemistry, Chemical Equilibrium, Henry Louis Le Chatelier, Karl Ferdinand Braun, Concentration, Temperature, Volume, Pressure, Lenz's Law, Homeostasis, Status Quo, Alphascript Publishing, 2009.
25. (a) N. Holmberg-Douglas and D. A. Nicewicz, Chem. Rev., 2022, 122, 1925–2016; (b) C. K. Prier, D. A. Rankic and D. W. C. MacMillan, Chem. Rev., 2013, 113, 5322–5363.
26. (a) A. R. Katritzky, O. Rubio, J. M. Aurrecoechea and R. C. Patel, J. Chem. Soc., Perkin Trans. 1, 1984, 941–945; (b) D. A. Shabalin, Org. Biomol. Chem., 2021, 19, 8184–8204.
27. Q.-Y. Meng, S. Wang and B. König, Angew. Chem., Int. Ed., 2017, 56, 13426–13430.
28. K. Kalyanasundaram, Coord. Chem. Rev., 1982, 46, 159–244.
29. (a) J. W. Tucker and C. R. J. Stephenson, J. Org. Chem., 2012, 77, 1617–1622; (b) M. H. Shaw, J. Twilton and D. W. C. MacMillan, J. Org. Chem., 2016, 81, 6898–6926.
30. N. A. Romero and D. A. Nicewicz, Chem. Rev., 2016, 116, 10075–10166.

---

Footnotes

† Electronic supplementary information (ESI) available: Optimization details, general procedures, spectral data, crystal data, and NMR experiments. CCDC 2160747 and 2161689. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3gc00143a

‡ These authors contributed equally.

---