Yinxiang 
            Jian‡
          
        
        
      ac, 
      
        
          
            Peng 
            Liang‡
          
        
      ab, 
      
        
          
            Xiaoyan 
            Li
          
        
      ac, 
      
        
          
            Huawu 
            Shao
          
        
      
*ac and 
      
        
          
            Xiaofeng 
            Ma
          
        
      
*ac
      
aNatural Products Research Centre, Chengdu Institute of Biology, Chinese Academy of Sciences, Chengdu, 610041, People's Republic of China. E-mail: maxf@cib.ac.cn; shaohw@cib.ac.cn
      
bSchool of Chemical Engineering, Institute of Pharmaceutical Engineering Technology and Application, Key Laboratory of Green Chemistry of Sichuan Institutes of Higher Education, Sichuan University of Science & Engineering, Xueyuan Street 180, Huixing Road, Zigong, Sichuan 643000, People's Republic of China
      
cUniversity of Chinese Academy of Sciences, Beijing 100049, People's Republic of China
    
First published on 22nd November 2022
Herein, an efficient and highly functional group-compatible procedure for controllable transformation of indoles by the combination of phenyliodine bis(trifluoroacetate) (PIFA) with n-Bu4NCl·H2O (TBAC) was exploited. Through controlling the amount of PIFA and TBAC from one to three equivalents, 3-chloro-indoles, 3-chloro-2-oxindoles, and 3,3-dichloro-2-oxindoles were obtained, respectively, in satisfactory to excellent yields. The advantages of the protocol include mild conditions, facile process with short reaction time, high yields, satisfactory functional group tolerance, and the use of PIFA, which is an air- and moisture-stable promoter. The mechanism studies showed that the reaction may proceed through a halonium ion species-mediated halogenation–elimination–halogenation stepwise process.
![]()  | ||
| Scheme 1 Biologically active molecules and synthetic applications of indole derivatives associated in this study. | ||
For the synthesis of 3-halo-indoles, the traditional procedures are based on the direct electrophilic halogenation of indoles using N-chlorosuccinimide (NCS)8a–d/N-bromosuccinimide (NBS)8e–g or Br2
9/I2,9d1,3-dichloro/dibromo/diiodo-5,5-dimethyl-2,4-imidazolidinedione (DCDMH)/DBDMH/DIDMH,10 SO2Cl2,11 and others.12 Moreover, in recent years, with the development of modern organic synthesis, many catalysed halogenation processes including electrochemical13 and photochemical8f reactions were developed, which realized greener and more sustainable halogenation procedures by improving the regioselectivity, averting overhalogenation, and minimizing the use of highly toxic reagents. The combination of hypervalent iodine reagents (such as (diacetoxyiodo)benzene (PIDA)) with halide anions is another much more attractive strategy because a wide range of 5-substituted indoles can be transformed to the corresponding 3-halo-indoles under very mild conditions with high efficiency (Schemes 2A and B).14
![]()  | ||
| Scheme 2 Synthesis of 3-chloro-indoles, 3-chloro-2-oxindoles, and 3,3-dichloro-2-oxindoles promoted by hypervalent iodine reagents. | ||
While they have been broadly employed in the construction of spirocyclic and 3,3-di-substituted oxindole scaffolds due to their bifunctional properties, the synthesis of 3-chloro-2-oxindoles is highly dependent on the FeCl3-prompted cyclization of β-nitrostyrene derivatives.15 The other procedures always involved multi-step transformation with narrow substrate scope and low yields. In 2018, Prathima and co-worker disclosed that in the presence of oxone and NaCl, indole-3-carboxaldehydes can be smoothly converted into 3-chloro-2-oxindoles. In 14 tested samples, products were produced in 79–92% yields, although it was a two-step process whereby it was necessary for the indole-3-carboxaldehydes to be prepared from indoles.16
In 2019, Yu's group developed an elegantly direct procedure for the synthesis of 3-chloro-2-oxindoles from indoles mediated by a hypervalent iodine reagent [1-chloro-1,2-benziodoxol-3-(1H)-one] (Togni-II-Cl) (Scheme 2C).17 The N-alkyl-substituted indoles and electron-donating groups or halide-substituted indoles were converted into 3-chloro-2-oxindoles in 61–95% yields, while simple 1H-indole offered the product in only 25% yield. Also, the reaction was performed in DMF using TFA as the co-solvent, which may result in the formation of unwanted byproducts when acid-sensitive groups are involved.
Lastly, 3,3-dichloro-2-oxindoles are extremely important oxindole derivatives that have wide biological activity, including anti-diabetic and anti-cancer activity. The synthesis of 3,3-dichloro-2-oxindoles from inexpensive and simple feedstocks remains rare. The first systematic study was from Yu's group.17 During their synthesis of 3-chloro-2-oxindoles from indoles, they found that when the amount of hypervalent iodine reagent was increased to three equivalents, they were able to isolate the corresponding 3,3-dichloro-2-oxindoles in 87% to 99% yields (Scheme 2D). This process seems more effective than the one used for the synthesis of 3-chloro-2-oxindoles because the N-unprotected indoles can be converted into the products in excellent yields, and some functional groups including halides, CN, and NO2 are well tolerated under the reaction conditions.
Despite these great successes, transformation of indoles to 3-halo-indoles, 3-chloro-2-oxindoles, and 3,3-dichloro-2-oxindoles in a controllable manner with satisfactory functional group tolerance and high selectivity remains a challenge. For example, halogenation of indoles is currently the most extensively studied among three transformations. However, most halogenation methodologies use C2-8a,e,f,14g or C5-substituted9d,14f indoles (among the most reactive positions after the C3-position1a,9b,9c), or indoles with electron-withdrawing groups protecting the N-atom,8c,12,14e to improve regioselectivity and minimize overhalogenation, while general protocols for the synthesis of 3-chloro-2-oxindoles and 3,3-dichloro-2-oxindoles from indoles, especially from 1-unsubstituted indoles, remain sparse.17 In these contexts, the systematic study of indole halogenation and oxidation reactions is thus particularly interesting and necessary.
In pursuit of our ongoing efforts on oxidative transformation of aromatic compounds,18 we previously described the hypervalent iodine reagent-mediated direct synthesis of 2-oxindoles from indoles.18b As the continuation of this work, herein, we report the results of our investigation of selective synthesis of 3-halo-indoles, 3-chloro-2-oxindoles, and 3,3-dichloro-2-oxindoles by the combination of phenyliodine bis(trifluoroacetate) (PIFA) and n-Bu4NX (TBAX, X = Cl, Br, I). By controlling the amount of PIFA-TBAC, the corresponding monohalogenation, chloridion–oxidation, and dichloridion–oxidation products were produced in satisfactory to excellent yield (Scheme 2F).
Compared to the procedures in Scheme 2A–D, the properties of this process include (i) the first examples for conversion of indoles into 3-halo-indoles, 3-chloro-2-oxindoles, and 3,3-dichloro-2-oxindoles by simply controlling the amount of reagents for halogenation; (ii) the most general protocol for regioselective 3-chlorination of indoles developed to date, with a variety of solvents that can be employed; (iii) likely reaction through halogenation–elimination–halogenation stepwise-delivered diversity of products with oxidation-sensitive groups such as aldehyde, alkenyl, alkynyl, and alcohol tolerated; and (iv) rapid reactions for 3-halo-indoles and 3-chloro-oxindoles that finish in less than 5 minutes.
| Entry | Oxidant | Cl source | Solvent | T (°C) | Yieldb (%) | 
|---|---|---|---|---|---|
| a General conditions: 1a (1.0 equiv.), oxidant (1.2 equiv.), Cl source (1.2 equiv.). b Isolated yield. | |||||
| 1 | PIFA | NaCl | CH2Cl2 | rt | N.R. | 
| 2 | PIFA | NH4Cl | CH2Cl2 | rt | N.R. | 
| 3 | PIFA | TBAC | CH 2 Cl 2 | rt | 93 | 
| 4 | PIFA | TBAC | DCE | rt | 73 | 
| 5 | PIFA | TBAC | Toluene | rt | 53 | 
| 6 | PIFA | TBAC | THF | rt | 56 | 
| 7 | PIFA | TBAC | 1,4-Dioxane | rt | 47 | 
| 8 | PIFA | TBAC | CH3CN | rt | 67 | 
| 9 | PIFA | TBAC | Et2O | rt | 73 | 
| 10 | PIFA | TBAC | Acetone | rt | 64 | 
| 11 | PIFA | TBAC | CHCl3 | rt | 60 | 
| 12 | PIFA | TBAC | EtOAc | rt | 48 | 
| 13 | PIDA | TBAC | CH2Cl2 | rt | Complex | 
| 14 | PIFA | TBAC | CH2Cl2 | 0 | 90 | 
| 15 | PIFA | TBAC | CH2Cl2 | 40 | 89 | 
The reaction can also be performed in a wide range of other solvents including 1,2-dichloroethane (DCE), toluene, tetrahydrofuran (THF), 1,4-dioxane, CH3CN, ether (Et2O), acetone, CHCl3, and ethyl acetate. In all of these solvents, the 3-chloro-indole product (2a) was isolated in 47% to 73% yield (Table 1, entries 4–12). Changing the oxidant to PIDA resulted in the formation of a complex mixture (Table 1, entry 13). When the reaction was carried out at 0 °C and 40 °C, respectively, the product can still be formed at approximately 90% yield (Table 1, entries 14 and 15).
After determining the optimal reaction conditions, we proceeded to investigate the substrate scope. As shown in Table 2, we found that the reaction exhibits remarkable functional group tolerance, and it is compatible with numerous varieties of substituents on the phenyl ring and N-substituent. The substrates with an electron-donating group on the N-atom, such as Me, allyl, and Bn, were smoothly chloridized at the 3-position in satisfactory to excellent yields with high regioselectivity (Table 2, 2b–2d).
The electron-donating (such as Me, MeO) and electron-withdrawing substituents (such as NO2, CN, F, Cl, Br, I) at the 4-, 5-, 6-, or 7-position of the phenyl ring were compatible with the chlorination reaction. In all the tested cases, the corresponding 3-chloro-indoles were provided in satisfactory to excellent yields, and all the reactions were finished in less than 3 minutes (Table 2, 2e–2v). The substrates substituted with a double bond (C
C), aldehyde, or β-nitrostyrene were all compatible under the reaction conditions, achieving the products in 92% to 94% yields (Table 2, 2c, 2w, 2x), which further demonstrated the efficiency and highly functional group tolerance of our approach for the synthesis of 3-chloro-indoles.
The tricyclic indole is also a suitable substrate for the reaction, and gave the corresponding 3-chlorination indole product 2y in 91% yield. We examined the influence of the substituted group on the 2-position, because the 2-substituted 3-chloro-indoles are broadly used as bioactive molecules, as shown in Scheme 1A. Obviously, there were negligible effects on the efficiency of the reaction by the substitutes in the 2-position of indole, and the tested three substrates offered the products in satisfactory to excellent yields (2z, 2aa, 2ab). 7-Azaindole was also converted into a 3-chlorination product (2ac) in 80% yield. To further investigate the scope of the reaction, we studied the influence of the different halogen sources, including n-Bu4NF·6H2O (TBAF), n-Bu4NBr (TBAB), and n-Bu4NI (TBAI). Even though no desired product (2ad) was detected when TBAF was employed, the halogen atom from TBAB and TBAI was smoothly transformed into indole, and the corresponding 3-bromo-1H-indole 2ae and 3-iodo-1H-indole192af were provided in 90% and 83% yield, respectively (Table 2). Because 3-iodo-indoles are extremely important building blocks for the preparation of 3-substituted indole derivatives that are prevalent in numerous biologically active molecules,19e our protocol would therefore provide a novel and useful strategy for the synthesis of 3-iodo-indoles.
We previously showed that the combination of PIFA and TBAC oxidized 3-substituted indoles to corresponding 2-oxindoles. The efficient halogenation and oxidation of various indoles under similar conditions fascinated us, and thus, we further explored the possibility of synthesizing 3-chloro-2-oxindoles directly from indoles. Interestingly, by simply increasing the amount of PIFA and TBAC to 2.2 equiv., we were able to isolate 3-chloro-2-oxindole from indole in 83% yield without altering any other reaction conditions,18b and the reaction was as efficient as the process for the synthesis of 3-chlorination indole, with both reactions finishing in less than 5 min.
Herein, we further investigated the substrate scope for the direct synthesis of 3-chloro-2-oxindoles. As illustrated in Table 3, varieties of 3-chloro-2-oxindoles were successfully obtained in moderate to satisfactory yields directly from indoles. The electron-donating substituents such as Me, Bn, and allyl on the N-atom of indole reacted well under the standard reaction conditions, offering the corresponding products in 70%–79% yields (Table 3, 3a–3c). The electron-donating groups at the 4-, 5-, 6-, and 7-position of indoles are also suitable for the oxidation reaction, and the corresponding 3-chloro-2-oxindole products were achieved in moderate to satisfactory yields (Table 3, 3d–3h). Indoles with electron-withdrawing groups, including NO2, CN, Cl, and Br at different positions of the phenyl ring, were also smoothly chlorinated and oxidized to 3-chloro-2-oxindoles in up to 86% yield (Table 3, 3i–3m). The reaction also tolerated substrates with tricyclic-containing indoles, despite producing the product (3n) with a diminished yield (47%). Finally, 3-chloro-7-azaoxindole 3o was successfully obtained in approximately 60% yield with this chlorination–oxidation protocol.
At this stage, we were curious what would happen if we further increased the amount of PIFA and TBAC. Much to our surprise, when 5-chloro-indole was treated with 3.2 equiv. of PIFA combined with same amount of TBAC, we observed a clear conversion after stirring the reaction mixture at room temperature for 1.5 hours. The major product was proved to be 3,3,5-trichloro-2-oxindole (4a), and the yield for this transformation was 86%. 5-Bromo-indole also delivered the product 4b in up to 95% yield. We then focused our attention on the functional group tolerance of this oxidation–dichlorination protocol. Some representative examples are collected in Table 4. Indoles with electron-withdrawing groups (including NO2, CN, and ester) substituted on the phenyl ring were all smoothly converted into dichloridized oxindole products in up to 96% yield (4c–4e). Indole substrates with electron-donating groups (such as Me) on the phenyl ring also generated the corresponding product (4f) in 60% yield under the standard conditions.
The high yield under the very mild reaction conditions encouraged us to further extend this protocol to tricyclic and N-substituted indoles. The functional groups at N-atoms such as alkynyl and free hydroxyl groups were all compatible under the reaction conditions, offering the products (4h, 4i) in 85% and 82% yield, respectively. N-Allyl-substituted groups were also successfully transformed into the 3,3-dichloro-2-oxindole 4j despite in the reduced yield. Moreover, azaindoles including 4-azaindole and 7-azaindole were smoothly converted into 3,3-dichloro-2-oxindoles (4k, 4l) in excellent yields.
To clarify the mechanism of the halogenation and oxidation processes, some control experiments were carried out (Scheme 3). It was found that the 3-chlorination product was formed in 71% and 90% yield when 1.0 equiv. of 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) and 2,6-di-tert-butyl-4-methylphenol (BHT), respectively, was added to the reaction mixture, which may rule out the possibility of a radical process (Scheme 3, Eqn. 1). The lower yield from TEMPO possibly occurred because it reacted with PIFA, inhibiting the formation of active halonium ion.20 Additionally, treatment of 5-bromo-3-chloroindole with 1.2 equiv. of PIFA and TBAC smoothly formed 3l in 53% yield (Scheme 3, Eqn. 2). When 3l was further reacted with PIFA and TBAC (1.2 equiv.), 4b was isolated as the only product in 69% yield (Scheme 3, Eqn. 3, Conditions A), which suggested that direct chlorination of 3-chloro-2-oxindole to form 3,3-dichloro-2-oxindole may take place in our case. We have previously proved that when PIFA was stirred with TBAC, it formed the highly active intermediate 5 (vide infra), which may trigger the subsequent chlorination and oxidation.18b Moreover, treatment of 3l with 5 or HOCl (in situ formed from the reaction of NaOCl and HClaq) produced 4b in 64% and 69% yield, respectively (Scheme 3, Eqn. 3, Conditions B and C). These control experiments suggest that intermediate 5 and HOCl may be involved in all three transformations.
On the basis of the above and the previous experiments, a plausible mechanism for the transformations is proposed in Scheme 4. The reaction between TBAC and PIFA generates a new and highly active hypervalent iodine intermediate 5, which is followed by reductive elimination to form Cl–OH species.18b Electrophilic chlorination of indole by 5 or Cl–OH produces intermediate 6,1a which then undergoes aromatization to form 3-chloro-indole product 2e.
Subsequently, 2e participates in the second chlorination by 5 or Cl–OH to provide intermediate 7, which is attacked by trifluoroacetate anion or H2O to form intermediate 8a and 8b, respectively. In the presence of H2O, intermediate 8a can also participate in hydrolysis reactions to produce 8b.21 Elimination of HCl from 8b provides 3-chloro-2-oxindole 3e. Lastly, chlorination of 3e by 5 or Cl–OH can offer the 3,3-dichloro-2-oxindole 4e. This is different from the Togni-II-Cl-promoted dichlorination–oxidation process reported by Yu's group, in which they proved that the oxidation of 8b can directly form 4e.17
The highly efficient, mild reaction conditions and controllable selectivity of our protocols were very useful for the synthesis of 3-chloro-indoles, 3-chloro-2-oxindoles, and 3,3-dichloro-2-oxindoles so that a wide range of indole-based biologically active molecules could be constructed (Scheme 5). For example, when indole (9) was treated with PIFA-TBAC under the standard conditions for the dichloridation–oxidation, the product 3,3-dichloro-2-oxindole (10) that would be used as an anticancer reagent17 was formed in 79% yield (Scheme 5A). The application of our methodology to the formal synthesis of ropinirole hydrochloride, an FDA-approved drug for the treatment of Parkinson's disease, is also shown in Scheme 5B. Chloridation–oxidation of alcohol 11 formed 3-chloro-2-oxindole 12 in 51% yield. Catalytic hydrogenation of 12 in the presence of H2 generated 2-oxindole 13 in 90% yield. Conversion of 13 to ropinirole hydrochloride (14) has been previously achieved after two-step simple transformations.22
Mechanism studies revealed that the reactions may undergo halogenation–elimination (formal oxidation)–halogenation stepwise, which, in turn, will lead to the formation of different products (3-chloro-indoles, 3-chloro-2-oxindoles, and 3,3-dichloro-2-oxindoles) by controlling the amount of PIFA-TBAC. The application of the protocol was demonstrated by the synthesis of biologically active molecules, including ropinirole hydrochloride, an FDA-approved drug that is used for the treatment of Parkinson's disease.
Footnotes | 
| † Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2ob01951e | 
| ‡ These authors contributed equally. | 
| This journal is © The Royal Society of Chemistry 2023 |