Xiaojun Pan,
Qiao Liu,
Liming Chang and
Gaoqing Yuan*
School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou, China. E-mail: gqyuan@scut.edu.cn
First published on 4th June 2015
A simple and efficient method has been developed for the synthesis of symmetrical and unsymmetrical pyridines via NH4I-promoted cyclization of ketones with DMSO and NH4OAc. It was found that methyl ketones always gave selective formation of the unsymmetrical pyridine, while non-methyl ketones gave unpredictable results (symmetrical or non-symmetrical product only, or a mixture of the two). In addition, the deuterium-labeling experiments indicated that the C4 or C6 of the target product pyridine rings resulted from DMSO.
Recently, Guan and co-workers reported a ruthenium-catalyzed cyclization of ketoxime acetates with DMF for the synthesis of symmetrical pyridine, in which DMF acts as both C4 source and reaction medium (Scheme 1a).5 Different from Guan's work, Deng and co-workers developed a method for construction of unsymmetrical pyridines by employing DMF as C6 source (Scheme 1b).6 DMSO, a cheap and low-toxic solvent, has been widely used in organic synthesis. In addition, DMSO can be used as a multipurpose precursor for the formation of –Me,7 –SMe,8 –SO2Me,9 –CHO,10 –CN,11 and –OH.12 Based on DMSO playing multiple roles in the organic synthesis, we have reported an efficient ammonium iodide-induced sulfonylation of alkenes for the synthesis of vinyl methyl sulfones by using DMSO and H2O as the source of –SO2Me,13 and the coupling of alkenes with DMSO and alcohols for the synthesis of β-alkoxy methyl sulfides by employing DMSO as the source of –SMe.14 In the experimental process, we found that DMSO was easily subjected to decomposition to produce MeSH and HCHO. The formation of formaldehyde makes the utilization of DMSO as a new carbon source for the synthesis of substituted pyridines become possible. Herein, we report an efficient ammonium iodide-promoted cyclization of ketones with DMSO and ammonium acetate for synthesis of unsymmetrical and symmetrical pyridines by employing DMSO as C4 or C6 source, respectively (Scheme 1c).
We initially employed acetophenone (1a) as model substrate to optimize the reaction conditions. In the presence of NH4I, acetophenone 1a was converted to 2,4-diarylpyridines (2a) in 22% yield without ammonium acetate in DMSO at 130 °C for 14 h (Table 1, entry 1). Instead of NH4I, 33% yield of 2a was obtained when equal amount of ammonium acetate was added into this reaction system (Table 1, entry 2). To our delight, the yield of 2a could be increased from 33% to 87% when the reaction was performed in the presence of 0.5 equiv. of NH4I and 1.0 equiv. of NH4OAc (Table 1, entry 3). However, the efficiency of this transformation decreased significantly when the reaction was carried out at 120 °C (Table 1, entry 4). Subsequently, the effect of solvents on the reaction was also investigated. Unfortunately, only a little or even trace amount of 2a was detected in DMF, DMA and acetonitrile (Table 1, entries 5–7). These results indicated that the solvent was probably involved in this transformation. In addition, the source of ammonia was also varied and it was observed that ammonium acetate was the most suitable ammonia source for this reaction (Table 1, entries 8–10). Compared to NH4I, a similar ammonium salt NH4Br or NH4Cl resulted in a lower yield (Table 1, entries 11 and 12). Neither NaI nor tetrabutylammonium iodide (n-Bu4NI) could be used to facilitate this reaction (Table 1, entries 13 and 14). When employing 0.25 equiv. of I2 instead of 0.5 equiv. of NH4I, only 25% yield of 2a was obtained (Table 1, entry 15). Compared to I2, HI could afford a higher yield of 2a (Table 1, entry 16). In addition, different ratios of HI and I2 have a great effect on this reaction (Table 1, entries 17–19). Obviously, a high initial concentration of I2 is unfavorable to this reaction (Table 1, entry 15). Based on these results, it was proposed that in situ generated I2 from NH4I was an active promoter in this reaction.
Entry | Halide (equiv.) | Ammonia source | Solvent | Yieldb (%) |
---|---|---|---|---|
a Reaction conditions: 1a (1 mmol), solvent (2.0 mL), ammonia source (1 equiv.), 130 °C, 14 h under air in a sealed tube.b Determined by GC based on 1a.c Reaction temperature 120 °C.d I2 (0.125 equiv.), HI (0.25 equiv.).e Under N2 in a sealed tube. | ||||
1 | NH4I (1) | — | DMSO | 22 |
2 | — | NH4OAc | DMSO | 33 |
3 | NH4I (0.5) | NH4OAc | DMSO | 87 |
4c | NH4I (0.5) | NH4OAc | DMSO | 23 |
5 | NH4I (0.5) | NH4OAc | DMF | 6 |
6 | NH4I (0.5) | NH4OAc | DMA | 5 |
7 | NH4I (0.5) | NH4OAc | CH3CN | Trace |
8 | NH4I (0.5) | (NH4)2CO3 | DMSO | 16 |
9 | NH4I (0.5) | NH4HCO3 | DMSO | 6 |
10 | NH4I (0.5) | NH4Cl | DMSO | 5 |
11 | NH4Br (0.5) | NH4OAc | DMSO | 53 |
12 | NH4Cl (0.5) | NH4OAc | DMSO | 60 |
13 | NaI (0.5) | NH4OAc | DMSO | 38 |
14 | n-Bu4NI (0.5) | NH4OAc | DMSO | 36 |
15 | I2 (0.25) | NH4OAc | DMSO | 25 |
16 | HI (0.5) | NH4OAc | DMSO | 50 |
17d | HI/I2 = 2![]() ![]() |
NH4OAc | DMSO | 61 |
18 | HI/I2 = 1![]() ![]() |
NH4OAc | DMSO | 42 |
19 | HI/I2 = 1![]() ![]() |
NH4OAc | DMSO | 30 |
20e | NH4I (0.5) | NH4OAc | DMSO | 75 |
With the optimal reaction conditions established, a number of ketones (1) were investigated to evaluate the generality and scope of this reaction. Very interestingly, methyl ketones always afforded unsymmetrical pyridines, as shown in Table 2. Using various methyl aryl ketones with electron-withdrawing and electron-donating groups as the substrates, the reaction could give the corresponding unsymmetrical products in moderate to good yields (Table 2, entries 1–9). In addition, functional group at the meta positions of the aromatic ring was also applicable for this transformation (Table 2, entries 12–15). However, when ortho-substituted of acetophenone was involved in this reaction, only 40% yield of the desired product was obtained (Table 2, entry 10). The similar results were observed when employing 1-(naphthalen-1-yl)ethanone and 1-(thiophen-2-yl)ethanone as the substrates (Table 2, entries 16 and 17). Besides, 3,4-dimethylacetophenone and 4-phenylbutan-2-one could smoothly perform in this process (Table 2, entries 11 and 18). However, only 21% yield of desired product was detected by GC when employing heptan-2-one as the substrate (Table 2, entry 21). Also, the product was difficult to be separated. No desired products could be obtained when 1-(pyridin-3-yl)ethanone or (E)-4-phenylbut-3-en-2-one was employed as the substrate (Table 2, entries 19 and 20).
Entry | Substrate | Product | Yield | |
---|---|---|---|---|
a Reaction conditions: 1 (1 mmol), NH4I (0.5 equiv.), NH4OAc (1 equiv.), DMSO (2.0 mL), 130 °C, 14 h.b Isolated yield.c GC yield. | ||||
1 | ![]() |
![]() |
R = H | 2a, 81% |
2 | R = F | 2b, 78% | ||
3 | R = Cl | 2c, 72% | ||
4 | R = Br | 2d, 77% | ||
5 | R = I | 2e, 80% | ||
6 | R = CH3 | 2f, 83% | ||
7 | R = OCH3 | 2g, 76% | ||
8 | R = n-butyl | 2h, 60% | ||
9 | R = isopropyl | 2i, 75% | ||
10 | ![]() |
![]() |
2j, 40% | |
11 | ![]() |
![]() |
2k, 65% | |
12 | ![]() |
![]() |
R = F | 2l, 85% |
13 | R = Cl | 2m, 80% | ||
14 | R = Br | 2n, 70% | ||
15 | R = OCH3 | 2o, 75% | ||
16 | ![]() |
![]() |
2p, 45% | |
17 | ![]() |
![]() |
2q, 45% | |
18 | ![]() |
![]() |
2r, 52% | |
19 | ![]() |
![]() |
2s, 0% | |
20 | ![]() |
![]() |
2t, 0% | |
21 | ![]() |
![]() |
2u, c21% |
For non-methyl ketone substrates, the reaction gave unpredictable results (symmetrical or non-symmetrical pyridine only, or a mixture of the two), as shown in Table 3. For example, the reactions utilizing propiophenone and valerophenone as the substrates gave a mixture of two regioisomers in a ratio of 1:
1.3 and 2.3
:
1, respectively (Table 3, entries 1 and 2). The same phenomenon occurred when 3,4-dihydronaphthalen-1(2H)-one and cyclooctanone were used as the substrates (Table 3, entries 3 and 4). Surprisingly, only one desired product was obtained when the substrate above was replaced by 2,3-dihydro-1H-inden-1-one or cycloheptanone (Table 3, entries 7 and 8). It was worth noting that the corresponding product was symmetrical pyridine for 2,3-dihydro-1H-inden-1-one and unsymmetrical pyridine for cycloheptanone. No desired products could be observed if 3-chloro-1-phenylpropan-1-one and 2-hydroxy-1-phenylethanone were employed as the substrates (Table 3, entries 5 and 6).
Several deuterium-labeling experiments were carried out to gain a preliminary insight into the reaction mechanism (Scheme 2). The reaction was carried out successfully under N2 to afford the corresponding product in 75% yield, which indicated that oxygen was not an important factor in this reaction system (Table 1, entry 20). Instead of DMSO, when DMSO-D6 was subjected to the procedure under the standard conditions, 6-deuterated pyridine 2c′ was obtained in 80% yield (Scheme 2, eqn (1)). This result confirmed that the C6 in the pyridine ring was provided by DMSO. In addition, the reaction occurred as well to give 6-deuterated unsymmetrical pyridine 2d′ and 4-deuterated symmetrical pyridine 3f when propiophenone was used, which could also provide the evidence of the source of C6 or C4 (Scheme 2, eqn (2)).
Based on the above experimental results and literature reports,6,15 two possible reaction pathways are proposed in Scheme 3. Initially, the intermediate 2 is formed by the condensation of ketone 1 itself, which can react with ammonium acetate to give imine intermediate 3. Then, intermediate 3 easily undergoes tautomerization leading to an intermediate 4. Meanwhile, the formaldehyde is generated via decomposition of DMSO (eqn (5)), followed by the reaction with 4 to afford intermediate 5. Subsequently, 5 is oxidized to intermediate 6 by I2 resulting from the reaction of DMSO with HI (eqn (4)). The final product 7 is formed by the intramolecular condensation of intermediate 6 (path a). According to the experimental results (Table 2), methyl ketones seem to exclusively follow path a. It is also possible that ketone 1 has priority to react with ammonium acetate to afford imine intermediate 8, which is easily converted to intermediate 9 via tautomerization. Subsequently, the intermediate 9 combines with formaldehyde provided by DMSO to give intermediate 10. With the help of I2, intermediate 10 is oxidized into intermediate 11. Addition of 1 to 11 generates intermediate 12, which gives intermediate 13 via a intramolecular condensation. Finally, losing a molecule of water, intermediate 13 is transformed into the desired product 14 (path b). For non-methyl ketones, the reaction may undergo path a and/or path b. In addition, it should be pointed out that if the imine intermediate 8 was formed from a methyl ketone it would be less likely to tautomerize to enamine 9 due to the lower stability of less-alkyl substituted CC bonds. So methyl ketones do not follow path b.
In conclusion, a convenient and efficient method has been developed for the preparation of symmetrical and unsymmetrical pyridines via ammonium iodide-promoted cyclization of ketones with DMSO and ammonium acetate. In this reaction system, DMSO is used not only as an effective reaction medium, but also as the source of C4 or C6 for the formation of pyridines. It is worth mentioning that a mixture of two regioisomers, or only symmetrical or unsymmetrical product is obtained when non-methyl ketones are used as substrates, while methyl ketones always gives unsymmetrical pyridines. The present work provides a new strategy for the synthesis of symmetrical and unsymmetrical pyridines.
Footnote |
† Electronic supplementary information (ESI) available: Detailed experimental procedures, characterization of products, and NMR spectral charts. See DOI: 10.1039/c5ra07584j |
This journal is © The Royal Society of Chemistry 2015 |