Open Access Article
Zhenbiao Luo
*,
Mulin Huang,
Xiangyong Wang,
Xinye Wang,
Zhihui Hu,
Liang Zhao and
Ting Huang
Department of Brewing Engineering, Moutai Institute, Guizhou, Renhuai 564507, China. E-mail: Luozhenbiao@mtxy.edu.cn
First published on 6th October 2025
A green and simple method for the synthesis of benzimidazoles by amino acid catalysis of o-phenylenediamines with α-keto acids has been developed. The reaction proceeded smoothly in water under air and metal catalyst-free conditions to obtain 2-substituted benzimidazoles. Furthermore, the method was performed in an open system at room temperature, making the experimental operation very simple and greatly enhancing the accessibility of the method.
The versatile applications of benzimidazoles have sparked a surge of interest among synthetic chemists, leading to the proliferation of modified synthetic methodologies in recent years.18–20 In addition to the commonly used iron,21,22 copper,23 silver,24 and palladium catalyzed25,26 methods for synthesizing benzimidazoles, photochemical27–29 and electrochemical30,31 approaches have also been explored. Some groups have even used novel materials.32–39 Some reports have introduced metal free methods.40,41 Recently, Biswas and co-workers demonstrated the synthesis of benzimidazole under visible light irradiation in the presence of 3,6-di(pyridin-2-yl)-1,2,4,5-tetrazine.42 Besides, the acid catalyzed method has also been developed by other researchers.43–45 Our group has previously developed an acid-promoted approach.46 Although this method carried out under ambient conditions in an open flask, it still required 1.0 equivalent of acid, and o-phenylenediamine necessitated a protecting group. Therefore, we wanted to explore a new method of synthesizing benzimidazole without protecting group in an open flask.
In our previous studies, we discovered that α-keto acids reacted with o-phenylenediamines to afford benzimidazoles, and some methods suggested that amino acids exhibit catalytic activity.47,48 Consequently, we explored the use of amino acids as catalysts (Scheme 1). Initially, the reaction between o-phenylenediamine (1a) and pyruvic acid (2a) was conducted in water at room temperature. Disappointingly, only 3-methylquinoxaline-2(1H)-one (4a) was obtained, which was obviously not the desired product (Table 1, entry 1). Similar results have also been reported in previous reports.49 Subsequently, each of the 20 standard amino acids was individually screened as an additive. For brevity, only those yielding 2-methyl-1H-benzo[d]imidazole (3a) were reported herein. Fortunately, the addition of L-Lysine and L-arginine afforded 3a (Table 1, entries 2,3), while the remaining 18 amino acids failed (see SI for details). It was evident that a significant amount of 4a was produced. Next, D-lysine and D-arginine were used for experiments respectively, and the yield of 3a did not enhance significantly (Table 1, entries 4, 5), even if (±)-lysine and (±)-arginine were added (Table 1, entries 6, 7). This suggested that the amino acid configuration did not play a decisive role in improving the yield. With the increase of the amount of amino acids added, the yield of 3a decreased, mainly because 1a was not completely consumed (Table 1, entries 8, 9). Subsequently, the test of adding two different amino acids was detected. The addition of L-lysine and L-arginine did not improve the yield of 3a (Table 1, entry 10), whereas the combination of L-lysine/L-Arg with L-tyrosine made it (Table 1, entries 11, 12). A reaction was conducted by adding L-phenylalanine with similar structure to L-tyrosine, but no increase in the yield was detected (Table 1, entry 13). Solvent screening was then performed: MeCN, DMF, and EtOH provided 3a, but with a lower yield (Table 1, entries 14–16). THF, DMSO or DCM were used as solvents, the target product was not obtained (Table 1, entries 17–19). Considering that decarboxylation might be involved in the reaction, temperature optimization was performed. The yield almost unchanged at low temperature and slowed down the reaction (Table 1, entry 20). As the temperature increased, the reaction time also shortened, and the yield was improved at 50 °C (Table 1, entries 21, 22). However, when the temperature rose to 60 °C, the yield slightly decreased (Table 1, entry 23), and then the temperature reached 80 °C, the yield showed a significant decrease (Table 1, entry 24). To verify the effectiveness of amino acids in the reaction at high temperatures, a control experiment was conducted at 50 °C without adding amino acids. As a result, 3a was not obtained but 4a (Table 1, entry 25). Considering the instability of pyruvic acid, it was divided into two and four batches. It was obvious that the effect of batch addition was better than that of single addition (Table 1, entries 26–28). In order to investigate whether the reaction involved free radicals, 2,2,6,6-tetramethyl-1-piperidinooxy free radical (TEMPO) was added, and the result showed that TEMPO had almost no effect (Table 1, entry 29). Because the pH value affected the chemical properties of amino acids, we adjusted the pH value with HCl solution and NaOH solution. No 3a was found in the acidic system, but the yield of 3a increased with the increase of pH value (Table 1, entries 30–33). We hypothesized that 3a was obtained in the presence of base. Further experiments with different bases showed that 3a was obtained when the amount of base was added to 1.2 equivalents. The buffer solution increased the yield of 3a in the absence of amino acids. This suggested that the addition of Tyr to form the buffer enhanced the reaction conditions, thereby improving the yield of 3a (Table 1, entries 34, 35). Water as solvent in the presence of NaOH, 3a was obtained at low yield (Table 1, entry 36), while ethanol as solvent in the presence of t-BuOK, the yield was slightly improved (Table 1, entry 37). Unfortunately, many substrates did not afford the target product in the presence of base, or the yield was very low, so this method was abandoned. Considering the convenience of the experimental operation, we concluded that the optimized reaction conditions as follows: o-phenylenediamine (1.0 eq), lysine (0.25 eq), tyrosine (0.1 eq) and α-keto acid (1.2 eq) in water and in an open flask at room temperature.
| Entry | Additive | Solvent | T (°C) | Yield (%)/3a | Yield (%)/4a |
|---|---|---|---|---|---|
| a Reaction conditions: substrate 1a (0.5 mmol), 1b (0.6 mmol), amino acids, water (15.0 mL) in an open flask.b Pyruvic acid was divided into two batches.c Pyruvic acid was divided into four batches. | |||||
| 1 | — | H2O | 25 | — | 91 |
| 2 | L-lys (0.25 eq) | H2O | 25 | 47 | 12 |
| 3 | L-Arg (0.25 eq) | H2O | 25 | 45 | 12 |
| 4 | D-Lys (0.25 eq) | H2O | 25 | 48 | 12 |
| 5 | D-Arg (0.25 eq) | H2O | 25 | 46 | 12 |
| 6 | (±)-Lys (0.25 eq) | H2O | 25 | 48 | 12 |
| 7 | (±)-Arg (0.25 eq) | H2O | 25 | 45 | 11 |
| 8 | L-Lys (0.50 eq) | H2O | 25 | 38 | 7 |
| 9 | L-Arg (0.50 eq) | H2O | 25 | 36 | 7 |
| 10 | L-Lys (0.06eq), L-Arg (0.06eq) | H2O | 25 | 46 | 12 |
| 11 | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 25 | 53 | 7 |
| 12 | L-Arg (0.25eq), L-Tyr (0.10eq) | H2O | 25 | 51 | 7 |
| 13 | L-Lys (0.25eq), L-Phe (0.10eq) | H2O | 25 | 41 | 5 |
| 14 | L-Lys (0.25eq), L-Tyr (0.10eq) | MeCN | 25 | 13 | 31 |
| 15 | L-Lys (0.25eq), L-Tyr (0.10eq) | DMF | 25 | 13 | 27 |
| 16 | L-Lys (0.25eq), L-Tyr (0.10eq) | EtOH | 25 | 21 | 13 |
| 17 | L-Lys (0.25eq), L-Tyr (0.10eq) | THF | 25 | — | 65 |
| 18 | L-Lys (0.25eq), L-Tyr (0.10eq) | DMSO | 25 | — | 63 |
| 19 | L-Lys (0.25eq), L-Tyr (0.10eq) | DCM | 25 | — | 51 |
| 20 | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 0 | 52 | 8 |
| 21 | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 37 | 53 | 7 |
| 22 | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 50 | 61 | 7 |
| 23 | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 60 | 59 | 7 |
| 24 | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 80 | 35 | 21 |
| 25 | — | H2O | 50 | — | 89 |
| 26b | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 25 | 65 | 7 |
| 27c | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 25 | 69 | 7 |
| 28c | L-Lys (0.25eq), L-Tyr (0.10eq) | H2O | 50 | 71 | 1 |
| 29 | L-Lys (0.25eq), L-Tyr (0.10eq), TEMPO (1.00 eq) | H2O | 25 | 63 | 8 |
| 30 | L-Lys (0.25eq), L-Tyr (0.10eq), pH = 4 | H2O | 25 | — | 80 |
| 31 | L-Lys (0.25eq), L-Tyr (0.10eq), pH = 6 | H2O | 25 | — | 83 |
| 32 | L-Lys (0.25eq), L-Tyr (0.10eq), pH = 8 | H2O | 25 | 51 | 8 |
| 33 | L-Lys (0.25eq), L-Tyr (0.10eq), pH = 10 | H2O | 25 | 53 | 8 |
| 34 | [0.1 M] PBS buffer, pH = 8 | H2O | 25 | 43 | 4 |
| 35 | [0.1 M] PBS buffer, pH = 10 | H2O | 25 | 48 | 5 |
| 36 | NaOH (1.2 eq) | H2O | 25 | 35 | 12 |
| 37 | t-BuOK (1.2 eq) | EtOH | 25 | 45 | 11 |
Under the optimized reaction conditions, we explored the substrate scope of α-keto acids and o-phenylenediamines to assess the practicality of the reaction (Scheme 2). Electron-rich 4,5-dimethyl- and 4-methoxy substituted o-phenylene-diamines afforded the target products with excellent yields (3b, 3c). 3d gave the desired product, indicating the tolerance of halogen groups. Moreover, 2,3-diaminonaphthalene also obtained the corresponding product 3e. It was noteworthy that o-phenylenediamines bearing electron withdrawing substituents such as cyano and nitro group afforded 3f and 3g with moderate yields. Subsequently, the scope of α-keto acids was further investigated. Using 2-ketobutyric acid or 4-methyl-2-oxovanoic acid, the yield was better than pyruvic acid (3h–3u). Obviously, the yields of benzimidazoles were significantly improved with the growth of alkyl chain of α-keto acids. Phenylpyruvic acid was adopted, but the target product was not isolated.
In order to explore the practicability of this method in the synthesis of other heterocyclic compounds (Table 2). 2-Aminophenol and 2-aminothiophenol reacted with α-keto acid respectively. Unfortunately, benzoxazoles were not obtained under this reaction condition, but benzothiazoles were easily afforded (6a–6e). Benzothiazoles were obtained without the addition of amino acids, which was confirmed in previous reports.50 Under the optimized conditions, the yields of benzothiazoles were appropriately increased by using water as solvent in an open flask at room temperature.
Based on the above results and previous report,51 a possible reaction mechanism was proposed (Scheme 3). Lysine reacted with the α-keto acid (2) to afforded α-oxocarboxylate A and Tyr⋯Lys+–H. Subsequently, Condensation of o-phenylene-diamine (1) with A to obtain hemiaminal B. The decarboxylation of B was then facilitated by the abstraction of H+ from Lys+–H to afford the tetrahedral intermediate C and Lys.52 Benzimidazoline D was formed by the favored 5-exo-tet cyclization.53 Finally, the dehydrogenation of D in air obtained the desired benzimidazole 3 (Path A). On the other hand, o-phenylenediamine (1) condensed with α-keto acid (2) to form hemiamines E. Subsequent two steps of dehydration, 4 was obtained (Path B).
| This journal is © The Royal Society of Chemistry 2025 |