Ongoing research aims to diversify and optimize the range of photocatalysts, thereby making photocatalysis more accessible and practical for researchers.15 Heterogeneous photocatalysis offers numerous advantages in chemical synthesis and environmental applications, which makes it an intriguing field of study.16–18 The significant utilities of heterogeneous photocatalysis are that they could successfully be used in flow reactor,19 water treatment,20 energy production,21 etc. In addition, the use of perovskite materials as heterogeneous photocatalysis has emerged as a popular and promising research topic in recent years.22 CeCl3, a well-known Lewis acid,23,24 is insoluble in acetonitrile and lacks light absorption. On the other hand, N-bromosuccinimide (NBS) is known to produce a bromonium ion in acetonitrile.25 However, when anhydrous CeCl3 is combined with NXS (NBS or NCS) in acetonitrile, a transient charge-transfer complex forms, enabling visible light absorption. The complex facilitates photoexcitation of Ce(III) ions, converting them into potent reductants that donate electrons to NXS, generating halide radicals from NXS radical anions. This facilitates the solar synthesis of gem-dihaloketones from terminal alkynes through C-X cross-coupling reactions (Fig. 1b). Remarkably, the catalyst could be recovered by simple filtration and reused without significant decomposition (vide infra).
Using phenylacetylene (1a) and 2.5 equiv of NBS (2a), the standard reaction condition was identified (Fig. 2a) for the synthesis of 2,2-dibromo-1-phenylethan-1-one (3a) under nitrogen atmosphere. The details of the optimization have been given in the supporting information (Table S1). The reactions were performed under blue LED (450–470 nm) or sunlight for 4 h. The yield of 3a was 92% under blue LED and 93% under sunlight. The compound 2,2-dichloro-1-(p-tolyl)ethan-1-one (4a, 86%) was synthesized from NCS (2b) under blue LED. However, the reaction was found to be inefficient in the absence of the CeCl3 photocatalyst. After 48 h, and it gave only 39% of the isolated product (entry 3, Table S1). The use of photocatalysts such as Eosin Y and Rose Bengal26 could not enhance the efficiency of the reactions. A yield of 20% and 23% were obtained after 12 h of reaction time using the photocatalysts Eosin Y27 and Rose Bengal, respectively. We have observed that the reaction failed under N2 atmosphere using dry CH3CN (entry 14). Interestingly, the reaction was successful either under an O2 atmosphere using dry acetonitrile or with wet-acetonitrile.
Anhydrous CeCl3 is a colorless crystalline, does not absorb visible light in the solar spectrum, and is insoluble in acetonitrile (Fig. 2b). Nevertheless, upon the addition of NBS, the solution underwent a notable transformation to dark yellow which was further validated by UV-Vis spectroscopy shown in Fig. 2c (λmax ~ 351 nm). Likewise, the mixture of NCS-CeCl3 exhibited a pale-green color (Fig. 2b) and displayed λmax ~ 370 nm (Fig. 2d). These observations provided important evidence for the formation of a transient charge-transfer (CT) complex.28–31 Figs. 2e and 2f exhibit significant changes in the cyclic voltammogram of NBS and NCS, after the addition of CeCl3, respectively. The excited state reduction potential of Ce(III) (E*red) are known to vary from − 2.19 V to -3.45 V depending on the nature of the ligands attached to Ce(III) ion.12 On the other hand, from the cyclic voltammetric data of NBS, it was noted that the oxidation potential of NBS (Eox) is + 1.4 V (vs Ag/AgCl electrode) which is much higher than the E*red of Ce(III). Moreover, in the presence of CeCl3, the oxidation potential of NBS (Eox) was increased to + 1.54 V. This fact indicates that the oxidizing power of NBS was further enhanced in the presence of Ce(III).32 Similar to the former, oxidation potential of NCS enhanced from + 1.07 V to + 1.15 V after addition of CeCl3 (Fig. 2f).
Furthermore, the ESI-MS spectral analysis of the NBS and CeCl3 mixture (Figure S92 and Figure S93, supporting information) revealed the presence of distinct peaks at m/z 918.982 (100) which attributed to the possible formation of the complexes, [CeCl(H2O)2(NBS)4]+ (Fig. 2g). A significant variation was observed in the 1H and 13C NMR spectroscopy of NBS upon the addition of CeCl3 (Figure S87-S90, supporting information). The changes in the NMR spectra indicated probable interactions occurring between NBS and CeCl3. As an example, the change of the –CH2 peak of NBS at 29.4 ppm is shown in Fig. 2h. The detailed analysis of the data obtained from ESI-MS, cyclic voltammetry, NMR spectroscopy, and UV-Vis spectroscopy provides significant insights into the potential complexation between NBS and CeCl3, which could trigger a cascaded bromination reaction after the generation of a bromide radical in the presence of light.
Terminal alkynes can undergo functionalization to yield gem-dibromoketones through various methods, such as utilizing Br2-HBr in ethyl acetate along with phenacyl bromide,33 conducting one-pot syntheses with DBHT and styrene,34 using an oxidizing agent like KBr-oxone,35 or employing FeCl3 as a catalyst,36 etc. Different gem-dibromoketones synthesized by utilizing various phenylacetylenes are shown in Fig. 3. When the mixture of phenylacetylene (1a) and NBS (2a) was exposed to sunlight, the compound 3a was isolated by 93%. Then, phenylacetylenes containing electron-donating alkyl groups, e.g., methyl, propyl, n-butyl, t-butyl, gave the corresponding dibromo acetophenones (3b-3g) up to 92% yield (Fig. 3a). Electron-withdrawing groups like fluoro, chloro, bromo, and phenyl congaing terminal alkynes when irradiated by blue LED (450–470 nm) or exposed to sunlight converted into the desired product (3h-3l) with an excellent yields (up to 91%). Under standard reaction conditions, phenylacetylene bearing electron-donating alkoxy groups such as -OMe or -OEt afforded the desired compounds 3n-3q with excellent yields (82–93%, Fig. 3b). Similarly, up to 95% yields of the products were obtained for poly-aromatic groups containing terminal alkynes (3r and 3s, Fig. 3c) and tri-substituted phenylacetylene (3m, Fig. 3a). Again, a nitro group containing phenylacetylene and heteroaromatic terminal alkynes (3t-3v) provided satisfactory yields (up to 80%) (Fig. 3c). The gram-scale synthesis of 3a is shown in Fig. 3a and the isolated yield was 90% (2.4 g, 8.27 mmol).
Using the terminal alkynes and N-chlorosuccinimide (2b) as the starting materials, the standard reaction condition was also helpful for the synthesis of gem-dichloroacetophenone, as shown in Fig. 3d. The corresponding, gem-dichloroketones (4a–4f) were produced with high yields (up to 91%) when various electron-donating alkyl groups, such as methyl, n-propyl, n-butyl, and tert-butyl, were present in the phenyl ring of the terminal alkynes. High yields (up to 89%) of the corresponding ketones were produced using the terminal alkyne having electron-withdrawing halide group (4g) and thiophenyl group (4h). Importantly, the volatile nature of the gem-dichloroacetophenones made their isolation difficult.
Figure 4 displays the control experiments conducted to gain insights into the reaction mechanism. Independently, the radical quenching experiments were carried out using TEMPO, BHT, and 1,1-diphenylethylene, and a trace amount of the desired product (3a) was observed (Fig. 4a). This outcome strongly suggests that the formation of 2,2-dibromoacetophenone primarily occurs through radical intermediates. No formation of 3a was observed when the reaction was carried out in a N2 atmosphere using dry CH3CN (Fig. 4b). Subsequently, the reaction was conducted in the presence of O2 in dry CH3CN, resulting in the isolation of 3a with a yield of 91%. Moreover, when the reaction was performed with H218O in dry CH3CN under an N2 atmosphere, the desired product was obtained with a yield of 92%. The formation of 3a-18O was confirmed by GC-MS analysis (Figure S91, supporting information). These observations strongly suggest that the formation of 3a necessitates the presence of O2 and/or H2O in the reaction conditions. The EPR study of the reaction mixture in the presence of DMPO revealed a significant change in the signal, as depicted in Fig. 4c. Conversely, in acetonitrile, CeCl3 did not exhibit any detectable signals. These findings further supported the rationalization of a radical-based mechanism. The light ON-OFF-ON experiment helped in elucidating the indispensability of light for this chemical transformation (Fig. 4d). The distinct advantage of heterogeneous catalysis lies in the ability to recover and reuse the catalyst for multiple cycles. In this study, we conducted five successive cycles for the synthesis of 3a, utilizing the catalyst recovered from the previous cycle by simple filtration. Remarkably, no significant change in photocatalytic activity was observed even after the fifth cycle, because 3a was isolated with a yield of 80% (Fig. 4e). The FESEM analysis (Figure S19, supporting information) demonstrates that the morphology of CeCl3 remains unaltered even after multiple cycles. This indicates that CeCl3 can be reused several times without any significant changes.
Figure 5 depicts a proposed reaction mechanism based on the evidence obtained from control experiments (Fig. 4) and existing literature reports.37 Even though NXS (X = Br, Cl) and CeCl3 do not independently absorb visible light, when they are combined in acetonitrile, they form a charge-transfer complex. This complex is capable of efficiently absorbing radiation within the visible range. As a result, Ce(III) ions can be photoexcited, and they serve as potent reductants by donating an electron to NXS.32 This mechanism emphasizes the significant influence of photoexcited Ce(III) ions, which facilitate the desired C-X cross-coupling reaction. As a result of this electron transfer, halide radicals were generated from the radical anion of N-halosuccinimide (Fig. 5a). These halide radicals play a key role in subsequent reactions, such as anti-Markovnikov cross-coupling with the phenylacetylene to give intermediate I (for NBS). Following, the intermediate I could follow the path a and get oxidized to vinylic carbocation II by Ce(IV) and possibly underwent a nucleophilic attack by H2O to generate III. The intermediate III might have undergone deprotonation and subsequently bromination by 2a, leading to the formation of product 3a. Throughout the reaction, NXS was consistently consumed, whereas CeCl3 remained unaffected and recyclable without significant decomposition (Fig. 4e).
In the presence of molecular oxygen in dry-acetonitrile, an alternative mechanism may have occurred as shown in path b of Fig. 5a. Consequently, intermediate I could undergo a reaction with molecular oxygen, resulting in the formation of intermediate IV. This intermediate IV then coupled with I to yield intermediate V. Following the homolytic cleavage of the peroxy bond in V, intermediate VI was generated, which subsequently reacted with NBS (2a) to produce 3a.
Figures 5b-5d display the synthetic modifications of gem-dibromoketones. Following refluxing conditions for 20 h in acetonitrile, treatment of 3g with silver acetate resulted in the formation of a yellow liquid, specifically 2-(4-(tert-butyl)phenyl)-2-oxoethane-1,1-diyl diacetate (5a), with a yield of 85% (Fig. 5b).38 Fig. 5c shows the nucleophilic displacement of the bromides of 3b by pyrrolidines, led to the formation of 2,2-di(pyrrolidin-1-yl)-1-(p-tolyl)ethan-1-one (5b) with a yield of 79%.39 By following the procedure outlined in Fig. 5d, 2-oxo-2-(p-tolyl)ethane-1,1-diyl bis(4-methylbenzenesulfonate) (5c) was obtained with a yield of 87%.40
Overall, we have achieved a remarkable breakthrough by developing a method to harness visible light using inert CeCl3, effectively transforming it into a recyclable heterogeneous photocatalyst. CeCl3 is a well-known Lewis acid that is insoluble in acetonitrile and non-absorbing to light. Similarly, N-bromosuccinimide (NBS) is also non-absorbing to light and produces a bromonium ion in acetonitrile. However, the combination NXS and CeCl3 in acetonitrile results in the formation of a transient charge-transfer complex capable of efficiently absorbing light, which enables the photoexcitation of Ce(III) ion, and functions as a potent reductant for electron transfer to NXS. Subsequently, the halide radicals generated from NXS played a pivotal role in driving the reactions forward. Throughout the reaction, there was a continuous consumption of NXS, while the catalyst was recovered after the reactions and showed no substantial decomposition, enabling its further reuse.