The development of efficient methods for the formation of new chemical bonds under mild and green conditions is one of the fundamental challenges of synthetic chemistry. Transition-metal catalyzed cross-coupling reactions have been proven to be powerful tools for the formation of C-C, C-N, C-O and C-S bonds, and reactions of this type have been widely applied in both academia and industry [1, 2, 3, 4, 5, 6]. However, classic cross-coupling reactions between electrophiles and nucleophiles generally require separate pre-functionalization and de-functionalization steps, which can reduce the overall efficiency of the transformation. A new strategy has recently been developed to overcome this difficulty involving the oxidative coupling of two nucleophiles [7], where the use of hydrocarbons as nucleophiles is considered by many to be an ideal approach. There has been a recent increase in number of oxidative cross-coupling reactions being reported recently involving single electron transfer processes with first-row transition metal catalysts (i.e., Cu and Fe), and radical oxidative coupling reactions of this type represent a promising development in green chemistry [8].
With respect to the development of new sustainable and green synthetic methods, the application of visible light as a renewable source of energy has attracted significant interest from organic chemist working in a variety of different fields [9, 10, 11, 12]. Based on photo-induced single electron transfer processes, visible light photocatalysis provides a new strategy for the oxidation of C-H bonds, and therefore represents an environmentally friendly approach to the construction of valuable chemicals under mild conditions. Although many organic molecules do not possess the structural features required to absorb visible light, the use of visible light photocatalysts (photosensitizers) in conjunction with electron/energy transfer processes could allow for a wide variety of efficient photochemical reactions to be conducted under visible light irradiation conditions.
Although significant developments have been achieved in transition-metal catalyzed oxidative cross-coupling reactions, synthetic strategies based on visible light photoredox reactions have emerged as efficient pathways for the formation of new chemical bonds between two nucleophiles during the last few years. Based on the mild and environmentally benign features of photoredox catalysis and the synthetic step economy of oxidative coupling reactions, the utilization of visible light photo-induced single electron transfer processes in oxidative cross-coupling reactions continues to attract increasing levels of attention (Scheme 1). Furthermore, the increased interest in this area has led to the development of many different types of photocatalyzed oxidative coupling reactions during the last 5 years, including the visible light-photocatalyzed oxidation/coupling reaction of amines, photo-induced oxidative decarboxylative coupling reactions and cross-coupling hydrogen evolution (CCHE) reactions. To date, several important reviews have been published providing a summary of fundamental organic transformations [13, 14, 15, 16, 17, 18, 19]. The aim of this review, however, is to highlight the recent development of visible light-photocatalyzed oxidative coupling reactions.
The visible light-photocatalyzed oxidation of an sp3 C-H bond adjacent to the N atom of a tertiary amine would provide access to an activated iminium ion, which could be trapped with a wide range of different nucleophiles to yield the direct coupling products. Photocatalyzed oxidative cross-coupling pathways of this type have been shown to be powerful protocols for the functionalization of amines. For example, Stephenson et al. [20] reported an oxidative aza-Henry reaction under photoredox conditions in 2010, which proceeded via the formation of an iminium ion intermediate (Scheme 2). Notably, this work provided the first example of the visible light mediated functionalization of an sp3 C-H bond adjacent to a nitrogen atom. The results of this study also revealed that visible light and the catalyst were both required for a successful conversion, and that the absence of oxygen resulted in much lower yields of the product. Furthermore, it was revealed that oxygen and CH3NO2 could both act as oxidants in this transformation. Based on a series of Stern-Volmer studies, the authors went on to propose a mechanism for this transformation. Briefly, visible light excitation of the Ir3+ catalyst followed by reductive quenching would lead to the formation of a radical cation of tetrahydroisoquinoline 2. The authors proposed that oxygen could play a critical role in the reaction. The corresponding superoxide radical anion could abstract a H atom from the trialkylammonium radical cation to give the iminium ion. Nucleophilic attack of 4 onto the iminium ion would provide the observed product 5.
A dual catalytic system was subsequently developed for this reaction by Rueping et al. [21] involving a combination of photoredox catalysis and organocatalysis (Scheme 3). Under these reaction conditions, it was possible to convert an enolizable ketone into a nucleophilic enamine intermediate 8 in the presence of L-proline, and this enamine intermediate could then react with the photocatalytically generated iminium ion 3 to give the Mannich-type addition product. A variety of different N-aryl-tetrahydroisoquinolines have been subjected to this photooxidation/Mannich reaction to provide the corresponding products in good to excellent yields. However, long-chain ketones were found to be unsuitable as substrates for this transformation.
In a series of subsequent reports, Rueping’s group went on to investigate a variety of other photocatalyzed oxidative coupling reactions, which culminated in the development of the oxidative α-cyanation [22] and α-phosphorylation [23] of tertiary amines 1 under visible light photoredox catalytic conditions (Scheme 4). Rueping’s group also reported the development of a visible light-photocatalyzed alkynylation reaction for the preparation of tetrahydroisoquinoline derivatives by combining photoredox catalysis and copper catalysis [24].
Ensuing works by the groups of König [25], Wu [26] and Tan [27] towards the α-C-H oxidation of tertiary amines using photoredox catalysis further extended this methodology to a wide range of nucleophiles, including nitroalkanes, dialkyl phosponates, malononitrile and dialkyl malonates, which were successfully employed as coupling partners with tetrahydroisoquinoline derivatives (Scheme 5). In these studies, eosin Y and Rose Bengal, which are famous organic dyes, were used as photocatalysts. Significantly, molecular oxygen has been shown to be responsible for accelerating these aerobic photocatalytic reactions. Furthermore, electron spin resonance (ESR) measurements collected by Wu’s group provided direct evidence for the formation of a superoxide radical anion (O2•-) rather than singlet oxygen during the visible light irradiation process. Notably, Tan’s group reported that the α-cyanation and α-trifluoromethylation reactions of tertiary amines could be achieved using a combination of graphene oxide and Rose Bengal. In 2014, Zhou et al. [28] reported the development of a visible light-induced oxidative cross-coupling reaction between tertiary amines and diazo compounds. In a similar manner to previous reports, air or oxygen was used as an oxidant in this study to afford various β-amino-α-diazo adducts in good yields.
In 2012, Stephenson et al. [29] reported the development of a new and more versatile approach for the functionalization of α-amino C-H bonds using visible light photoredox catalysis, and improved their initial system using BrCCl3 as an external oxidant to replace oxygen (Scheme 6). Notably, this method also exhibited a high level of compatibility towards a broad range of nucleophiles. Furthermore, the reaction system was removed from blue LED irradiation immediately after the complete conversion of the tertiary amines to the corresponding iminium intermediate 3 to avoid the formation of any undesired by-products.
In 2012, Rovis et al. [30] reported a productive dual-catalysis mode for the asymmetric α-acylation of tertiary amines (Scheme 7). Based on this powerful combination of N-heterocyclic carbene (NHC) catalysis and visible light photoredox catalysis, Rovis’s group successfully achieved the direct asymmetric functionalization of sp3 C-H bonds with a wide range of aldehydes. Using the methodology developed by Stephenson for the visible light-mediated formation of iminium ions with m-dinitrobenzene (m-DNB) as an oxidant, Rovis’s group investigated the reactions of these intermediates with nucleophiles generated in situ using an NHC catalyst to give the desired C-C coupling products. However, further work is still required to expand the substratescope of this reaction to more challenging aromatic aldehydes.
The photochemical properties of Au and Pt(II)-complexes are fascinating. In 2012, Che’s group [31] and Zhu’s group [32] reported their efforts towards the use of organogold(III) complexesto catalyze the direct sp3 C-H functionalization of tertiary amines using visible light irradiation conditions under an oxygen atmosphere (Scheme 8). More recently, Wu et al. [33] reported the use of a Pt(II)-complex as an efficient photocatalyst for the functionalization of sp3 C-H bonds following the addition of 2.0 equivalents of FeSO4. In the presence of FeSO4, the corresponding amide could not be detected under visible light irradiation conditions, but the desired cross-coupling product was exclusively obtained under ambient air conditions.
Iminium ions generated via visible light photocatalysis have been reported to undergo [3+2] cycloaddition reactions with azomethine ylides (Scheme 9) [34]. In 2011, Xiao’s group reported a tandem photocatalytic oxidation/[3+2] cycloaddition/oxidative aromatization sequence for the synthesis of pyrrolo[2,1-a]isoquinolines, which was initiated by visible light. It is noteworthy that this method provided access to a wide range of biologically active natural products. According to this process, the azomethine ylide generated from tetrahydroisoquinoline 25 were reacted with maleimides 26 under air or pure oxygen to give the [3+2] cycloaddition product 27. Rueping’s group also reported a similar [3+2] cycloaddition reaction using visible light photoredox catalysis [35].
As demonstrated by the examples provided above, the substrates for these photocatalytic reactions are mainly limited to tetrahydroisoquinolines, and further work aimed at expanding the substrate scope of these visible light-mediated reactions is highly desirable. The reduction potential of amides is greater than that of the corresponding amines, making it much more difficult for amides to be directly oxidized. In 2012, Stephenson’s group reported the development of a visible light-mediated Friedel-Crafts amidoalkylation reaction using persulfate, which proceeded via the oxidative quenching cycle of Ru(bpy)3Cl2 (Scheme 10) [36]. The key step in this process was found to be oxidation of the dialkylamides. Furthermore, alcohols and electron-rich arenes performed as effective nucleophiles for this reaction, with the corresponding C-O and C-C bonds being formed with high yields and selectivities via a reactive N-acyliminium intermediate. Persulfate (S2O82-) can be used as effective oxidative quencher for the excited-state of Ru(bpy)32+, and the resulting sulfate radical anion can be used to abstract a hydrogen atom from the amide to provide an α-amino radical, which would allow for the generation of the N-acyliminium ion 32 via a photocatalyzed oxidation.
In 2011, Xiao et al. [37] reported the development of a visible light-photocatalyzed intramolecular cyclization reaction. The reactive iminium ion intermediate involved in this reaction was generated from an amine substrate in the presence of a photoredox catalyst and O2 under visible light irradiation conditions (Scheme 11). The iminium ion then underwent an intramolecular cyclization reaction to afford a highly substituted tetrahydroimidazole derivative with high diastereoselectivity in good yield. The stereochemical outcome of this reaction was rationalized on the basis that the Re-face of the iminium ion would be attacked much more readily than the Si-face because of steric repulsion from the pendant aryl groups.
Although the generation of iminium ions by the two electron oxidation of amines and the subsequent reaction of the iminium ions with various nucleophiles has been studied extensively, reports pertaining to the use of α-aminoalkyl radicals formed by single electron oxidation processes as reactive intermediates in photocatalyzed oxidative cross-coupling reactions are scarce. In 2012, Nishibayashi et al. [38] reported the development of a reaction for the direct sp3 C-H amination of benzocyclic amines via an α-aminoalkyl radical using photoredox catalysis (Scheme 12). Di-tert-butyl azodicarboxylate was selected as the N source in this particular case and the visible light-mediated addition of the α-aminoalkyl radical 39 to azodicarboxylate ester 36 gave the corresponding C-N bond formation product 37. According to the proposed mechanism for this reaction, the amine would be oxidized by the excited photocatalyst to give the radical cation 38, which would be deprotonated to give the α-aminoalkyl radical 44. Subsequent reduction of azodicarboxylate ester 36 by Ir(II), would afford the corresponding radical anion 40, which would undergo a radical-radical coupling reaction with 40 to give 41. Subsequent protonation of 41 would then give the product 37. It is also possible to direct the radical addition of 39 to 36 to give intermediate 41.
In early 2013, Bian et al. [39] reported that N,N-dimethylaniline derivatives could react with N-aryl- and N-benzylmaleimides to give the corresponding tetrahydroquinoline products using visible light photoredox catalysis with air as the terminal oxidant (Scheme 13). The formation of the tetrahydroquinoline products observed in this study proceeded via the addition of α-aminoalkyl radicals to the C=C double bond of maleimide 43. The resulting radical intermediate 44 then underwent a free radical cyclization to give the desired product 45. Around the same time, Rueping et al. [40] also reported a similar tandem protocol for the application of α-aminoalkyl radicals in organic synthesis (Scheme 14). Two different products could be obtained from the same starting material in this reaction depending on whether the transformation was conducted in the presence or absence of oxygen. In a similar manner to the work reported by Bian’s group, Rueping suggested that compound 47 was formed via a radical addition/cyclization pathway in the presence of O2. However, in the absence of O2, the α-aminoalkyl radical underwent an intermolecular addition process to the electron-deficient alkene bond to afford 48.
In 2014, Zhou et al. [41] used a photoredox strategy to generate α-aminoalkyl radicals for the synthesis of 3-acylindoles via the intramolecular oxidative cyclization of O-alkynylated N,N-dialkylamines (Scheme 15). According to the proposed mechanism for this reaction, a single-electron transfer (SET) between substrate 49 and Ir(III)* would lead to the formation of Ir(II) and the corresponding radical cation 51, which would undergo a deprotonation step to give the α-amino alkyl radical 52. The subsequent intramolecular radical addition reaction of the carbon radical in 52 to the triple bond of the alkyne would give the vinyl radical 53, which would be trapped by oxygen to generate the superoxide radical 54. The reduction of 54 with Ir(II) would then regenerate the Ir(III) catalyst together with the formation of intermediate 55. Another possible route to 55 would involve the addition of the superoxide radical anion O2•− to vinyl radical 53. Finally, the protonation of 55 would give the vinyl hydrogen peroxide 56, which would undergo an intramolecular hydrogen abstraction step to afford the 3-acylindole 50.
In 2012, Rueping et al. [42] developed a relay catalytic system for the functionalization of glycine derivatives and dipeptides (Scheme 16). Importantly, this methodology extended the scope of photoredox-catalyzed C-H functionalization reactions beyond the use of tertiary amines. The authors found that the addition of Zn(OAc)2 led to significant improvements in the efficiency of the C-H arylation reaction. In a similar manner to the studies described above, this reaction begins with a SET process from the secondary amine substrate to the excited-state of the photocatalyst to give the radical cation 60, which would be converted to intermediate 61. Alternatively, the glycine derivatives could be directly oxidized to 61 by the superoxide anion generated during the course of the reaction. The activated electrophile 62 could be formed in the presence of the Lewis acid catalyst, and could subsequently react with nucleophile 58 to give the desired product 59.
As mentioned above, N-centered radical cations can be generated by the direct oxidation of amines using a photoredox catalyst. Under visible light photoredox conditions, the fate of nitrogen-centered radical cations has been shown to follow one of two reaction pathways, including (1) the conversion of the radical into an iminium ion with concomitant release of a hydrogen radical or (2) the conversion of the radical into an α-amino radical by deprotonation. Further to these two pathways, it also possible that the radical could undergo a direct electrophilic addition to an alkene, as reported by Zheng’s group in 2012 (Scheme 17) [43]. Zheng’s group developed a new approach to N-arylindoles via a tandem oxidative C-N bond formation/aromatization sequence. The Ru(bpz)32+ photocatalyst used in this sequence was found to be capable of oxidizing N-p-methoxyphenylanilines such as 63 to the corresponding radical cations. It is noteworthy that the addition of silica gel to the reaction mixture led to a significant increase in the rate of the reaction. This increase in the rate was attributed to the silica gel adsorbing oxygen and behaving as a source of protons, as well as facilitating the oxidation of the substrate through a proton-coupled electron transfer process. The mild aerobic oxidation conditions of this reaction were compatible with variety broad range of functional groups, although a p-alkoxyphenyl group on the N atom was found to be critical for the reaction.
In 2012, Li et al. [44] reported the development of an aerobic visible light-mediated reaction for the formation of 2-substituted benzothiazoles via the radical cyclization of thioanilides (Scheme 18). Oxygen served as a terminal oxidant in this oxidative cyclization reaction. However, this reaction was found to be particularly sensitive to the concentration of oxygen in the reaction system. This reaction worked well under low oxygen conditions, although the intramolecular cyclization was almost completely suppressed in the absence of oxygen, with 71 being formed as the major product. The proposed mechanism for this reaction starts with the oxidative quenching of the excited-state of Ru(II) species by O2. The subsequent deprotonation of 66 would give the corresponding imidothiolate anion 67, which is much susceptible to oxidation than 66. The single-electron oxidation of 67 by Ru(III) would then give the S-centered radical 68, which would undergo a radical addition reaction to the benzene ring to give dienyl radical 69. Superoxide would then abstract a H atom from radical 69 to give the desired benzothiazole product 70 and hydrogen peroxide.
The oxidative decarboxylation of carboxylic acids is a fundamental and important process in nature. The availability, high stability and low cost of carboxylic acids make them extremely promising raw materials for chemical synthesis. Compared with transition-metal-catalyzed decarboxylation processes, photocatalyzed oxidative decarboxylation reactions represent a novel and efficient approach for the construction of chemical bonds under mild conditions.
In 2013, Zhu et al. [45] reported a visible light-photocatalyzed decarboxylation/radical C-H functionalization reaction that proceeded efficiently at room temperature (Scheme 19). This strategy therefore provided a facile approach for the construction of 3,3-disubstituted oxindoles, which represent a privileged class of heterocyclic scaffolds that can be found in a wide range of natural products and biologically active drugs. Given that carboxylic acids can readily undergo ligand metathesis with phenyliodine diacetate, a wide variety of primary, secondary and tertiary aliphatic carboxylic acids have been reacted in this way to afford the corresponding oxindoles in good yields. Furthermore, a broad range of different functional groups were found to be compatible with these conditions.
In 2013, using O2 as the sole oxidant, Lei et al. [46] reported the radical oxidative decarboxylative coupling of α-keto acids with amines by photocatalysis to give the corresponding amides (Scheme 20). In this case, the α-keto acids were used as precursors for acyl radicals, which were subsequently reacted with a variety of different amines to give the corresponding amide radical anions. The radical anions were then subjected to a further radical oxidation to give the desired amide products. Based on the results of an EPR experiment, the authors proposed a reductive quenching mechanism for this reaction. According to the mechanism, the irradiation of Ru(Phen)3Cl2 with visible light would lead to the formation of the excited state of the photocatalyst, which would be reduced by an amine to form Ru(I). A single-electron-transfer (SET) from Ru(I) to oxygen would provide the superoxide radical anion and regenerate the Ru(II). This method could also be applied to the construction of heterocyclic compounds such as benzimidazoles, benzoxazoles and benzothiazolesusing aniline substrates bearingan NH2, OH or SH group at their ortho-position, respectively. The results of a DFT calculation also supported that the formation of the radical anion was a facile process in this reaction.
Chen et al. [47] recently reported the development of a creative catalytic system based on the combination of hypervalent iodine reagents with a photocatalyst (Scheme 21) [47]. This new system allowed for a series of alkyl-substituted alkenes to be synthesized via a deboronation/decarboxylation sequence under mild conditions. During the course of this study, Chen’s group isolated a novel benziodoxole-vinyl carboxylic acid intermediate form the reaction mixture, which they could be playing a critical role in the catalytic cycle. A mechanism for this reaction was proposed based on a series of experimental investigations. According to this mechanism, the vinyl carboxylic acid substrate would react with BI-OAc to form a benziodoxole vinyl carboxylic acid complex in situ, which would subsequently decompose to generate a carbon-based radical. This radical would then oxidize the photoexcited [Ru(bpy)3]2+* to [Ru(bpy)3]3+, which would be reduced by the alkyl trifluoroborate or boronic acid to complete the catalytic cycle. At the same time, an alkyl R radical would be generated, which would add to the iodine intermediate to form a new radical species. Adduct 86 would then undergo a benziodoxole-facilitated decarboxylation to release a benziodoxole radical together with the alkene product 84.
Along with the development of photocatalyzed oxidative coupling reactions, there have been an increasing number of reports pertaining to new bond formation modes. In 1993, a new chiral photocatalyst, Δ-Ru(menbpy)32+, was synthesized and utilized in the oxidative dimerization of naphthol (Scheme 22) [48]. Using Co(acac)3 as the stoichiometric oxidant, this reaction proceeded smoothly to give 1,1’-bi-2-napthol (88) in good yields. An oxidative quenching mechanism was proposed, and the formation of the α-carbonyl radical was considered to be the key step.
The activation of the sp2 C-H bonds of phenols was recentlyexplored by Yoon’s groups [49], which resulted in the development of a new protocol for the oxidative [3+2] cycloaddition of phenols and alkenes that provided access to a series of dihydrobenzofuran products (Scheme 23). In this report, ammonium persulfate was used as an easy-to-handle, benign terminal oxidant. This oxidative coupling reaction required electron-rich phenols bearing alkoxy substituents at the 2- or 4-position, although a broad range of coupling partners were well tolerated under the reaction conditions. It was reported that an orange precipitate formed in the reaction mixture that could be used to catalyze the [3+2] cycloaddition, and so the authors proposed that the precipitate was Ru(bpz)3(S2O8). According to the proposed mechanism for this reaction, the excited state of this salt would undergo an oxidative quenching step to give the active oxidant Ru(bpz)33+. The oxidation of the phenol substrate would generate the corresponding phenoxonium cation 92, which would be trapped by olefin 90 to give the dihydrobenzofuran product 91.
In 2014, MacMillan et al. [50] developed a photoredox-mediated reaction for the direct α-arylation of cyclic and acyclic ethers with electron-deficient heteroarenes, which proceeded according to a Minisci-type mechanism (Scheme 24). This method showed a broad scope with regard to both the dialkyl ether and heteroarene substrates to give the α-oxyalkylated arene products in high yields. Oxidation of the excited state of Ir(III) by a persulfate anion afforded the Ir(IV) complex together with the sulfate radical anion. It was reported that the α-oxyalkyl radical was generated via a hydrogen-atom-transfer (HAT) pathway between the dialkyl ether and the sulfate radical anion. The radical addition reaction would then occur according to a Minisci-type pathway to provide the amine radical cation, which would lose a proton and an electron to give the α-aryl ether products.
Direct oxidative C-H functionalization reactions involving arenes are important transformations in metal-catalyzed synthetic organic chemistry, and generally require the addition of a stoichiometric amount of a suitable metal salt. In 2014, Rueping et al. [51] reported the development of a combined Rh/photoredox catalytic system for the C-H functionalization of arenes to provide the oxidative Heck products using O2 as the terminal oxidant (Scheme 25). The use of Ru(bpy)3(PF6)2 in the Rh-catalyzed sp2 C-H activation of benzamides with olefins afforded excellent yields of the olefinated products. Subsequent mechanistic studies revealed that air rather than chlorobenzene worked well as the terminal oxidant.
Subsequent reports from the same group revealed that the combination of Pd and photoredox catalysis in the presence of visible light allowed for the C-H olefination of aromatic enamines and the synthesis of the indoles (Scheme 26). [52] Using catalytic amounts of the photocatalyst, it was possible to avoid the typically high loadings of the external oxidant and metal salt. Control experiments showed that no other oxidant was present during this reaction. Furthermore, the superoxide anions formed in situ in the presence of oxygen and the photoredox catalyst functioned as the external oxidant. When this reaction was performed with 100 mol% of the photoredox catalyst under an Ar atmosphere, the corresponding product was isolated in 46% yield. This result therefore demonstrated that the Pd catalyst could be reoxidized by the photoredox catalyst.
In 2014, Rueping’s group reported the development of a photoredox reactionfor the olefination of phenol ethers using a Ru catalyst (Scheme 27) [53]. Rueping’s group also evaluated the effects of the different components on the success of the reaction, including the ruthenium dimer, the silver salt, the photoredox catalyst and molecular oxygen. The highlight of this method was found to be the unique interaction between the metal and the photoredox catalyst, which allowed for the direct re-oxidation of the metal hydride intermediate to complete the catalytic cycle. It is noteworthy that this system can be used for oxidant-sensitive molecules.
Oxidative coupling reactions are some of the most desirable and powerful tools for the direct formation of new chemical bonds from two carbon nucleophiles. For example, oxidative coupling reactions can be used to achieve the direct formation of a new C-C bond from two different C-H bonds using a suitable oxidant. However, these reactions generally require the addition of a sacrificial oxidant, which can lead to the development of numerous waste products and oxidative side reactions. Cross-coupling hydrogen evolution (CCHE) reactions using photoredox catalysis have recently emerged as a powerful approach for the formation of C-C bonds, while circumventing the need for stoichiometric oxidants or the pre-functionalization of substrates.
In 2013, Wu et al. [54] reported the development of an external oxidant-free cross-coupling reaction for the formation of C-C bonds with H2 evolution under visible light irradiation conditions (Scheme 28). In this particular study, the organic dye eosin Y was used as a photosensitizer to convert the tertiary amine substrates to the corresponding iminium ions, which were trapped by nucleophiles to form the new C-C bonds. It is noteworthy that a graphene-supported RuO2 nanocomposite (G-RuO2) was used as a catalyst to reduce the proton eliminated from the C-H bonds of the substrates. A deuterium-labeling experiment was conducted, which revealed that the proton exchange process between the protons released from the substrate and water was quick. Therefore, when D2O was used as the solvent, D2 was generated instead of H2 as the only byproduct.
Wu’s group went on to extend this strategy to the homo-coupling reactions of thiols to give the corresponding disulfides (Scheme 29) [55]. The irradiation of CdSe quantum dots (QDs) with visible light was found to result in the quantitative coupling of thiols to give disulfides products and H2 without the addition of any external sacrificial oxidants. Furthermore, the addition of nickel(II) salts to the system led to a significant improvement in the efficiency and rate of conversion. Subsequent mechanistic studies indicated that the CdSe QDs possessed many active surface sites that could be used to induce the photochemical transformation.
Wu et al. [56] also developed a homogenous catalyst, Co(dmgH)2Cl2 (dmgH=dimethylglyoximate), which they used to catalyze the cross-coupling reaction between isoquinolines and indoles (Scheme 30). In a similar manner to other studies in this area, the organic dye eosin Y was used as a photosensitizer to convert the tertiary amine substrates to the corresponding iminium ions, and the proton released during this process was reduced by a cobalt catalyst to give an almost identical amount of H2. In contrast to other heterogeneous systems reported in the literature, this system could be used in a mixture of organic solvent and water, making it possible to expand the scope of this reaction to water-insoluble substrates.
Photoredox catalysis has been proven to be a valuable tool for the construction of new chemical bonds in synthetic chemistry. Along with the development of photoredox catalysis, a large number of novel visible light photoredox-catalyzed oxidative cross-coupling reactions featuring mild and efficient conditions have been reported in recent years. The mechanisms of these photocatalyzed oxidative coupling reactions often involve a single electron transfer (SET) process between the photocatalyst and the substrate. The reductive quenching or oxidative quenching of the excited photocatalyst generally leads to the formation of a diverse range of reactive intermediates, followed by the further coupling/oxidation of these intermediates to allow for the formation of new chemical bonds. As we have shown, these active intermediates include iminium ions,α-aminoalkyl radicals, S-centered radicals, α-carbonyl radical and α-oxyalkyl radicals, among others. Furthermore, the combination of a photoredox catalyst with a metal catalyst, organocatalyst or proton-reductive-catalyst has recently been used to achieve direct photocatalyzed C-H activation and oxidant-free C-H/X-H functionalization reactions, thereby highlighting the potential of photocatalyzed oxidative coupling reactions.
There are still a large number of challenges in the field of photoredox-catalyzed oxidative coupling reactions. For example, the general scope of photoredox-catalyzed oxidative coupling reaction involving readily oxidized C‒H bonds remains limited. Furthermore, the extension of this methodology to non-amine-based substrates remains a significant challenge, given that amines have been utilized almost exclusively as electron donors in photocatalytic reactions.