Iron(III)-Catalyzed Highly Regioselective Halogenation of 8-Amidoquinolines in Water

A simple protocol of iron(III)-catalyzed halogenation of 8-amidoquinolines in water under mild conditions was developed, affording the 5-halogenlated products in good to excellent yields up to 98%. The reaction mechanism most likely involves a single-electron transfer (SET) process.


Results and Discussion
Initially, N-(quinolin- 8-yl)pivalamide, N-bromo-succinimide (NBS) or Br2 were treated as model substrates to optimize the reaction conditions. As shown in Table 1, no desired product was observed under an inert atmosphere (entry 1), indicating the indispensability of the oxidant. Furthermore, although the product could be obtained with 91% yield by NBS without a catalyst, the yield was only 35% with Br2 as a halogen source (entry 2). Thus, in order to make both halogen reagents react well, a variety of metal salts including Pd(II), Cu(II), Co(II), and Fe(III) were then evaluated. To our delight, metal salts were helpful for the catalysis, especially in the case of Br2 as a halogen source, the reactivity of which was improved to the same level as NBS (entries 3-7). Considering the efficiency, low cost, and environmental friendliness, Fe(NO3)3·9H2O was selected as a catalyst for the further studies. The oxidant was another important factor that affected the results. Simple silver salts such as Ag2O and Ag2CO3 gave similar yields as air (entries 8 and 9), while AgOAc raised the yield to about 75% (entry 10). Interestingly, the yield increased dramatically to about 90% in the case of CH3(CH2)5COOAg (entry 11), indicating the positive effect of a long chain carboxylic acid ion, which might function as a phase transfer reagent [42]. Indeed, when the air was used as an oxidant, the combination of CH3(CH2)5COOH and NaHCO3 resulted in the best yield of 95% for both halogen reagents (entry 12). In summary, the optimal conditions consist of quinolines (0.3 mmol), NBS or Br2 (0.6 mmol),

Results and Discussion
Initially, N-(quinolin-8-yl)pivalamide, N-bromo-succinimide (NBS) or Br 2 were treated as model substrates to optimize the reaction conditions. As shown in Table 1, no desired product was observed under an inert atmosphere (entry 1), indicating the indispensability of the oxidant. Furthermore, although the product could be obtained with 91% yield by NBS without a catalyst, the yield was only 35% with Br 2 as a halogen source (entry 2). Thus, in order to make both halogen reagents react well, a variety of metal salts including Pd(II), Cu(II), Co(II), and Fe(III) were then evaluated. To our delight, metal salts were helpful for the catalysis, especially in the case of Br 2 as a halogen source, the reactivity of which was improved to the same level as NBS (entries 3-7). Considering the efficiency, low cost, and environmental friendliness, Fe(NO 3 ) 3 ·9H 2 O was selected as a catalyst for the further studies. The oxidant was another important factor that affected the results. Simple silver salts such as Ag 2 O and Ag 2 CO 3 gave similar yields as air (entries 8 and 9), while AgOAc raised the yield to about 75% (entry 10). Interestingly, the yield increased dramatically to about 90% in the case of CH 3 (CH 2 ) 5 COOAg (entry 11), indicating the positive effect of a long chain carboxylic acid ion, which might function as a phase transfer reagent [42]. Indeed, when the air was used as an oxidant, the combination of CH 3 (CH 2 ) 5 COOH and NaHCO 3 resulted in the best yield of 95% for both halogen reagents (entry 12). In summary, the optimal conditions consist of quinolines (0.3 mmol), NBS or Br 2 (0.6 mmol), Fe(NO 3 ) 3 ·9H 2 O (5 mol%), NaHCO 3 (0.3 mmol), and CH 3 (CH 2 ) 5 COOH (0.3 mmol) at room temperature for 24 h in the air. With the optimized conditions in hand, we subsequently examined the scope of quinoline derivatives, as shown in Scheme 3. Overall, different substrates provided moderate to excellent yields, and both halogen reagents could efficiently realize the reaction, while much lower reactive activity was found in the previous reports [26]. The length of the linear alkyl chain showed few effects on the reaction, affording similar results around 90% yields (Scheme3a-d). Various branched chain alkyl groups also gave excellent yield (Scheme3e-j). Meanwhile, different aryl groups were also compatible in this system. The substrates bearing para-methyl chloro groups gave excellent yields up to 95% (Scheme 3l,m), para-trifluoromethyl groups gave a moderate yield of about 73% (Scheme 3n), and the meta-chloro and methoxyl group gave a good yield (Scheme 3o,p). Replacement of the aryl with the ethylphenyl and thienyl groups were also well-tolerated (Scheme 3q,r). The methyl group on the C2 position of the quinoline ring were also compatible (Scheme 3s). Moreover, the structure of the product (Scheme 3l) was confirmed by X-ray crystallography ( Figure 1)  With the optimized conditions in hand, we subsequently examined the scope of quinoline derivatives, as shown in Scheme 3. Overall, different substrates provided moderate to excellent yields, and both halogen reagents could efficiently realize the reaction, while much lower reactive activity was found in the previous reports [26]. The length of the linear alkyl chain showed few effects on the reaction, affording similar results around 90% yields (Scheme 3a-d). Various branched chain alkyl groups also gave excellent yield (Scheme 3e-j). Meanwhile, different aryl groups were also compatible in this system. The substrates bearing para-methyl chloro groups gave excellent yields up to 95% (Scheme 3l,m), para-trifluoromethyl groups gave a moderate yield of about 73% (Scheme 3n), and the meta-chloro and methoxyl group gave a good yield (Scheme 3o,p). Replacement of the aryl with the ethylphenyl and thienyl groups were also well-tolerated (Scheme 3q,r). The methyl group on the C2 position of the quinoline ring were also compatible (Scheme 3s). Moreover, the structure of the product (Scheme 3l) was confirmed by X-ray crystallography ( Figure 1 Fe(NO3)3·9H2O (5 mol%), NaHCO3 (0.3 mmol), and CH3(CH2)5COOH (0.3 mmol) at room temperature for 24 h in the air. With the optimized conditions in hand, we subsequently examined the scope of quinoline derivatives, as shown in Scheme 3. Overall, different substrates provided moderate to excellent yields, and both halogen reagents could efficiently realize the reaction, while much lower reactive activity was found in the previous reports [26]. The length of the linear alkyl chain showed few effects on the reaction, affording similar results around 90% yields (Scheme3a-d). Various branched chain alkyl groups also gave excellent yield (Scheme3e-j). Meanwhile, different aryl groups were also compatible in this system. The substrates bearing para-methyl chloro groups gave excellent yields up to 95% (Scheme 3l,m), para-trifluoromethyl groups gave a moderate yield of about 73% (Scheme 3n), and the meta-chloro and methoxyl group gave a good yield (Scheme 3o,p). Replacement of the aryl with the ethylphenyl and thienyl groups were also well-tolerated (Scheme 3q,r). The methyl group on the C2 position of the quinoline ring were also compatible (Scheme 3s). Moreover, the structure of the product (Scheme 3l) was confirmed by X-ray crystallography (Figure 1  In an endeavor to expand the scope of this methodology, NIS and I2 were treated as halogen reagents. As shown in Scheme 4, iodination reaction could also be fulfilled (although with low yields around 40%) by using I2 or NIS (3c, 3f). The lower yield of iodination than that of bromination might have been due to the lower reactivity of the iodine free radical and the instability of iodo products [43,44].   In an endeavor to expand the scope of this methodology, NIS and I2 were treated as halogen reagents. As shown in Scheme 4, iodination reaction could also be fulfilled (although with low yields around 40%) by using I2 or NIS (3c, 3f). The lower yield of iodination than that of bromination might have been due to the lower reactivity of the iodine free radical and the instability of iodo products [43,44]. In an endeavor to expand the scope of this methodology, NIS and I 2 were treated as halogen reagents. As shown in Scheme 4, iodination reaction could also be fulfilled (although with low yields around 40%) by using I 2 or NIS (3c, 3f). The lower yield of iodination than that of bromination might have been due to the lower reactivity of the iodine free radical and the instability of iodo products [43,44]. Furthermore, the scaled-up reaction was carried out, giving quantities of the 2c in 90% (Scheme 5), which indicated it as a facile route to the desired product on a more synthetically useful scale. In order to expand the application of this protocol, N-(5-bromoquinolin-8-yl)pivalamide was reacted with boronic acid to give a series of derivatives by simple Suzuki coupling reactions in moderate to good yields ranging from 54% to 84%(Scheme 6) [45,46]. Scheme 6. Suzuki coupling reaction.
Next, radical trapping experiments were carried out, as shown in Scheme 7, and the addition of 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) drastically hampered the reaction. In addition, an EPR Furthermore, the scaled-up reaction was carried out, giving quantities of the 2c in 90% (Scheme 5), which indicated it as a facile route to the desired product on a more synthetically useful scale. Furthermore, the scaled-up reaction was carried out, giving quantities of the 2c in 90% (Scheme 5), which indicated it as a facile route to the desired product on a more synthetically useful scale. In order to expand the application of this protocol, N-(5-bromoquinolin-8-yl)pivalamide was reacted with boronic acid to give a series of derivatives by simple Suzuki coupling reactions in moderate to good yields ranging from 54% to 84%(Scheme 6) [45,46]. Scheme 6. Suzuki coupling reaction.
In order to expand the application of this protocol, N-(5-bromoquinolin-8-yl)pivalamide was reacted with boronic acid to give a series of derivatives by simple Suzuki coupling reactions in moderate to good yields ranging from 54% to 84%(Scheme 6) [45,46]. Furthermore, the scaled-up reaction was carried out, giving quantities of the 2c in 90% (Scheme 5), which indicated it as a facile route to the desired product on a more synthetically useful scale.

Scheme 5. Gram-scale iron-catalyzed C-H bromination of 1c.
In order to expand the application of this protocol, N-(5-bromoquinolin-8-yl)pivalamide was reacted with boronic acid to give a series of derivatives by simple Suzuki coupling reactions in moderate to good yields ranging from 54% to 84%(Scheme 6) [45,46]. Scheme 6. Suzuki coupling reaction.
Next, radical trapping experiments were carried out, as shown in Scheme 7, and the addition of 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO) drastically hampered the reaction. In addition, an EPR experiment was done (for detailed EPR spectra, see SI). Both results suggested that the radical mechanism might be involved in the reaction.
Molecules 2019, 24, 535 6 of 13 experiment was done (for detailed EPR spectra, see SI). Both results suggested that the radical mechanism might be involved in the reaction.

Scheme 7. Radical trapping experiments.
At last, three analogous substrates were also investigated. As shown in Scheme 8, 8aminoquinoline and quinoline gave 40% and none of the product, respectively, which suggested the necessity of the protected amino group during the reaction (Scheme 8a). Moreover, N-(1naphthyl)carboxamide generated a mixture of brominated byproducts, indicating that a chelation of iron with N,N-bidentate 8-aminoquinoline might play a predominant role in the reaction (Scheme 8c). Based on our work, as well as existing literatures [20][21][22][23][24][25][26][27], a plausible reaction pathway was proposed, as shown in Scheme 9. At first, substrate 1 and Fe(III) species formed complex A, which was transformed to be complex B after deprotonation [41,47]. Then, B, which may have influenced the electron density of the quinoline ring at the C5-H position [47,48], was attacked by a bromine radical from the halogen reagent to form complex C by a single electron transfer process (SET). The complex C soon transformed into D through oxidation. After generation of the intermediate E through the proton transfer process (PT), a metal dissociation process gained the terminal product 2 and Fe(III) species, and the catalytic cycle was completed. Furthermore, considering that the reaction could be carried out without a catalyst, although the yield was low, a metal-free halogenation mechanism reported by Xu [32] also may have been involved in the reaction. At last, three analogous substrates were also investigated.
As shown in Scheme 8, 8-aminoquinoline and quinoline gave 40% and none of the product, respectively, which suggested the necessity of the protected amino group during the reaction (Scheme 8a).
Moreover, N-(1-naphthyl)carboxamide generated a mixture of brominated byproducts, indicating that a chelation of iron with N,N-bidentate 8-aminoquinoline might play a predominant role in the reaction (Scheme 8c).
Molecules 2019, 24, 535 6 of 13 experiment was done (for detailed EPR spectra, see SI). Both results suggested that the radical mechanism might be involved in the reaction.

Scheme 7. Radical trapping experiments.
At last, three analogous substrates were also investigated. As shown in Scheme 8, 8aminoquinoline and quinoline gave 40% and none of the product, respectively, which suggested the necessity of the protected amino group during the reaction (Scheme 8a). Moreover, N-(1naphthyl)carboxamide generated a mixture of brominated byproducts, indicating that a chelation of iron with N,N-bidentate 8-aminoquinoline might play a predominant role in the reaction (Scheme 8c). Based on our work, as well as existing literatures [20][21][22][23][24][25][26][27], a plausible reaction pathway was proposed, as shown in Scheme 9. At first, substrate 1 and Fe(III) species formed complex A, which was transformed to be complex B after deprotonation [41,47]. Then, B, which may have influenced the electron density of the quinoline ring at the C5-H position [47,48], was attacked by a bromine radical from the halogen reagent to form complex C by a single electron transfer process (SET). The complex C soon transformed into D through oxidation. After generation of the intermediate E through the proton transfer process (PT), a metal dissociation process gained the terminal product 2 and Fe(III) species, and the catalytic cycle was completed. Furthermore, considering that the reaction could be carried out without a catalyst, although the yield was low, a metal-free halogenation mechanism reported by Xu [32] also may have been involved in the reaction. Based on our work, as well as existing literatures [20][21][22][23][24][25][26][27], a plausible reaction pathway was proposed, as shown in Scheme 9. At first, substrate 1 and Fe(III) species formed complex A, which was transformed to be complex B after deprotonation [41,47]. Then, B, which may have influenced the electron density of the quinoline ring at the C5-H position [47,48], was attacked by a bromine radical from the halogen reagent to form complex C by a single electron transfer process (SET). The complex C soon transformed into D through oxidation. After generation of the intermediate E through the proton transfer process (PT), a metal dissociation process gained the terminal product 2 and Fe(III) species, and the catalytic cycle was completed. Furthermore, considering that the reaction could be carried out without a catalyst, although the yield was low, a metal-free halogenation mechanism reported by Xu [32] also may have been involved in the reaction.

General Experimental Procedures
Unless otherwise noted, all the reactions were performed under air atmosphere. All reagents were used without purification as commercially available. All reactions were monitored by thin layer chromatography. Analytical thin layer chromatography (TLC) was performed using silica gel GF254 plates. Chemical yields refer to pure isolated substances. Column chromatography was performed using silica gel (200-300 mesh or 300-400 mesh) eluting with petroleum ether and ethyl acetate. All products were characterized by their NMR spectra. 1 H-NMR spectra were recorded at 400 MHz and 13 C-NMR spectra at 100 MHz (Bruker DPX, Bruker, Madison, WI, USA) with CDCl3 as a solvent. Chemical shifts were reported in ppm using TMS as the internal standard.

Synthesis of Starting Materials
To a 50 mL single neck flask charged with CH2Cl2 (20 mL) was added 8-aminoquinoline (10 mmol) and triethylamine (11 mmol) and stirred at room temperature for 5 min, then the reaction solution was cooled in an ice bath. The acid chloride (12 mmol) was added dropwise (if solid, it was dissolved with CH2Cl2). The reaction solution was stirred overnight. When it was completed monitored by TLC, the mixture was filtered through a pad of Celite, the solid was washed with ethyl acetate (30 mL), and the organic layer was washed with 1 M NaHCO3 of aqueous solution (3 × 15 mL), then the organic layer was dried with Na2SO4, filtered, and the solvent was removed under reduced pressure. The product was finally obtained by column chromatography on silica gel (PE/EtOAc = 20/1).

General Procedures for Iron-Catalyzed Halogenation C5-H of 8-Amidoquinolines under Mild Conditions in Water
Reaction conditions A: A mixture of 1 (0.3 mmol), NBS (0.6 mmol), Fe(NO3)3·9H2O(5 mol%), CH3(CH2)5COOH (0.3 mmol), NaHCO3 (0.3 mmol) in water (1.0 mL) in a 20 mL Schlenk tube was stirred at room temperature for 24 h. Then, the mixture was extracted with EtOAc (10 mL × 4). The combined organic layer was dried with Na2SO4 and filtered through a pad of Celite. The solvent was removed under reduced pressure and the residue was purified by silica gel column chromatography using PE/EtOAc (20/1) as an eluent to afford the products. Scheme 9. Plausible mechanism for C5 halogenation of quinolines.

General Experimental Procedures
Unless otherwise noted, all the reactions were performed under air atmosphere. All reagents were used without purification as commercially available. All reactions were monitored by thin layer chromatography. Analytical thin layer chromatography (TLC) was performed using silica gel GF 254 plates. Chemical yields refer to pure isolated substances. Column chromatography was performed using silica gel (200-300 mesh or 300-400 mesh) eluting with petroleum ether and ethyl acetate. All products were characterized by their NMR spectra. 1 H-NMR spectra were recorded at 400 MHz and 13 C-NMR spectra at 100 MHz (Bruker DPX, Bruker, Madison, WI, USA) with CDCl 3 as a solvent. Chemical shifts were reported in ppm using TMS as the internal standard.

Synthesis of Starting Materials
To a 50 mL single neck flask charged with CH 2 Cl 2 (20 mL) was added 8-aminoquinoline (10 mmol) and triethylamine (11 mmol) and stirred at room temperature for 5 min, then the reaction solution was cooled in an ice bath. The acid chloride (12 mmol) was added dropwise (if solid, it was dissolved with CH 2 Cl 2 ). The reaction solution was stirred overnight. When it was completed monitored by TLC, the mixture was filtered through a pad of Celite, the solid was washed with ethyl acetate (30 mL), and the organic layer was washed with 1 M NaHCO 3 of aqueous solution (3 × 15 mL), then the organic layer was dried with Na 2 SO 4 , filtered, and the solvent was removed under reduced pressure. The product was finally obtained by column chromatography on silica gel (PE/EtOAc = 20/1).

General Procedures for Iron-Catalyzed Halogenation C5-H of 8-Amidoquinolines under Mild Conditions in Water
Reaction conditions A: A mixture of 1 (0.3 mmol), NBS (0.6 mmol), Fe(NO 3 ) 3 ·9H 2 O(5 mol%), CH 3 (CH 2 ) 5 COOH (0.3 mmol), NaHCO 3 (0.3 mmol) in water (1.0 mL) in a 20 mL Schlenk tube was stirred at room temperature for 24 h. Then, the mixture was extracted with EtOAc (10 mL × 4). The combined organic layer was dried with Na 2 SO 4 and filtered through a pad of Celite. The solvent was removed under reduced pressure and the residue was purified by silica gel column chromatography using PE/EtOAc (20/1) as an eluent to afford the products.
Reaction conditions B: A mixture of 1 (0.3 mmol), Br 2 (0.6 mmol), Fe(NO 3 ) 3 ·9H 2 O (5 mol%), CH 3 (CH 2 ) 5 COOH (0.3 mmol), NaHCO 3 (0.3 mmol) in water (1.0 mL) in a 20 mL Schlenk tube was stirred at room temperature for 24 h. Then, the mixture was extracted with EtOAc (10 mL × 4). The combined organic layer was dried with Na 2 SO 4 and filtered through a pad of Celite. The solvent was removed under reduced pressure and the residue was purified by silica gel column chromatography using PE/EtOAc (20/1) as an eluent to afford the corresponding halogenation products.
Reaction conditions C: A mixture of 1 (0.3 mmol), NIS (0.6 mmol), Fe(NO 3 ) 3 ·9H 2 O(5 mol%), CH 3 (CH 2 ) 5 COOH (0.3 mmol), NaHCO 3 (0.3 mmol) in water (1.0 mL) in a 20 mL Schlenk tube was stirred at room temperature for 24 h. Then, the mixture was extracted with EtOAc (10 mL × 4). The combined organic layer was dried with Na 2 SO 4 and filtered through a pad of Celite. The solvent was removed under reduced pressure and the residue was purified by silica gel column chromatography using PE/EtOAc (20/1) as an eluent to afford the products.