2024 Volume 72 Issue 1 Pages 75-79
The oxidative cleavage reaction of pyrrolidine-2-methanols to γ-lactams has been described. In this reaction, [4-iodo-3-(isopropylcarbamoyl)phenoxy]acetic acid and powdered Oxone (2KHSO5·KHSO4·K2SO4) were employed as the catalyst and co-oxidant, respectively. The reaction is efficient and environmentally benign because it produces various lactams from readily available substrates in moderate to excellent yields using organocatalyst and inorganic non-toxic co-oxidant.
Lactams are ubiquitous structural motifs found in both natural products and pharmaceuticals. To date many synthetic methods have been reported for their synthesis.1–6) The oxidative transformations of azacycles 1 and 2 with carboxy and hydroxymethyl groups, respectively, at the 2-position have been developed to synthesize the corresponding lactams 3 (Chart 1). These transformations are useful because several substrates, such as 1 and 2, are readily available from commercial source. Moreover, the preparation of 1 and 2 with various substituents and/or additional stereocenters can be easily accomplished from commercially available compounds. In addition, oxidation can occur selectively at the 2-position to produce regioselective lactams, such as 2-pyrrolidone and 2-pipelidone. Oxidative transformations of carboxylic acids 1 to 3 have already been reported, including stoichiometric oxidations,7–11) transition-metal catalyzed oxidations,12–15) photochemical16,17) and electrochemical18,19) reactions. The transformation of methanols 2 to 3 has been achieved using transition metal-catalyzed20–22) and non-catalytic23) oxidations. However, these methods have several limitations such as low versatility and low yield. They also require toxic reagents, multiple steps, and special equipment. Therefore, efficient oxidative transformations of 1 or 2 to 3 with environmentally benign oxidants under heavy metal-free conditions should be developed.
We previously reported the oxidative cleavage of tetrahydrofuran-2-methanols 4 to γ-lactones 5 using a catalytic amount of 2-iodobenzamide 6 in the presence of an inorganic co-oxidant, Oxone (2KHSO5·KHSO4·K2SO4)24) (Chart 2a). This reaction did not require the use of heavy metal reagents and proceeded smoothly under environmentally benign conditions to produce lactones in high to excellent yield. We also applied this reaction to the total synthesis of tanikolide.25) We envisioned that this reaction could be applied to azacycles as an efficient oxidative transformation for the synthesis of lactams. Herein, we report the oxidative cleavage of pyrrolidine-2-methanols 7 to γ-lactams 8 using 2-iodobenzamide 6 as the catalyst (Chart 2b).
For the initial screening of the oxidative cleavage of azacycles, we first performed the reactions of N-carbobenzyloxy(Cbz)-L-prolinol (7a), L-prolinol (9), and N-Cbz-L-proline 10 under the reported conditions24) for the oxidative cleavage of 4 to 5 with catalyst 6 (Chart 3). Cbz-prolinol 7a was treated with 5 mol% 6 in the presence of 4 equivalents (equiv.) of Oxone in a 10 : 1 mixture of MeNO2 and N,N-dimethylformamide (DMF) at 50 °C. Fortunately, the reaction was completed within 7 h, generating the corresponding γ-lactam 8a in 78% yield (Chart 3a). Catalyst 6 was easily separated from the product 8a without chromatography and recovered quantitatively. Without catalyst 6 the oxidation of 7a did not proceed (Chart 3b). N-Unprotected prolinol 9 was oxidized using 6, but only decomposed products were obtained (Chart 3c). The oxidation of N-protected proline 10 produced only 4% of 8a after 7 h (Chart 3d). Proline (11) did not react under these reaction conditions after 36 h (Chart 3e), although 11 slowly reacted with the trivalent iodine reagent [(PhIO)n] to produce 2-pyrrolidone in moderate yield after 2 d.8) These results indicate that the oxidative cleavage reaction of N-protected pyrrolidine-2-methanols with a catalytic amount of 6 and Oxone proceeded smoothly to give N-protected 2-pyrrolidone.
Next, we investigated the effect of a co-oxidant on the reaction by using 7a as the substrate (Table 1). When n-Bu4NHSO5, which was prepared from Oxone and n-Bu4NHSO4 according to the literature,26) was used as an acid-free co-oxidant, the oxidation of 7a did not occur after 5 h. Furthermore, the addition of 1 equiv. of KHSO4 as an acid led to the formation of the desired lactam 8a (entry 2). However, a longer reaction time was required, resulting in lower yield. The use of powdered Oxone27) increased the yield (80%) of 8a in a shorter reaction time (5 h) (entry 3). In contrast, the use of buffered Oxone,28) which is powdered Oxone neutralized with K2CO3, resulted in a longer reaction time (36 h) (entry 4). These results suggest that an acid is necessary for faster oxidation and powdered Oxone is the best co-oxidant.
 | ||||
|---|---|---|---|---|
| Entry | Co-oxidant | Time (h) | Yield of 8a (%) | Recovery of 7a (%) |
| 1 | Oxone | 7 | 7 | — |
| 2b) | n-Bu4NHSO5c) | 36 | 44 | 15 |
| 3 | Powdered Oxoned) | 5 | 80 | — |
| 4 | Buffered Oxonee) | 36 | 80 | — |
In the oxidative cleavage of tetrahydrofuran-2-methanols 4 to lactones 5 using a catalyst 6,24) the choice of solvent is very important. Therefore, we investigated the effect of different solvents on oxidative cleavage reaction using catalyst 6 and powdered Oxone (Table 2). When 7a was oxidized in a 1 : 1 mixture of MeNO2 and DMF, the reaction was incomplete even after 36 h (entry 2). The yield (55%) of lactam 8a was very low compared to that obtained using 10 : 1 MeNO2–DMF (entry 1). We hypothesized that it would be impossible to increase the yield within a short reaction time in a mixed solvent system of MeNO2 and DMF. Therefore, we employed MeNO2 as the sole solvent; this reaction was almost complete within 36 h, affording 8a in 62% yield (entry 3). The reaction in which DMF was used as the sole solvent did not afford the desired product (entry 4). The use of MeCN resulted in a short reaction time (9 h) (entry 5). However, the yield (62%) of 8a was still low. Interestingly, when dimethyl carbonate [CO(OMe)2]28) was used as the solvent, 8a was obtained in 95% yield after 46 h (entry 6). To decrease the reaction time, we employed a mixed solvent system containing CO(OMe)2 and DMF. When this reaction was conducted in a 10 : 1 mixture of CO(OMe)2 and DMF, 8a was obtained in 93% yield after 9 h (entry 7). Different ratios of CO(OMe)2 to DMF did not improve the yield of 8a or decrease the reaction time (entries 8 and 9). The use of a 10 : 1 mixture of CO(OMe)2 and MeNO2 resulted in a lower yield (81%) of 8a over a longer reaction time (36 h) (entry 10). Based on these results, the oxidative cleavage of pyrrolidine-2-methanols 7 with 5 mol% of 6 and 4 equiv. of powdered Oxone® in a 10 : 1 CO(OMe)2–DMF mixture at 50 °C would be the optimal reaction conditions.
 | ||||
|---|---|---|---|---|
| Entry | Solvent (ratio) | Time (h) | Yield of 8a (%) | Recovery of 7a (%) |
| 1 | MeNO2–DMF (10 : 1) | 5 | 80 | — |
| 2 | MeNO2–DMF (1 : 1) | 36 | 55 | 37 |
| 3 | MeNO2 | 36 | 62 | trace |
| 4 | DMF | 36 | 0 | 33 |
| 5 | MeCN | 9 | 62 | — |
| 6 | CO(OMe)2b) | 46 | 95 | — |
| 7 | CO(OMe)2–DMF (10 : 1) | 9 | 93 | — |
| 8 | CO(OMe)2–DMF (1 : 1) | 36 | 10 | 64 |
| 9 | CO(OMe)2–DMF (20 : 1) | 13 | 84 | — |
| 10 | CO(OMe)2– MeNO2 (10 : 1) | 36 | 81 | 5 |
a) All reactions were performed using 7a (0.25 mmol), 6 (5 mol%), and powdered Oxone (1.0 mmol) in solvent (1.1 mL) at 50 °C unless otherwise specified. b) Ref. 28.
With the optimized reaction conditions in hand, we examined the oxidative cleavage of various pyrrolodine-2-methanols 7b–n (Table 3). We first focused on prolinols 7b–h containing various N-protecting groups as substrates. When N-Fmoc (9-fluorenylmethoxycarbonyl) prolinol 7b was oxidized with 5 mol% of 6 and 4 equiv. of powdered Oxone in a 10 : 1 mixture of CO(OMe)2 and DMF at 50 °C, the oxidation was completed within 11 h to afford the corresponding lactam 8b in 84% yield (entry 1). In contrast, the reaction of N-tert-butoxycarbonyl (Boc) prolinol 7c produced a complex mixture (entry 2), presumably because of the deprotection of the N-Boc group under acidic reaction conditions. To prevent this deprotection, we employed buffered Oxone as a co-oxidant. Interestingly the reaction of 7c successfully afforded lactam 8c in 60% yield after 15 h (entry 2). N-Acylated prolinols 7d and 7e were oxidized to afford the corresponding lactams 8d and 8e, respectively. However, the yields of 8d (56%) and 8e (34%) were relatively low (entries 3 and 4). N-Sulfonylated substrates were also investigated. The oxidation of N-Ts(tosyl) prolinol 7f afforded 8f in 84% yield after 13 h (entry 5). However, the reaction of N-nitrobenzenesulfonyl prolinols 7g and 7h resulted in lower yields (8g: 55%, 8h: 58%) and longer reaction times (51 h) (entries 6 and 7). These results suggest that urethane-type protecting groups and tosyl group are suitable for this reaction. Next, we examined the reactions of N-Cbz pyrrolidine-2-methanols with various substituents on the pyrrolidine ring. The oxidation of 5-propylpyrrolidine-2-methanols 7i afforded the corresponding lactam 8i in 87% yield after 9 h (entry 8). The 4-acetoxy derivative 7j was reacted to produce 8j in 81% yield after 13 h (entry 9). The 4-benzoyloxy derivative 7k was oxidized within 24 h to afford 8k in 73% yield (entry 10). Oxidation of the benzyloxy-substituted derivative 7l produced 8l in 93% yield; however, this reaction required di-tert-butyl-4-methylphenol (BHT)24,25) to prevent the undesired C–H oxidation at the benzylic position (entry 11). Pyrrolidine-2-methanol 7m, which contained an ethoxycarbonylmethoxy group at the 4-position was oxidized to give 8m in 83% yield after 10 h (entry 12). The oxidation of the 4-cyano derivative 7n also afforded lactam 8n in 74% yield although this transformation required a long reaction time (40 h) (entry 13). In case of oxidation of indoline-2-methanol 7o, indole derivative 12 was obtained instead of the desired lactam 8o (entry 14). Unfortunately, the reaction of N-Cbz piperidine-2-methanol under the same conditions gave complex mixture after 24 h.
 |
a) All reactions were performed using 7 (0.25 mmol), 6 (5 mol%), and powdered Oxone (1.0 mmol) in a 10 : 1 mixture of CO(OMe)2 and DMF (1.1 mL) at 50 °C unless otherwise specified. b) Buffered Oxone was used as the co-oxidant. c) One equiv. of di-tert-butyl-4-methylphenol (BHT) was added. d) Indole derivative 12 was obtained in 37% yield. 
To understand the reaction mechanism, we attempted the reaction of 13 with powdered Oxone without a catalyst as a control (Chart 4). The reaction afforded the corresponding lactam 8a in 80% yield. A possible mechanism for the oxidative cleavage of pyrrolidine-2-methanols 7 to γ-lactams 8 using catalyst 6 and powdered Oxone is essentially the same as that of tetrahydrofuran-2-methanol 4 to lactone 529) (Chart 5). Thus, the reaction of catalyst 6 with powdered Oxone produces pentavalent iodine 15. Substrate 7 is oxidized with 15 to generate aldehyde 16. Subsequently, 16 is further oxidized with Oxone to directly afford lactam 8 (see Supplementary Materials, Chart S1), and the intermediate, carboxylic acid 1, is not formed. The resulting trivalent iodine 14 is oxidized again by Oxone to form pentavalent iodine 15.
We found the oxidative cleavage of pyrrolidine-2-methanols 7 to γ-lactams 8. Various types of pyrrolidine-2-methanols were reacted with 5 mol% 2-iodobenzamide 6 and 4 equiv. powdered Oxone in a 10 : 1 mixture of CO(OMe)2 and DMF at 50 °C to produce 8 in moderate to excellent yields. This reaction is an efficient and environmentally benign oxidative transformation for the synthesis of lactams; it proceeds smoothly without the use of heavy metals, organic oxidants, or special equipment to afford lactams in moderate to excellent yields.
The authors declare no conflict of interest.
This article contains supplementary materials.
