Hf(OTf)4 as a Highly Potent Catalyst for the Synthesis of Mannich Bases under Solvent-Free Conditions

Hf(OTf)4 was identified as a highly potent catalyst (0.1–0.5 mol%) for three-component Mannich reaction under solvent-free conditions. Hf(OTf)4-catalyzed Mannich reaction exhibited excellent regioselectivity and diastereoselectivity when alkyl ketones were employed as substrates. 1H NMR tracing of the H/D exchange reaction of ketones in MeOH-d4 indicated that Hf(OTf)4 could significantly promote the keto-enol tautomerization, thereby contributing to the acceleration of reaction rate.


Introduction
Mannich reaction has been recognized as one of the most classic multicomponent reactions (MCRs) and utilized for the synthesis of β-amino carbonyl compounds (Mannich bases) via one-pot condensation-addition of aldehyde, amine, and ketone since its discovery in 1917 [1]. Mannich bases are versatile synthetic intermediates [2][3][4][5] and widely applied in the synthesis of natural products [6] and pharmaceutical chemistry [7,8].
Previous research on Group IVB transition metal (Zr(IV) and Hf(IV)) catalysts revealed their superior activity on many carbonyl-transformation reactions [30,43]. Our ongoing research in this field showed that Hf(IV) salts are even more potent than Zr(IV) salts in many carbonyl-transformation Previous research on Group IVB transition metal (Zr(IV) and Hf(IV)) catalysts revealed their superior activity on many carbonyl-transformation reactions [30,43]. Our ongoing research in this field showed that Hf(IV) salts are even more potent than Zr(IV) salts in many carbonyltransformation reactions [44][45][46]. However, the catalytic effect of Hf(IV) salts on Mannich reaction has never been explored. We report herein the identification of Hf(OTf)4 as a highly potent catalyst for efficient synthesis of a diversity of β-amino carbonyl compounds under solvent-free conditions at room temperature. The alkyl ketone-based Mannich reaction catalyzed by Hf(OTf)4 exhibited excellent regioselectivity and diastereoselectivity. The H/D exchange experiments showed that Hf(OTf)4 catalyst could significantly promote the rate of keto-enol tautomerization.

Catalyst (5 mol%) Reaction Time (h) Yield of 1 (%)
The solvent effect was investigated in the presence of 5 mol% Hf(OTf)4. As listed in Table 2 (entry 1-4), the reactions in THF, DME, benzene, and CH2Cl2 proceeded much slower with good to moderate yields. Compared to the reaction in acetonitrile, the one performed in EtOH resulted in shorter reaction time and higher yield (94% yield, 5 h). But when the reaction in EtOH was elevated to 80 °C, to our surprise, the reaction rate and yield of Hf(OTf)4-catalyzed Mannich reaction was not significantly affected like many other reactions [42][43][44]. Hf(OTf) 4 6 89 a Benzaldehyde/aniline/acetophenone are in a 1:1:2 molar ratio.
The solvent effect was investigated in the presence of 5 mol% Hf(OTf) 4 . As listed in Table 2 (entry 1-4), the reactions in THF, DME, benzene, and CH 2 Cl 2 proceeded much slower with good to moderate yields. Compared to the reaction in acetonitrile, the one performed in EtOH resulted in shorter reaction time and higher yield (94% yield, 5 h). But when the reaction in EtOH was elevated to 80 • C, to our surprise, the reaction rate and yield of Hf(OTf) 4 -catalyzed Mannich reaction was not significantly affected like many other reactions [42][43][44].
As expected, reducing the amount of Hf(OTf) 4 catalyst to 2 mol% for the reaction in EtOH resulted in prolonged reaction time (12 h). In contrast, under solvent-free conditions, 2 mol% Hf(OTf) 4 furnished efficient formation of 1 in only 2 h at room temperature. Further optimization determined that the amount of Hf(OTf) 4 could be reduced to as low as 0.5 mol% for efficient production of 1 (93%, 4 h).
With the optimized conditions, 0.5 mol% Hf(OTf) 4 was applied to the synthesis of a diversity of aryl ketone-derived β-amino carbonyl compounds (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16). As shown in Scheme 1, the current method exhibits good generality to a variety of substrates, and the aryl ketone-derived Mannich bases were obtained in excellent yields (87-94%) within 4-7 h. Electron-donating and electron-withdrawing substituents at the ortho, meta, and para positions of the phenyl rings of ketone, aldehyde, and aniline are well tolerated by this method. 1-4), the reactions in THF, DME, benzene, and CH2Cl2 proceeded much slower with good to moderate yields. Compared to the reaction in acetonitrile, the one performed in EtOH resulted in shorter reaction time and higher yield (94% yield, 5 h). But when the reaction in EtOH was elevated to 80 °C, to our surprise, the reaction rate and yield of Hf(OTf)4-catalyzed Mannich reaction was not significantly affected like many other reactions [42][43][44]. As expected, reducing the amount of Hf(OTf)4 catalyst to 2 mol% for the reaction in EtOH resulted in prolonged reaction time (12 h). In contrast, under solvent-free conditions, 2 mol% Hf(OTf)4 furnished efficient formation of 1 in only 2 h at room temperature. Further optimization determined that the amount of Hf(OTf)4 could be reduced to as low as 0.5 mol% for efficient production of 1 (93%, 4 h).

Alkyl Ketone-Based Mannich Reaction Catalyzed by Hf(OTf)4: Regioselectivity and Diastereoselectivity
Since the reactivity of alkyl ketones are typically higher than that of aryl ketones, the reaction of benzaldehyde, aniline, and acetone (2:1:1 molar ratio) only took 10 min in the presence of 0.5 mol% Hf(OTf)4 under solvent-free conditions. Further optimization determined that only 0.1 mol% Hf(OTf)4 is sufficient to catalyze the high-yielding formation of the Mannich base 17 (94%, 30 min).
However, when more complicated alkyl ketones such as 2-pentanone and 1,1-dimethylacetone were used, the reaction rate was notably slower. It took 4-5 h to yield the corresponding Mannich bases 18 and 24 in good yields. More importantly, comparison with the reactions without catalyst indicated that the presence of Hf(OTf)4 not only promoted the reaction rate but also resulted in high regioselectivity. As shown in Table 3, in contrast to the uncatalyzed reactions, which yielded both isomers (a:b = ~1: 0.5 molar ratio, determined by 1 H NMR), only the less substituted, namely the methyl-derived, isomer (a) was obtained when Hf(OTf)4 was used. The application of the optimized conditions to linear alkyl ketone substrates afforded a diversity of Mannich bases 18-29 in excellent yields (82-91%) within 4-7 h (Scheme 2).

Alkyl Ketone-Based Mannich Reaction Catalyzed by Hf(OTf) 4 : Regioselectivity and Diastereoselectivity
Since the reactivity of alkyl ketones are typically higher than that of aryl ketones, the reaction of benzaldehyde, aniline, and acetone (2:1:1 molar ratio) only took 10 min in the presence of 0.5 mol% Hf(OTf) 4 under solvent-free conditions. Further optimization determined that only 0.1 mol% Hf(OTf) 4 is sufficient to catalyze the high-yielding formation of the Mannich base 17 (94%, 30 min).
However, when more complicated alkyl ketones such as 2-pentanone and 1,1-dimethylacetone were used, the reaction rate was notably slower. It took 4-5 h to yield the corresponding Mannich bases 18 and 24 in good yields. More importantly, comparison with the reactions without catalyst indicated that the presence of Hf(OTf) 4 not only promoted the reaction rate but also resulted in high regioselectivity. As shown in Table 3, in contrast to the uncatalyzed reactions, which yielded both isomers (a:b =~1:0.5 molar ratio, determined by 1 H NMR), only the less substituted, namely the methyl-derived, isomer (a) was obtained when Hf(OTf) 4 was used. The application of the optimized conditions to linear alkyl ketone substrates afforded a diversity of Mannich bases 18-29 in excellent yields (82-91%) within 4-7 h (Scheme 2).   In the following research, we investigated the diastereoselectivity of Hf(OTf)4-catalyzed synthesis of cycloketone-derived Mannich bases 30-32 under solvent-free conditions. We noticed that even residual Hf(IV) cation in the used round-bottom flask may significantly affect the outcome of the anti/syn ratio. To avoid false results from the contamination of the trace amount of residual catalyst, the control reactions without catalyst were all performed in new reaction vessels with new stir bars. In addition, the ratio of anti/syn isomers was determined directly from 1 H NMR of the crude product. The results listed in Table 4 showed that the Mannich reaction of cyclopentanone did not yield the desired product at all after 12 h without catalyst. For cyclohexanone and cycloheptanone, the corresponding reactions were sluggish and poor yielding. When 0.1 mol% Hf(OTf)4 was applied, both the reaction rates and yields of these reactions were notably improved. The presence of Hf(IV) cation dramatically increased the ratio of anti/syn isomers from 63:37 up to 96:4 when cyclohexanone was used. For cyclopentanone, Hf(IV) cation also favored the formation of anti isomer (anti/syn = 92:8). But increasing the amount of Hf(OTf)4 to 1 mol% did not further improve the ratio of anti/syn isomers. Interestingly, in the case of cycloheptanone, solvent-free conditions alone favored the formation of syn isomer, but addition of 0.1 mol% Hf(OTf)4 increased the ratio of anti/syn isomers from originally 20:80 to 59:41. It was observed that higher Hf(OTf)4 loading resulted in remarkable increase in both reaction rate and anti/syn ratio, which could reach up to anti/syn = 86:14 when 50 mol% Hf(OTf)4 was used.    In the following research, we investigated the diastereoselectivity of Hf(OTf)4-catalyzed synthesis of cycloketone-derived Mannich bases 30-32 under solvent-free conditions. We noticed that even residual Hf(IV) cation in the used round-bottom flask may significantly affect the outcome of the anti/syn ratio. To avoid false results from the contamination of the trace amount of residual catalyst, the control reactions without catalyst were all performed in new reaction vessels with new stir bars. In addition, the ratio of anti/syn isomers was determined directly from 1 H NMR of the crude product. The results listed in Table 4 showed that the Mannich reaction of cyclopentanone did not yield the desired product at all after 12 h without catalyst. For cyclohexanone and cycloheptanone, the corresponding reactions were sluggish and poor yielding. When 0.1 mol% Hf(OTf)4 was applied, both the reaction rates and yields of these reactions were notably improved. The presence of Hf(IV) cation dramatically increased the ratio of anti/syn isomers from 63:37 up to 96:4 when cyclohexanone was used. For cyclopentanone, Hf(IV) cation also favored the formation of anti isomer (anti/syn = 92:8). But increasing the amount of Hf(OTf)4 to 1 mol% did not further improve the ratio of anti/syn isomers. Interestingly, in the case of cycloheptanone, solvent-free conditions alone favored the formation of syn isomer, but addition of 0.1 mol% Hf(OTf)4 increased the ratio of anti/syn isomers from originally 20:80 to 59:41. It was observed that higher Hf(OTf)4 loading resulted in remarkable increase in both reaction rate and anti/syn ratio, which could reach up to anti/syn = 86:14 when 50 mol% Hf(OTf)4 was used. In the following research, we investigated the diastereoselectivity of Hf(OTf) 4 -catalyzed synthesis of cycloketone-derived Mannich bases 30-32 under solvent-free conditions. We noticed that even residual Hf(IV) cation in the used round-bottom flask may significantly affect the outcome of the anti/syn ratio. To avoid false results from the contamination of the trace amount of residual catalyst, the control reactions without catalyst were all performed in new reaction vessels with new stir bars. In addition, the ratio of anti/syn isomers was determined directly from 1 H NMR of the crude product. The results listed in Table 4 showed that the Mannich reaction of cyclopentanone did not yield the desired product at all after 12 h without catalyst. For cyclohexanone and cycloheptanone, the corresponding reactions were sluggish and poor yielding. When 0.1 mol% Hf(OTf) 4 was applied, both the reaction rates and yields of these reactions were notably improved. The presence of Hf(IV) cation dramatically increased the ratio of anti/syn isomers from 63:37 up to 96:4 when cyclohexanone was used. For cyclopentanone, Hf(IV) cation also favored the formation of anti isomer (anti/syn = 92:8). But increasing the amount of Hf(OTf) 4 to 1 mol% did not further improve the ratio of anti/syn isomers. Interestingly, in the case of cycloheptanone, solvent-free conditions alone favored the formation of syn isomer, but addition of 0.1 mol% Hf(OTf) 4 increased the ratio of anti/syn isomers from originally 20:80 to 59:41. It was observed that higher Hf(OTf) 4 loading resulted in remarkable increase in both reaction rate and anti/syn ratio, which could reach up to anti/syn = 86:14 when 50 mol% Hf(OTf) 4 was used.

The Catalytic Role of Hf(OTf)4 on Keto-Enol Tautomerization
In our previous research, we have revealed Hf(IV) cation's strong capability on activating benzaldehyde for the fast formation of the imine intermediate [45]. Many previous reports had also proposed that the interactions of metal cations with ketone are also involved in the catalysis of Mannich reaction. However, not much evidence has been provided to support this point. In the current research, we utilized 1 H NMR to examine the activation effects of Hf(IV) on both aryl ketone and alkyl ketone. Interestingly, when 5 mol% Hf(OTf)4 was added to acetophenone in MeOH-d4, a remarkable H/D exchange process was promoted. As shown in Figure 1A, 86.6% of the proton of the active methyl group was exchanged to deuterium over 36 h. For cyclopentanone, only 1 mol% Hf(OTf)4 was needed to result in a comparable H/D exchange process (87.5%, 36 h), which is in agreement with the result that less catalyst is required for alkyl ketone-based Mannich reaction. Similar to the promoted tautomerization of dimethylphosphite in Kabachnik reaction [45], the coordination of Hf(IV) dramatically accelerated the tautomerization between the ketone and enol forms of both aryl and alkyl ketone, thereby increasing the overall reaction rate.

The Catalytic Role of Hf(OTf) 4 on Keto-Enol Tautomerization
In our previous research, we have revealed Hf(IV) cation's strong capability on activating benzaldehyde for the fast formation of the imine intermediate [45]. Many previous reports had also proposed that the interactions of metal cations with ketone are also involved in the catalysis of Mannich reaction. However, not much evidence has been provided to support this point. In the current research, we utilized 1 H NMR to examine the activation effects of Hf(IV) on both aryl ketone and alkyl ketone. Interestingly, when 5 mol% Hf(OTf) 4 was added to acetophenone in MeOH-d 4 , a remarkable H/D exchange process was promoted. As shown in Figure 1A, 86.6% of the proton of the active methyl group was exchanged to deuterium over 36 h. For cyclopentanone, only 1 mol% Hf(OTf) 4 was needed to result in a comparable H/D exchange process (87.5%, 36 h), which is in agreement with the result that less catalyst is required for alkyl ketone-based Mannich reaction. Similar to the promoted tautomerization of dimethylphosphite in Kabachnik reaction [45], the coordination of Hf(IV) dramatically accelerated the tautomerization between the ketone and enol forms of both aryl and alkyl ketone, thereby increasing the overall reaction rate.

The Catalytic Role of Hf(OTf)4 on Keto-Enol Tautomerization
In our previous research, we have revealed Hf(IV) cation's strong capability on activating benzaldehyde for the fast formation of the imine intermediate [45]. Many previous reports had also proposed that the interactions of metal cations with ketone are also involved in the catalysis of Mannich reaction. However, not much evidence has been provided to support this point. In the current research, we utilized 1 H NMR to examine the activation effects of Hf(IV) on both aryl ketone and alkyl ketone. Interestingly, when 5 mol% Hf(OTf)4 was added to acetophenone in MeOH-d4, a remarkable H/D exchange process was promoted. As shown in Figure 1A, 86.6% of the proton of the active methyl group was exchanged to deuterium over 36 h. For cyclopentanone, only 1 mol% Hf(OTf)4 was needed to result in a comparable H/D exchange process (87.5%, 36 h), which is in agreement with the result that less catalyst is required for alkyl ketone-based Mannich reaction. Similar to the promoted tautomerization of dimethylphosphite in Kabachnik reaction [45], the coordination of Hf(IV) dramatically accelerated the tautomerization between the ketone and enol forms of both aryl and alkyl ketone, thereby increasing the overall reaction rate.

General Methods
General chemical reagents and solvents were obtained from commercial suppliers. All reactions were monitored by thin layer chromatography on plates coated with 0.25 mm silica gel 60 F254 (Qingdao Haiyang Chemicals, Qingdao, China). TLC plates were visualized by UV irradiation (254 nm, Shanghai Peiqing Sci & Tech, Shanghai, China). Flash column chromatography employed silica gel (particle size 32-63 µm, Qingdao Haiyang Chemicals, Qingdao, China). Melting points were determined with a Thomas-Hoover melting point apparatus (Thomas Scientific, Swedesboro, NJ, USA) and uncorrected. NMR spectra were obtained with a Bruker AV-400 instrument (Bruker BioSpin, Faellanden, Switzerland) with chemical shifts reported in parts per million (ppm, δ) and referenced to CDCl 3 or DMSO-d 6 .