Dinesh
Singla
,
Vijay
Luxami
and
Kamaldeep
Paul
*
School of Chemistry and Biochemistry, Thapar Institute of Engineering and Technology, Patiala, India. E-mail: kpaul@thapar.edu
First published on 23rd November 2022
C–C and C–S bonds are valuable in organic transformation for synthesizing various bioactive heterocycles. Owing to the increasing awareness of the step and atom economy of a reaction, organic chemists have promulgated the idea of sustainable and green chemistry for organic synthesis. Hence, photoredox catalysis using organic dyes as photocatalysts has attracted constant attention over the last decade. Herein, recent advances and developments (over the previous ten years, 2012–2022) in eosin Y mediated photoredox catalyzed synthetic organic transformations for C–C and C–S bond formation along with their different mechanistic pathways (such as SET, HAT, and energy transfer) are discussed. This review is classified into two major sections, (i) C–C and (ii) C–S bond formations, which are further sub-categorized into various types of functionalizations including arylation, alkenylation, annulation, thionation, and sulfonylation.
Eosin Y, a cheap and widely-used organic dye, has been employed as an economically and environmentally friendly substitute for many transition-metal-based photocatalysts. Depending on the pH of the solution, eosin Y has four different forms: spirocyclic form, monoanionic form, neutral form and di-anionic form.5 As eosin Y has heavy atoms such as Br, it has efficient intersystem crossing (φISC = 0.33) and strong oxidizing and reducing abilities.6 Eosin Y catalyzed reactions through the SET mechanism have some electrochemical requirements for the successful oxidation and reduction of a substrate, as the redox potential of the reacting substrate is comparable to that of eosin Y.7 In contrast, substrates having no comparability of redox potential could be functionalized through an energy transfer pathway or a hydrogen atom transfer mechanism.8 Photoinduced eosin Y catalyzed reactions that proceed through energy transfer pathways are not controlled by the electrochemical or redox potential of the substrate and mainly depend on three factors: (i) the triplet energy of the organic substrate (energy acceptor, EA) must be less than the energy of the photocatalyst (PC) energy donor (ED), (ii) the φISC of the ED must be greater than that of the EA, and (iii) energy must be released from the donor to the acceptor.9
Here the PC gets excited by absorption of visible light to the lowest singlet state S1 from its ground state S0. Next, the PC moves to its triplet state (T1) by efficient intersystem crossing (ISC), where it transfers its energy to the EA (substrate) and excites the substrate to its triplet state. After significant endeavours by several groups, it has been observed that eosin Y has the potential to carry out a broad range of organic reactions such as arylation, alkenylation, and sulphonation for the synthesis of important compounds by constructing C–C and C–X (where X = S, N, O, P) bonds.6
In this review, we mainly emphasize on eosin Y mediated photoinduced C–H functionalization for forming C–C and C–S bonds with mechanistic pathways of some selected reactions. We categorize the review into sections according to synthetic transformations and the goals achieved. Classification is mainly based on two sections: (i) C–C bond formation and (ii) C–S bond formation; these sections are further categorized into subsections based on organic transformations.
Similarly, aryl ketones are privileged motifs in natural products and are widely used as building blocks in synthesizing various pharmaceutical agents. In this context, aryl diazonium salts (1) have given access to aryl ketones (5) by photoredox carbonylation of arenes. Gu and his coworkers in 2015 used photoredox catalysis in the presence of eosin Y to establish a metal-free carbonylation process. The sequential functioning of SET reduction, carbonylation, and reverse electron transfer to produce an acylium cation is supported by mechanistic studies. At room temperature, aryl ketones (5) were synthesized from aryl diazonium salts (1), CO, and other heterocyclic arenes under mild reaction conditions. Several additional (hetero)arenes were likewise subjected to visible-light-induced acylation and gave the corresponding aryl ketones. Moreover, the authors also showed that the efficiency of the reaction was affected by steric effects. Notably, electron-rich aromatic rings have more compatibility and give the desired products in 63%–84% yields (Scheme 3).12 Like aryl ketones, benzoates are also privileged motifs in agrochemicals, materials, and various pharmaceuticals.
Kim et al. in 2018 demonstrated a direct C–C bond formation using a photoredox reaction where various 3-aryl quinoxaline-2(1H)-ones (7) were designed via the direct C–H arylation of quinoxaline-2(1H)-ones (6) with diazonium salts (1). A free radical trapping experiment (Scheme 4) was conducted using TEMPO as a radical scavenger, which indicated the SET mechanism. Due to steric hindrance, Kim's group reported that ortho-substituted diazonium salts were less compatible. Moreover, the electronic effects of quinoxaline-2(1H)-ones (6) were also investigated, where both electron-donating and withdrawing groups were consistent and gave good yields. On the other hand, various N-protecting groups were examined, and no significant effect on the reaction was observed (Scheme 5).13
In the same year, the Yadav group extended Kim's method of direct arylation for the construction of the C–C bond to develop an efficient strategy for synthesizing amino-substituted biphenyl compounds by the straightforward installation of a phenyl ring on anilines; herein, they performed the reaction between anilines (9) and diazonium salts (1) in acetonitrile under green light photoredox catalysis. Substitution on aryldiazonium salts with alkyl, methoxy, fluoro, and cyano groups gave the corresponding diaryl anilines (10) in moderate to good yields. However, when cyano and fluoro groups were installed on N,N-dimethylaniline at the para-position, the corresponding compounds were obtained in low yields. Similarly, when N,N-diphenylaniline was used, again, a low yield of the final compound was obtained (Scheme 6).14 During investigation studies using TEMPO, they obtained a TEMPO-adduct which was further confirmed by HRMS, and indicated a radical pathway for the reaction.
It is noteworthy that Na2–eosin Y (Na2EY) upon irradiation with visible light can get excited to its higher singlet state with t1/2 = 1.9 ns, whereas neutral eosin Y has t1/2 = 2.1 ns due to which Na2EY has more potential to undergo intersystem crossing (ISC) for the generation of its long-lived triplet state (lowest) compared to neutral eosin Y. Several studies have been reported using Na2EY as the photocatalyst having a high efficiency to improve the yield of the final product.15
Hashmi and coworkers in 2020 revealed the C-3 direct arylation of valuable precursor anthranils (11) using various aryldiazonium salts (1) in the presence of the Na2EY photocatalyst at room temperature under the irradiation of blue light. Halogen (F, Br) substituted aryldiazonium salts reacted smoothly and afforded moderate yields, while substituents such as ester, acyl, and electron-withdrawing groups displayed better compatibility to afford good yields (Scheme 7). In addition, they showed the synthetic utility by synthesizing indole and quinoline scaffolds via the formation of α-imino gold carbenes, followed by their addition to alkynes.15 The authors performed radical trapping experiments and found an arene-TEMPO adduct that underwent a stress-free radical reaction pathway by a SET mechanism. The photoinduced triplet state (3[Na2EY]*) having a reduction potential (E1/2 ([Na2EY]/3[Na2EY]*) = −1.11 V vs. SCE) reduced the diazonium salt 1 to give aryl radical A. The attack of the aryl radical on anthranil gave radical intermediate B stabilized by benzisoxazole. SET by Na2EY˙− having an oxidation potential (E1/2 [Na2EY]˙−/[Na2EY] = 0.78 V) regenerated Na2EY along with the formation of a carbocation which gave the final product after deprotonation (Scheme 8).
In 2018, Ranu and co-workers carried out C–H functionalization of cyclic ether for the formation of the C–C bond. They demonstrated remote C–H functionalization of 8-aminoquinoline amide by cyclic ethers under dual catalysis. They catalyzed the reaction under visible light using eosin Y as the main cocatalyst for the generation of free radicals of cyclic ether (17). It formed a cobalt chelate complex followed by interaction with the excited radical eosin Y and abstraction of H+ to give the desired product. They also demonstrated the hybridization of ether with the quinoline amide unit. This approach has synthesized several functionalized ethers (18) in good yields (Scheme 10).17 The reaction proceeded with the generation of a tert-butyl hydrogen radical (Int-A), generated by the dissociation of tert-butyl hydrogen peroxide using photoexcited eosin Y. The free radical abstracted the α-hydrogen of THF to give another radical (Int-B), which interacted with the cobalt–quinoline complex A and gave complex B. Cationic intermediate C was obtained from complex B followed by deprotonation of complex C and the final product was formed (Scheme 11).
In the same year, the Yadav group developed an eosin Y-based dual catalytic system for synthesizing heterobiaryl-pyrazolo[3,4-b] pyridines (21). The authors demonstrated the efficiency of a catalytic system of eosin Y and Yb(OTf)3 (dual catalytic system) by carrying out the reaction of 3-(1H-indol-3-yl)-1-phenyl-1H-pyrazole-5-amine (19) with 1H-indole-3-carbaldehyde (20) for the synthesis of nitrogen-containing heterocyclic compounds. The reaction involved electrocyclization and opening of the indole ring under photoredox catalytic conditions (Scheme 12).18 The authors also demonstrated the TEMPO study and found complete reaction quenching, which means that the reaction involved a radical mechanism.
However, to access various 3-aminoimidazo[1,2-a]pyridines (29), a multicomponent photoinduced cascade transformation was developed in 2020 by Singh et al. They successfully performed the reaction with benzylamine (26), 2-aminopyridine (27), and t-butyl isocyanide (28) using eosin Y as the photocatalyst at room temperature. They demonstrated the formation of a C–C bond. This protocol has outstanding functional group tolerance to benzylamine and gave the desired products with notable yields. Electron-withdrawing groups such as nitro and fluoro gave higher yields than electron-donating groups (methyl and methoxy) (Scheme 14).20 Benzylamine is unable to undergo oxidation, due to the absence of a valid SET mode for this reaction. Here, eosin Y (EY) gets excited upon irradiation of visible light and gives excited eosin Y (EY*), responsible for hydrogen extraction from benzylamine to give an intermediate (Int A). Subsequently, intermediate Int B is formed by the oxidation of Int A. Reaction of 2-aminopyridine with Int B affords an imine intermediate (Int C) followed by the nucleophilic addition of isocyanide and rearrangement, where Int D is formed, and after tautomerism, the desired compound (29) is obtained (Scheme 15).
König and colleagues in 2016 reported the direct arylation of arenes for the synthesis of polyfluorinated biaryl compounds (31). Various arenes, having mono substitution, naphthalene and N-methyl-pyrrole gave the corresponding compounds in yields ranging from 57% to 91%. Moreover, electron-donating and withdrawing substituents were well tolerated to synthesize a number of halogenated biaryl compounds in the presence of eosin Y and showed tremendous significance over transition metal catalysts (Scheme 16).21 This approach has also been used for the late-stage functionalization of the natural product brucine (toxic alkaloid) (32) and demonstrated the functional group compatibility of this methodology (Scheme 17).21
It is noteworthy that the reaction proceeded through the reductive quenching cycle, where triethylamine was used as a sacrificial reductant which transferred electrons to the excited photocatalyst.
Zhan's group reported aerobic oxidative cyclization for synthesizing tetrahydroquinolines (35) using atmospheric oxygen and a metal-free photocatalyst. They carried out a one-pot reaction between N,N-dimethylanilines (9) and maleimides (34) under visible light as an energy source with 2 mol% loading of eosin Y (relatively low loading) at room temperature. Substituted N,N-dimethylanilines (9) with –Br and –CH3 groups reacted smoothly to afford the corresponding products in good yields. Furthermore, N-arylmaleimides (34) with –CH3, –OCH3, –Cl, and –Br gave related compounds with good yields (Scheme 18).22
Here the reaction started with electron transfer from amine to photoexcited eosin Y, which gave an amine radical cation along with [eosin Y]˙− which underwent SET in the presence of oxygen, was converted into neutral eosin Y and then formed a superoxide radical anion. Amine radical cation A underwent deprotonation to give α-aminoalkyl radical B, which further interacted with compound 34 and generated the corresponding radical C which underwent cyclization to produce cation intermediate D. The final product was obtained by proton and electron transfer during interaction between the superoxide radical anion and F along with HOO− which was further protonated to give H2O2 as the by-product (Scheme 19). H2O2 was confirmed using a KI/starch indicator.
Siddiqui and his coworkers have recently reported a multicomponent one-pot synthesis of functionalized indole (38) using eosin Y as an organophotocatalyst. The reaction was performed using substituted anilines (9), 4-hydroxy-6(2H)-pyran-2-one (37) and 2,2-dihydroxy-1-(4-methoxyphenyl) ethan-1-one (36) and gave the corresponding compounds in moderate to good yields via the SET mechanism (Scheme 20).23 This methodology also reveals that eosin Y-mediated photocatalysis has the potential for the multicomponent reaction for synthesizing highly modified compounds.
Yasuda and coworkers devised a feasible method for synthesizing 1,4-dicarbonyl compounds (49) by accelerating a reaction between α-halocarbonyls (48) and silyl enol ethers (47) with the inexpensive photocatalyst eosin Y under visible light irradiation. The chemoselectivity was optimal for the halo-substitution reaction (Scheme 23). Triethanolamine was discovered to be an effective reductant for the regeneration of eosin Y. This methodology was carried out with various silyl enol ethers and α-bromocarbonyl compounds. Furthermore, they envisioned secondary bromoketone, bromoamide, and bromoester, all giving the respective compounds (49) in moderate to good yields.26
Yasuda's group also established exclusive selectivity for C–C bond formation in the reaction between α-bromocarbonyl (48) and allyl trifluoroborate (50) under photoredox catalysis (Scheme 24). In the presence of either Bu4NF or CsF, eosin Y catalyzed the reaction to form the desired product (51). Single-electron transfer (SET) from photoexcited eosin Y to halocarbonyls is aided by metal fluorides which behave as27 Lewis acids. The methodology was likewise successful when using allyl pinacol boronates as allylating reagents. For the allylation of α-halocarbonyls, this process is believed to be both economical and safe. In the same year, König's group also developed a unique approach in which metal-free eosin Y catalyzed decarboxylative alkylation for C–C bond formation. This approach was applied to α-oxy acids, amino acids, and various fatty acids and thus has a vast substrate scope (Scheme 25). Benzylic and non-benzylic α,β-unsaturated esters, and ketones (53) gave the corresponding compounds (54) in moderate to good yields. The reaction worked well for various carboxylic acids (such as amino acids, fatty acids, primary and tertiary acids). It has been demonstrated that several compounds could be synthesized from renewable biomass with the present methodology. Irradiation of visible light on eosin Y (EY) generated the excited photocatalyst (EY*), which underwent reductive quenching in the presence of the sacrificial reductant DIPEA to generate DIPEA˙+. Eosin Y (EY) regenerated by the reduction of N-(acyloxy)phthalimide gave the corresponding radical (B). In addition, decarboxylation gave an alkyl radical (C), which reacted with a Michael acceptor to form the respective intermediate (D) followed by hydrogen atom abstraction from DIPEA˙+, enabling the formation of the desired compound (54) (Scheme 26).28
In one instance, Horner precursors were synthesized via C–C bond formation using triethyl 2-bromophosphonoacetate (56) and indoles (13) under photo-redox catalysis in 2018 by Opatz and co-workers. The authors used triethylphosphonoacetate as a radical precursor. N-Acylated or N-sulfonylated tryptamines reacted efficiently and gave a high yield of the corresponding compounds (56). However, 3-unsubstituted indoles afforded 2-substituted products with high regioselectivity. Due to steric effects, N-functionalized indoles were not converted into the corresponding products. This developed methodology has the potential for the total synthesis of various natural products and provides access to the late-stage modification of peptides (Scheme 27).29
In 2018, Xuan and co-workers disclosed the synthesis of succinic ester (59) from diazo compounds (58) as 1,4-dicarbonyl precursors. During optimization studies, various solvents and bases were screened. Furthermore, control experiments also shed some light on the significant role of the photocatalyst and light source. Radical trapping experiments using TEMPO indicated that the reaction proceeded through a free radical pathway. In the presence of a deuterium source, a deuterated product was obtained which revealed proton-coupled electron transfer for the generation of an α-carbonyl methylene radical from diazoacetate. The reaction showed great compatibility with electron withdrawing and donating groups to afford the desired compounds in up to 82% yield (Scheme 28).30
In 2019, Xu et al. performed the alkylation of benzothiazole (60) using N,N-dimethyl amides (61) via a xanthene-based dye-mediated photoredox system. The reaction was performed under solvent-free conditions using K2S2O8 as an oxidant. It was found that the substitution of benzothiazole at the 4 and 7 positions showed steric hindrance that could limit the potential of the reaction. In addition, they also found that electron-donating groups have more compatibility and give higher yields compared to electron-withdrawing groups (Scheme 29).31 During the introduction of TEMPO in the reaction, only a DMF-TEMPO adduct was obtained, revealing that the reaction proceeded through the free radical mechanism.
Zhang's group in 2020 developed an efficient synthesis of spiro[oxindole-3,4-(4H-pyran)] (65) scaffolds through the visible light-mediated one-pot multicomponent reaction and the formation of multiple C–C bonds. They performed a reaction between isatins (63), 1,3-dicarbonyl (64), and malononitrile (47) in the presence of eosin Y as an organo-photoredox catalyst to afford the final compounds (65) in excellent yields. They also explored substitution on the phenyl ring of isatin and revealed that both electron-donating and withdrawing groups have compatibility with the designed methodology (Scheme 30).32
Leng's group in 2021 disclosed the practical synthesis of α-alkylated ketones (67) by eosin Y mediated alkylation of enol acetates through decarboxylation of N-hydroxyphthalimide acetate. A broad substrate scope and high site selectivity were the prominent features of this strategy. Enol acetate having nBuO, OBn, and OPh as functional groups were also well tolerated. In addition, they demonstrated the synthetic utility of this method by synthesizing dyclonine hydrochloride (Scheme 31).33 The reaction involving the radical decarboxylation addition of the alkyl radical to enol acetates followed by oxidation gave the desired products.
In the same year, direct C3-acylation of quinoxaline-2(1H)-one (68) under visible light using eosin Y as the catalyst was disclosed in 2021 by the Zhao group. Here, they started with readily available aldehydes (69), acting as radical precursors, generated by excited eosin Y, followed by the addition of the radical to the heterocyclic compound quinoxaline-2(1H)-one which underwent oxidation to afford the desired compounds (70) in excellent yields. This approach of acylation has a broad substrate scope. However, electron-donating and -withdrawing substitutions on aldehyde gave the corresponding compounds in lower yields. In addition, protected and unprotected quinoxaline-2(1H)-ones showed compatibility with the reaction (Scheme 32).34
A promising technique to access cycloalkanes via the photoinduced cyclization of enones under the irradiation of visible light using an organic dye as a photocatalyst was developed by Benaglia and coworkers. They synthesized enantiomerically functionalized cyclopentane rings (72) using eosin Y. The reaction was performed in a homemade photoredox coil reactor and a 3D-printed meso reactor and obtained excellent yields of various scaffolds (Scheme 33).35
Hydrazones (76) can also be functionalized using photo-redox catalysis. Wang and his coworkers displayed one such report in 2019.37 They developed a photo-redox methodology for the perfluoroalkylation of hydrazones having wide functional group tolerance and a broad substrate scope (Scheme 35). During optimization, it has been observed that the base was an essential component of the reaction. In addition, the reaction was performed without a light source and organic dye, but no reaction efficacy was accomplished. Notably, hydrazones (76) having para-substituted phenyl rings with different halogens (F, Cl, Br) showed good reactivity. Hydrazones having an electron-withdrawing group (cyano, trifluoromethyl) gave a higher yield than the electron-donating group (methyl, methoxy). In addition, sterically hindered at the ortho-position gave the corresponding compounds (78) in good yields. Moreover, the author showed the methodology's utility at an industrial scale and the discovery of valuable scaffolds via a gram-scale reaction. The reaction proceeded through a single electron transfer mechanism.
Allylic trifluoromethylation using CF3SO2Na (80) (bench stable Langlois reagent) of Baylis–Hillman acetates (79) under photoredox catalysis was developed by Yadav et al. This strategy was accomplished using eosin Y as the organophotocatalyst under the irradiation of green light at room temperature to give trisubstituted alkenes. During preliminary investigation, the authors found no significant effect of the atmosphere on the reaction, hence, reactions were performed in a sealed tube. In addition, in the presence of TEMPO, complete quenching was observed, which confirmed the radical pathway of the reaction. Among electron-withdrawing groups, the NO2 group has no compatibility for this reaction, while others and electron-donating groups have shown remarkable tunning with the developed protocol (Scheme 36).38
Yang and coworkers in 2019 developed a facile synthesis of perfluoroalkylated oxindoles (83) via a photoredox process. They have performed a reaction between N-methyl-N-arylmethacrylamide (82) with n-C4F9I (77). It was observed that both electron-withdrawing and -donating groups reacted smoothly (Scheme 37), and N-arylacrylamides having two methyl groups on a double bond also underwent the reaction smoothly. Still, this time, an anti-isomer of the product was observed. Substrates with N-protecting groups showed excellent compatibility and afforded the corresponding products (83) in good yields, while no efficacy was maintained with unprotected substrates. Based on control experiments and literature precedents, they proposed that the reaction proceeded through the SET mechanism.39
Recently eosin Y-based photoredox catalysis has brought a revolution in organic synthesis when the first MOL (metallic–organic layer) Hf–EY–Fe, containing eosin Y, was designed in 2020 by Lin and his coworkers. This MOL has provided an efficient 2D material for incorporating organic dyes for effective photoredox transformation. Hf–EY–Fe is compatible with large biomolecules and catalyzes various reactions effectively such as aminotrifluoromethylation, hydroxytrifluoromethylation, and chlorotrifluoromethylation. Furthermore, the designed methodology has allowed functionalization on bioactive molecules such as nootkatone, adapalene, estrone, etc. (Scheme 38).40
Similarly, Shin and coworkers developed a facile approach for the chlorotrifluoromethylation of terminal alkenes (73). They performed the reaction between various terminal alkenes (including aromatic heterocycles) and CF3SO2Cl (87), which was based on the SET mechanism; here free radicals formed from CF3SO2Cl by the interaction with photoexcited eosin Y. Several chlorotrifluoromethylated derivatives (88) were obtained in good yields (Scheme 40).42
In 2018, the Wu group demonstrated for the first time C–H functionalization using a direct hydrogen-atom transfer photocatalyst and successfully formed a C–C bond. The reaction was performed using a xanthene-based dye (eosin Y) under mild reaction conditions and this methodology provided high selectivity by forming relatively stable intermediates. After the intensive screening of solvents, and catalyst loading, various functional groups (such as ethers, thioethers, alcohols, and aldehydes) were screened which showed great tunning with the reaction conditions (Scheme 41).43 Control experiments such as potential analysis and quenching studies suggested that C–H activation did not initiate the SET mechanism. Furthermore, a decrease in life time after adding alkene also supported the nonchain pathway. Here, the reaction proceeded through photoinduced eosin Y which helped in the formation of a carbon-centred radical (A) via the HAT process. This radical was trapped by olefin to give radical adduct B. Subsequent reactions between THF and radical B led to the reversible HAT process to give the desired alkylation product (91) (Scheme 42). In the same year, the Xiao group inspired by the initial work of the Wu group, also demonstrated C–H activation of glycals (92) via the hydrogen atom transfer (HAT) mechanism. They have synthesized several 2-deoxyglycosides (94) from glycals having broad functional group tolerance (Scheme 43).44
Very recently, Kong and coworkers demonstrated selective functionalization of unactivated C(sp3)–H bonds. As the transformation of unactivated C(sp3)–H bonds faces hurdles in synthetic organic chemistry, an effective methodology needs to be developed. In this context, the authors developed HAT photocatalysis. The reaction was performed using eosin Y as the photocatalyst in the presence of a Brønsted acid to promote the catalytic efficiency under an argon atmosphere at 60 °C for 48 h. The reaction showed wide substrate tolerance and functional group compatibility and gave the corresponding compounds (91) in moderate to good yields (Scheme 44). To find mechanistic insight they performed several control experiments that suggested the hydrogen atom transfer mechanism.45
Ene-γ-lactam moieties are important synthons as they are an integral structural unit in important medicinal compounds. Liu's group reported that eosin Y, a standard photo-redox catalyst, has shown excellent properties to catalyze thermal redox reactions in the presence of a small amount of Cu(OAc)2. Vinyl azides (103) reacted with ketene silyl acetals (104) in the developed system via [3 + 2] cycloaddition in the dark to give the desired cycloadduct product (105). Electron-withdrawing groups (acetyl, fluoro) on the phenyl ring were more compatible and gave the desired product up to 94% yield (Scheme 48).49 Liu's group also proposed a mechanism for the developed method, as shown in Scheme 49. Initially, the generation of an eosin Y radical took place by the reaction of eosin Y and Cu(OAc)2, which oxidized ketene silyl acetals (KSA) (104) under SET to afford α-ester radicals (Int-A). Furthermore, an iminyl radical (Int-B) was formed via an attack of α-ester radicals on unsaturated vinyl azides. The iminyl radical underwent reduction to give the corresponding anion (Int-D), and then intramolecular rearrangements and isomerization delivered the desired product (105).
During control experiments, complete quenching of the reaction has been observed in the presence of TEMPO, indicating the free radical mechanism of the reaction. This reaction gives an atom-economical and efficient approach towards a bifunctional motif with a broad substrate scope.50 In this approach, the photocatalyst was excited under visible light and oxidatively quenched with 106 to give [Na2–eosin Y]+ and radical intermediate A, which underwent regioselective addition to alkynes and the formation of intermediate B. Vinyl carbocation C was obtained by oxidation of intermediate B which reacted with bromide anions and gave the final product vinyl bromides (108) (Scheme 51).
Molander and his coworkers used eosin Y as a photo-redox catalyst to protect allylic amines and alkyl nitriles by C–C bond formation. They carried out reactions between Boc-protected potassium α-aminomethyltrifluoroborates (109) and several alkenyl sulfones (110, 111). During optimization, a number of photocatalysts and light sources were examined, Na2–eosin Y and green LEDs showed the desired results. Addition of 2,6-lutidine as an additive to increase the yield was unsuccessful. In this reaction, several allylic amines (112, 113) were synthesized via the addition and elimination of an aminoethyl radical in moderate to good yields with high stereoselectivity (Scheme 52).51 But the reaction did not proceed in the case of α-aminomethyltrifluoroborates.
Multiple C–C bonds were formed in a one-pot reaction to synthesize pyrazolo[1,2-a] pyrazoles (116). The dual catalytic system comprising eosin Y and Cu(0) was developed in 2020 by Petek et al. They demonstrated the cycloaddition reaction between azomethine imines (114) and alkynes (115) in a dual catalytic system. The use of TFA under optimized reaction conditions provided acidic medium to eosin Y and the absorption maximum under these conditions was examined, which was at 470 nm. In addition, cyclic voltammetry of azomethine showed that E1/2(111˙+/111) = 0.73 V vs. SCE, hence, it could be oxidized by excited eosin Y having E1/2(3EY*/EY˙−) = 0.83 V vs. SCE. Furthermore, the reaction with TEMPO gave a TEMPO–pyrazolone adduct which was confirmed by HRMS analysis [(m/z); 349.2247 corresponding to C21H32ClN3O2], also suggesting the free radical pathway. This strategy also worked smoothly for both electron-donating and -withdrawing groups and gave the final products in excellent yields (Scheme 53).52
Recently the Yu group synthesized C-alkynyl glycosides via eosin-mediated photoinduced catalyzed reductive coupling between glycosyl bromides (117) and alkynyl bromides (118).
They performed the reaction in the presence of the organic base DIPEA, acting as a reducing agent at ambient temperature under blue light irradiation. Stern–Volmer fluorescence quenching data and radical trapping experiments supported the free radical pathway of the reaction. This methodology has a broad substrate scope and functional group tolerance (Scheme 54).53 Electron-donating groups (–Me, –OMe) and electron-withdrawing groups (CF3, NO2) at the different positions of the phenyl ring showed compatibility and gave products (119) in moderate to good yields. In addition, they demonstrated the synthetic versatility of the final compounds.
Jiang's group has demonstrated a unique approach for simultaneous formation of two C–C bonds along with the construction of a C–S bond. The developed method synthesized sulfone-containing benzo[a]fluoren-5-ones (128) by eosin Y mediated photocatalysis via cyclizing C(sp3)-tethered 1,7-enynes (126) with sulfinic acids (127). This approach has advantages of the potential formation of many functionalized compounds in good yields and high tolerance to a broad range of functional groups (Scheme 57).56 By constructing multiple bonds (C–C and C–S), this approach has triggered the efficient synthesis of polycyclic aryl sulfones.
König's group synthesized vinyl sulfones (132) by photoredox catalysis of aryl sulfinates (130) in 2015. Notably, they synthesized only aryl vinyl sulfones in moderate to excellent yields (Scheme 59).58 However, after one year in 2016, they could achieve synthetic diversity and synthesized many alkyls and heteroaryl vinyl sulfones with high yields (Scheme 60). Various complex cyclic sulfones (134) were synthesized via two photocatalytic sequential steps to generate vinyl sulfones, followed by intramolecular cyclization.59 A broad substrate scope was explored and the corresponding compounds were obtained in good yields. In addition, complete regioselectivity was observed with mechanistic studies and gave two possible mechanisms: (i) reductive quenching cycle and (ii) oxidative quenching cycle as shown in Scheme 61. Here, EY could enter any of the two quenching cycles based on the electronic environment. This was confirmed by calculating the thermodynamics (Gibbs free energy) of both the mechanisms, indicating that both reductive and oxidative quenching cycles (photoinduced electron transfer) were equally exothermic (ΔGPeT = −60 kJ mol−1).
Inspired by Gu et al., Kshirsagar and co-workers performed regiospecific annulation of 2-alkynyl-azidoarenes (40) with sodium sulfinate salts (130) to synthesize 3-sulphonylindoles (135) via cascade cyclization with eosin Y as the photocatalyst. During optimization, combinations of solvents were used, and DMF/H2O gave the best results. Afterward, various azido benzenes having alkyl alkynyl or aryl alkynyl substituents were tested—substituted azidobenzenes with p-Me, p-Et, p-Bu, and p-Br afforded the respective products in good yields. A number of compounds were also synthesized with a variety of sodium sulfinate salts in moderate to good yields (Scheme 62). This reaction proceeded by the reductive quenching cycle, where oxidation of the sulfinate anion by excited eosin Y (EY*) gave an S-centred radical (B). After the regioselective attack of the sulfonyl radical on the alkyne, an alkene radical intermediate (C) was obtained, followed by intramolecular cyclization; an N-centred intermediate (E) was also obtained, which underwent proton abstraction to give the desired compound (135) (Scheme 63).60
The direct synthesis of aryl sulfonyl quinolines (138) via multicomponent one-pot cycloaddition of N-propargyl arylamines (136), SO2, and diaryliodonium salts (137) was achieved by Zhang's group in 2018.61 Based on control experiments of radical trapping it was concluded that the reaction proceeded through the free radical mechanism. The role of DABSO was crucial for SO2 incorporation. Here, the direct formation of the C–S bond afforded the decorated quinoline in one step. Moreover, this approach also exhibited broad functional group tolerance and a broad substrate scope, giving the corresponding sulfur-containing quinolines in moderate to good yields (Scheme 64).
Volla's group in the same year reported the synthesis of thiosulfonates (140) by introducing the SO2 group. The reaction occurred through cross-coupling photoinduced sulfenyl radicals from thiols (139), aryl sulfonyl radicals from aryl diazonium salts (1), and DABCO·(SO2)2. This coupling was accomplished through complete regioselectivity. Furthermore, this protocol also demonstrated that allyl tethered aryldiazonium salts reacted smoothly with thiols to form dihydro benzofuran thiosulfonates (Scheme 65).62 It was also observed that aliphatic thiols have significant compatibility compared to aromatic thiols and gave excellent yields. In addition, the substitution of diazonium salts with electron-donating and -withdrawing groups revealed that all reacted smoothly, but electron-donating groups have a better reaction profile.
Multicomponent sulfonylation of imidazoheterocycles via the formation of two C–S bonds has been developed by the Piguel group. C-3 sulfonylated imidazoheterocycles (143) were synthesized from different imidazopyridines (141) and diaryliodonium salts (142) through a straightforward synthesis, demonstrating an efficient C–H functionalization using eosin Y as the photo-redox catalyst and DABCO·SO2 as the SO2 surrogate. This strategy gave the desired compounds in moderate to good yields with a broad substrate scope. Various electron-donating and -withdrawing groups on the aryl ring of imidazopyridines and aryl diazonium salts were well tolerated to afford the respective products in notable yields (Scheme 66).63
Under the irradiation of green light, photoexcited eosin Y (EY*) was formed and transferred one electron by a SET mechanism to a diphenyl iodonium salt to afford a phenyl radical (A). Then, SO2 trapped the phenyl radical and gave a sulfonyl radical (B), which reacted regioselectively with 2-phenylimidazo[1,2-a]pyridine (141) to give a radical intermediate (C), which transferred the electron to the eosin Y cation and was converted into a carbocation (D), followed by re-aromatization via the loss of a proton and gave the desired sulfone 143 (Scheme 67).
Hydrosulfonylation of allenes was elaborated using organophotoredox catalysis by Voskressensky's group. They performed the reaction using allenes (144) with sodium p-tolylsulfinate in the presence of eosin Y as a photoinduced catalyst. Control experiments suggested the free radical pathway. The authors also demonstrated a deuterium labelled experiment and observed deuterium exchange at the α-position. Furthermore, addition of acetic acid could fasten the reaction and a mixture of vinyl sulfones (regio-isomers) was obtained. Substituted indoles gave a lower yield of the final product compared to substituted sulfinates. The authors also revealed that aliphatic allenes have very low compatibility for the developed method due to the instability of free radicals (Scheme 68).64
Oxime having an active OH unit has attracted significant attention from organic chemists for its number of applications in synthesizing organic compounds. Inspired by the work of Loh's group,55 Li's group explored a 1,5-hydrogen radical shift after adding a sulfur-centered radical to the double bond of the oxime. The sulfonylation of alkenes (146) was developed, assisted by oxime with sulfonyl hydrazides (147) using eosin Y as the photocatalyst at room temperature. Notably, the addition of Ag2O under the reaction conditions was important for the generation of a free radical from sulfonyl hydrazide which was captured by TEMPO and further confirmed by HRMS analysis. This strategy established excellent regioselectivity with wide functional group tolerance (Scheme 69).65
N-Substituted enamisulfones (150) were accessed by reacting vinyl azides (149) with sodium sulfinates (130) under photoredox catalysis. This strategy was developed by Mulina et al. in 2021. The reaction proceeded with a number of steps viz., radical sulfonyl formation, combination with the double bond followed by the generation of an iminyl radical and then reduction, which led to the final product. Electron-donating groups on the sulfinate phenyl ring showed better results than electron-withdrawing groups. In addition, various alkyl sulfinates have also been investigated, and the products were found in good yields (Scheme 70).66
Inspired by the work of Nevado et al., a new strategy that involved an organophotoredox catalyst was developed by Patel and his co-workers in 2021, which resulted in an efficient synthesis of β-substituted vinyl sulfones (153). Regioselective difunctionalization of aryl alkynes (151) was carried out in the presence of sodium aryl sulfonates (130) and carboxylic acids (152). During optimization, it was found that the use of KI in the reaction was crucial as an iodinating agent. In addition, initial investigation of eosin Y in ethanolic solution also suggested the use of green light. To confirm the radical nature of the developed protocol, various radical trapping and Stern–Volmer fluorescence quenching experiments were performed which confirmed the radical pathway. It has been revealed that this strategy worked smoothly with a steric bulk substrate (Scheme 71).67 Formation of the C–S bond and simultaneous introduction of different functional groups such as ester and sulfone made this strategy superior to other previously reported approaches.
Using easily accessible and affordable NH4SCN as the source of SCN, Yadav and co-workers developed an efficient process for the synthesis of α-C(sp3)–H thiocyanate of N-disubstituted amines at ambient temperature. Tertiary amines also showed significant compatibility for this protocol. This approach was also used to produce a C-thiocyanate product from the aza-Baylis–Hillman adduct; the scope of nucleophiles amenable for iminium ion chemistry is expanded. The reaction required 1 mol% of eosin Y, while the greenest and most sustainable reagents are ambient oxygen and visible light. The current technique provides a new methodology for constructing C–S bonds (Scheme 73).69
Later on, the authors also developed a photoredox methodology for synthesizing 6-thiocyanatophenanthridines (157) through a sequential reaction of radical cyclization followed by aromatization of isocyanobiphenyls (57) with ammonium thiocyanate (Scheme 74).70 It is noteworthy that in all developed, convenient and efficient strategies with photoredox catalysts, Yadav's group has used eosin Y as an inexpensive photoredox catalyst, which indicated its potential for thiolation reaction. Recently, they developed a methodology to catalyze the aerobic photoinduced formation of the C–S bond via denitration of β-nitrostyrene (158) using ammonium thiocyanate to give (E)-vinylthiocyanates (159) in up to 87% yield (Scheme 75).71 Substitution at the different positions of the aryl ring with electron-donating groups showed tunning with the reaction and gave the corresponding compounds in good yields. However, substitution with electron-withdrawing groups only at the ortho-position gave compounds in moderate to good yields, while substitution at the meta- and -para positions with these groups gave lower yields of the final compounds.
A click reaction between thiol (139) and alkyne (151) for C–S bond formation via a photo-redox reaction using eosin Y as a catalyst was reported by the Ananikov group in 2016. They demonstrated regioselectivity and broad tolerance of aryl thiols, giving valuable sulfur-containing scaffolds in moderate to good yields. Electron-donating groups (–Me, –OMe) on thiols gave higher yields of the final products (160). However, ortho-substituted thiols with halogens (–Cl, –Br) showed more compatibility and gave better yields. Notably, hexane has been used as a solvent to overcome the degradation of eosin Y. They also designed a photoreactor that could potentially improve the reactivity of the catalytic system (Scheme 76).72
The reaction proceeded through the excitation of eosin Y generating excited eosin Y (EY*), which underwent oxidation and helped in the generation of a radical cation (A) of thiophenol. Subsequently, abstraction of a proton took place from this species by pyridine to give a thiyl radical (B). This radical is reacted with alkyne (151) to provide the desired product (160) (Scheme 77).
However, the Krishnagar group in 2018 gave access to 3-thiocyanatobenzothiophenes (162) by forming multiple C–S bonds. An efficient cascade methodology has been designed, starting with methyl(2-(phenylethynyl)phenyl) sulfane (161) which underwent cascade annulation on reaction with ammonium thiocyanate under visible light-mediated photoredox catalysis of eosin Y. They screened other thiocyanates as well, but NH4SCN showed superior results. Control experiments with TEMPO indicated a radical free mechanism for this reaction. This strategy was favourable for a number of alkynylarylsulfanes and gave good yields of the products. It was also found that the yields dropped significantly under a nitrogen atmosphere (Scheme 78).73
In the same year, 3-sulfenylindolines (163) were also synthesized by thioamination of alkynes. Here, they performed photoinduced radical cascade annulation by reacting 2-alkynyl-azidoarenes (40) with aryl thiols (Scheme 79).74 This reaction also proceeded through the free radical mechanism. However, diaryl sulfides (162) were synthesized in 2017 by the Lee group, in which the reaction was carried out between thiol (139) and aryl diazonium salts (1) using eosin Y as a photocatalyst. A C–S bond formation took place which gave access to the desired sulfides; electron-donating and electron-withdrawing substituted thiols were equally compatible with this approach and had excellent yields up to 92%. They further transformed diaryl sulfide to sulfone, which might be used as an amyloid aggregate inhibitor (Scheme 80).75 A control experiment has also been performed for the mechanistic studies and it was observed that the reaction proceeds through a reductive quenching cycle.
Natarajan and co-workers reported a simple and efficient synthesis strategy of benzothiazole by constructing a C–S bond using eosin Y as a photocatalyst. In this strategy, thiophenol (139) reacted with nitriles (165) to afford the corresponding benzothiazole (166). A number of derivatives were synthesized in good yields. The reaction proceeded through a combination of a radical of thiophenol and nitriles to give an iminyl radical, which underwent intramolecular cyclization and loss of a hydrogen radical to give the final compounds (Scheme 81).76
However, in 2020, they developed the synthesis of 2-mercaptobenzothiazoles (169) by reacting 2-azidoarenediazonium salts (167) and CS2 (168). Generation of an azido arene radical via the SET mechanism and reduction of the azido arene diazonium ion followed by CS2 insertion and intramolecular cyclization gave the desired products in high yields (Scheme 82).77
In another work they demonstrated that the reaction proceeded through the addition of a self-centred radical, generated from thiophenol by the SET mechanism followed by the attack of 1,3-diarylpropynones (170); the reaction was compatible with several thiophenols (139) having p-CH3, p-OCH3, p-NO2, and p-F groups as well as 1,3-diarylpropynones (Scheme 83).78 Notably, the reaction was carried out in open air without using any base and external oxidant.
In the same year, the Singh group reported the sustainable and efficient synthesis of thioethers (172) using photoredox catalysis. The authors proposed that the reaction proceeded through a single electron transfer mechanism. They constructed a direct C–S bond by the reaction of substituted thiols (139) and a number of functionalized hydrazones (76) to access various thioethers (172) in good yields. In addition, a low amount of base and thiol might decrease the yield of the final compound (Scheme 84).79 The cyclic voltammogram of thiophenol showed an oxidation potential of −0.37 V and a reduction potential of −1.3 V, clearly indicating that eosin Y could allow the oxidation of thiophenol via the SET mechanism. Based on control experiments, it was proposed that photoinduced eosin Y oxidized thiol and gave the corresponding thiyl radical cation A and underwent arial oxidation to give neutral eosin Y along with a superoxide radical. Proton abstraction from the thiyl radical cation using the superoxide radical gave thiyl radical B along with the generation of a hydroperoxyl radical (H2O˙). Reaction of hydrazone with a hydroxide ion afforded diazo compound Int-A, which further gave carbene Int-B by the removal of N2; subsequent coupling between carbene Int-B and thiyl radical B gave intermediate C. The final product was obtained by the hydrogen abstraction from C (Scheme 85).
To access highly decorated cis-dihydrochromenone compounds (174), the Volla group has developed an efficient protocol for tandem sulfenylation annulation of the alkyne chain containing cyclohexadienones (173). This protocol demonstrated high diastereoselectivities, a broad substrate scope, and good yields (Scheme 86).80 In addition, in the presence of excess thiophenol and a prolonged time up to 16 h, bisulfenylated scaffolds were synthesized as a single diastereomer. Moreover, various tricyclic derivatives were obtained through self-centred radical addition and the intramolecular Michael addition of amines.
In 2020, the protection of ketones with mercapto alcohols under organophotoredox catalysis was demonstrated by Lee and co-workers. An effective strategy has been developed for the oxathiacetalization of aldehydes and ketones (175) and demonstrated the formation of the C–S bond. The employed conditions were also compatible with various aldehydes such as aromatic, aliphatic, and α,β-unsaturated aldehydes. Based on control experiments it was proposed that the reaction followed a free radical pathway. The eminent feature of the developed strategy was high functional group compatibility (Scheme 87).81
Jia's group has recently developed a photoinduced Atom Transfer Radical Addition (ATRA) reaction for accessing (E)-β-aryl sulfonyl vinyl sulfides (179) with high regio- and stereo-selectivities. They started with aryl alkynes (148) and thiosulfonates (178) under eosin Y catalysis at room temperature. The authors demonstrated the application of the developed strategy by late-stage modification of naturally occurring bioactive motifs (Scheme 88). Based on investigation studies it was revealed that thiyl radicals formed SET and followed the free radical pathway.82
Roy et al. in 2021 developed a prominent synthesis of thioester (180) by forming a C–S bond through photoinduced cross dehydrogenative coupling (PCDC) between aldehydes and thiols at ambient temperature. This strategy proceeded smoothly with both aryl/alkyl aldehydes (69) and alkyl/aryl-thiols (139) to give the final motifs in good to high yields (Scheme 89). Furthermore, during mechanistic studies, it has been revealed that the generation of an acyl radical from an aldehyde was intercepted by a disulfide intermediate to afford the desired thioester.83
Recently, Khade et al. have reported the practical synthesis of 1,3-dithiolanes (182). These scaffolds were obtained without using any base and external additives from terminal aromatic and heteroaromatic alkynes (151) via the photoinduced catalyst eosin Y. Various electron-donating, withdrawing, and deactivating groups showed compatibility with the employed protocol and gave the final products in good yields (Scheme 90).84 The authors also performed various control experiments such as cyclic voltammetry and Stern–Volmer constant calculation to gain mechanistic insight and concluded the free radical pathway via SET oxidation.
Zhang et al. performed direct sulfenylation to construct highly sulfenylated amides (185) via a reaction between thiosulfonates (183) and alkyl/aryl/heteroaryl amides (184) in the presence of TBHP and potassium carbonate, with eosin Y being used as the photocatalyst under mild reaction conditions. The reaction in the presence of TEMPO (radical trapping experiment) confirmed the radical free pathway of the reaction. Different sulfenyl amides were obtained from various amides and sulfones via direct construction of the C–S bond in moderate to good yields (Scheme 91).85
Recently, the Patel group synthesized thio-functionalized pyrroles (187) using visible light and an organophotocatalyst. They gave an inimitable example of visible-light-induced access to thio-functionalized pyrroles. They established the methodology by reacting β-ketodinitriles (186) with thiophenols (139) using a xanthene-based dye (eosin Y) as the photocatalyst. Here the authors also revealed the industrial utility of this large-scale reaction and its synthetic utility by further modification of the final products (Scheme 92).86 Dhar and coworkers demonstrated the thiolation of 1,4-naphthoquinones (188) under photoredox catalysis. They showed the direct formation of the C–S bond and synthesized several thiol substituted naphthoquinones (189) in excellent yields. Various aromatic and aliphatic thiols have been investigated and the radical generation pathway has also been demonstrated by HRMS analysis of the TEMPO-adduct (Scheme 93).87
Recently, Singh and coworkers demonstrated the trifluoromethylation of Bunte salts through the direct formation of the C–S bond under visible light irradiation using eosin Y as the photocatalyst. During control experiments it was observed that only green LEDs gave the desired product in a good yield. Various other dyes were also employed but no efficacy was maintained. In addition, several electron-donating and -withdrawing groups were investigated and it was found that electron-rich (–CH3, –OCH3) groups on the aryl ring afforded high yields, while electron-deficient groups (p-CN, p-NO2) showed less compatibility (Scheme 94).88 Furthermore, the reaction was performed in the presence of a radical trapping scavenger (TEMPO) and complete quenching of the reaction was observed, suggesting the free radical pathway of the reaction.
Despite unprecedented developments, many challenges still exist in this field. Firstly, in some reports, the authors claimed sustainability of the reaction, while the use of solvents such as DMSO, DCE, and DMF limited their sustainability. Secondly, the use of a base in some reactions completely quenches the reaction; as a consequence, the yield of the final product is lowered. Notably, in the present era, regioselective C–H functionalization has attracted incessant attention from the organic community, hence the developed methodologies must have fulfilled the condition of regioselectivity. In addition, it was found that ortho-substituted aryl diazonium salts were less active for the free radical mechanism, which could limit the substrate scope. Various methodologies to accomplish remote C–H functionalization have merged metal catalysis with photocatalysis and hence limited the concept of green chemistry.
Furthermore, reactions of various functional groups such as alcohols, urea, etc. are rare and not much explored for novel coupling partners or biologically active heterocyclic scaffolds. Single electron transfer and hydrogen atom transfer mechanisms are the basic mechanisms followed by most reactions catalyzed by eosin Y. However, very rare organic transformations using eosin Y as a photocatalyst follow the energy transfer mechanism. Hence, there is a significant scope to form C–C and C–S bonds using eosin Y via an energy transfer mechanism. Secondly, more effective methodologies need to be developed to avoid the degradation of eosin Y, as photodegradation is a common process. In this context, efficient immobilization with a covalent organic framework may solve this issue. Thirdly, there is a need for specific light for single electron reduction; in this vein, visible light-mediated catalysis is more favourable. Last but foremost, more profound knowledge of photoinduced organic transformation is demanded. New theories, methodologies, and mechanisms should be developed to deal with the current issues. There is also a need to develop reactions which can make use of water as the solvent for the concept of green chemistry and atom economy. Undoubtedly, a promising development of photoinduced organic transformation for forming C–C and C–S bonds will take place in the future to advance sustainable and green chemistry.
This journal is © the Partner Organisations 2023 |