Chengkou Liua,
Zheng Fanga,
Zhao Yangb,
Qingwen Lia,
Shiyu Guoa and
Kai Guo*ac
aCollege of Biotechnology and Pharmaceutical Engineering, Nanjing Tech University, 30 Puzhu South Road, Nanjing, 211816, China. E-mail: guok@njtech.edu.cn; Fax: +86 2558139926; Tel: +86 2558139935
bCollege of Engineering China Pharmaceutical University, 24 Tongjiaxiang, Nanjing, 210003, China
cState Key Laboratory of Materials-Oriented Chemical Engineering, Nanjing Tech University, 30 Puzhu South Road, Nanjing, 211816, China
First published on 1st March 2016
A practical two-step synthesis of α-ketoamides and α-amino ketones via direct oxidative coupling between 2° benzylic alcohols and amines was developed. Hydrogen peroxide, an economic and environmentally friendly oxidant, was used, and a metal catalyst was unnecessary. Moreover, the continuous-flow technique was employed to increase the functional group tolerance, efficiency and safety.
In the recent past, many chemists have done tremendous amount of work about one-pot preparation of α-ketoamides. The one-pot synthesis of α-ketoamides from different starting materials, such as acetophenone,35–39 ethylbenzene,46 styrene,47,48 phenylacetylene,49 1-arylethanol,45 phenylglyoxal,40,50–52 phenylacetaldehyde53 and 2-hydroxy-1-phenyl-ethanon,43 was developed successfully as shown in Fig. 1. Among these starting materials, acetophenone and ethylbenzene were more economical and easy available. However, greater scope in terms of coupling partners was highly significant to the organic synthesis. After summarizing the literatures, we found that molecular oxygen (O2) and tert-butyl hydroperoxide (TBHP) were used as the terminal oxidant mostly. Molecular oxygen is an ideal oxidant. Nevertheless, the metal catalyst (CuI, CuBr) was necessary when molecular oxygen was served as the terminal oxidant, which was not environmental friendly and increased the possibility of metal residual. As for TBHP, it was a little costly compared with some other oxidants, such as hydrogen peroxide (H2O2) or sodium hypochlorite (NaClO). So, it was meaningful to build a synthesis methodology of one-pot preparation of α-ketoamides using economical oxidant without any metal catalyst.
Initially, optimization studies were based on the model substrates of DL-1-phenethylalcohol (1a) and 1-oxa-4-azacyclohexane (2a) to examine the feasibility of the oxidation system of H2O2/I−/H+ as shown in Table 1. Firstly, the mixing way of the reactants was screened. The reaction was carried out using a 25 mL round bottom flask charged with DL-1-phenethylalcohol (1 mmol), 1-oxa-4-azacyclohexane (3 mmol), potassium iodide (0.2 mmol), sulfuric acid (0.2 mmol), hydrogen peroxide (6 mmol), DMF (3 mL) and a magnetic stir bar. The reaction mixture was stirred at 70 °C. Unfortunately, only trace amount of 1a had been transformed into the corresponding ketone and α-ketoamide 3a after 24 h (Table 1, entry 1). Next, the hydrogen peroxide (a solution of H2O2 (6 mmol) in DMF (1.5 mL)) was added to the reaction mixture by syringe with the help of a syringe pump within 12 h instead of adding the oxidant (H2O2) in a whole. It was a pity that only a little (<10%) desired product was obtained (Table 1, entry 2). Afterwards, a two-step loading way was tested. A solution of DL-1-phenethylalcohol (1 mmol), potassium iodide (0.2 mmol), sulfuric acid (0.2 mmol) and hydrogen peroxide (2 mmol) in DMF (1.5 mL) was stirred at 70 °C. After 12 h, 1-oxa-4-azacyclohexane (3 mmol) was added in a whole. Then, the hydrogen peroxide (a solution of H2O2 (4 mmol) in DMF (1.5 mL)) was syringed addition within 2 h. Inspiringly, excellent yield of desired oxidative coupling product 3a was isolated (Table 1, entry 3). Considering the functional group tolerance and reaction time consumed, continuous-flow technique was employed, which was possessed of high surface-to-volume ratio, efficient mass transfer and heat transfer and high selectivity. Firstly, one-step synthesis of α-ketoamides from the oxidation of alcohol and direct coupling of α-carbonyl with functionalized amine was tested (Fig. 1). However, the desired product was detected with a very low conversion (Table 1, entry 4). Afterwards, a two-step of synthesis methodology was performed shown in Fig. 2. To our delight, an almost quantitative conversion was obtained (Table 1, entry 5). It seemed that the high efficiency of H2O2 in this oxidation coupling was inconsistent with many literatures, which could be explained by a phenomenon observed. After H2O2 was added into the reaction mixture only one time, the mixture was brownish red and much colourless gas was produced. The colour of the mixture changed into white quickly and no more gas observed, which indicated that H2O2 was decomposed. So, a high conversion and yield was obtained when H2O2 was added slowly with the help of syringe pump within a certain time. A series of reactions were carried out to examine the different catalysts, including KI, NH4I, TBAI, I2, NaBr, KBr and NaCl. According to the results shown in Table 1, this reaction could be promoted to give the expected products in good yields by all iodine salts (Table 1, entries 3, 5–8). When bromine salts were involved, the yields dropped to minute quantity (Table 1, entries 9 and 10). And, NaCl failed to promote this reaction (Table 1, entry 11). Subsequently, various solvents were examined. To our delight, excellent isolated yield (>90%) of 3a was obtained when the reaction was performed in DMF, ACN or dioxane (Table 1, entries 5, 12 and 14). However, DCE was less effective (Table 1, entry 13). When the reaction temperature was decreased from 70 °C to 40 °C, an obvious decrease in yield was observed (Table 1, entries 5, 15–17). Among different temperature examined, 60 °C turned out to be the proper choice. Finally, the optimizations of the amount of potassium iodide, sulfuric acid and hydrogen peroxide were carried out. To our delight, the full conversions and good yields were obtained even though the amount of potassium iodide and sulfuric acid were dropped to 5 mol% (Tables S1 and S2, ESI†). Moreover, less than 5.6 equiv. of hydrogen peroxide led to the incompletion of the reaction (Table S3, ESI†) (Fig. 3).
Entry | Catalyst | Solvent | Temp (°C) | Yieldb (%) |
---|---|---|---|---|
a The experimental details were listed in ESI, DMF: N,N-dimethylformamide, ACN: acetonitrile, DCE: 1,2-dichloroethane, TBAI: tetrabutylammonium iodide.b Isolated yield. | ||||
1a | KI | DMF | 70 | Trace |
2a | KI | DMF | 70 | <10 |
3a | KI | DMF | 70 | 95 |
4a | KI | DMF | 70 | Trace |
5a | KI | DMF | 70 | 95 |
6a | NH4I | DMF | 70 | 94 |
7a | TBAI | DMF | 70 | 96 |
8a | I2 | DMF | 70 | 96 |
9a | NaBr | DMF | 70 | <10 |
10a | KBr | DMF | 70 | <10 |
11a | NaCl | DMF | 70 | NO |
12a | KI | ACN | 70 | 92 |
13a | KI | DCE | 70 | 62 |
14a | KI | Dioxane | 70 | 95 |
15a | KI | DMF | 60 | 94 |
16a | KI | DMF | 50 | 80 |
17a | KI | DMF | 40 | 45 |
In the course of the investigation into the scope of different substituents, a series of substituted aryl methyl ketones and amines were explored (Table 2, 3). Fortunately, the direct coupling of 1-oxa-4-azacyclohexane with aryl ketones containing electron-withdrawing or electron-donating groups proceeded successfully to afford the desired products with moderate to excellent yields (Table 2, 1, 2, 4–9). However, an obvious lower yield (50%) was obtained when 1-(4-methoxyphenyl)ethanol containing an instable substituent group was involved (Table 2, entry 4). Remarkably, 4-(1-hydroxyethyl)phenol failed to afford the corresponding α-ketoamide, which was similar to the previous literatures (Table 2, entry 3). And, a little iodine byproduct in the aryl ring was detected, which consumed the catalyst and restrained the process of coupling. As for 3-nitroacetophenone, a strong electron-deficient aromatic ketone, there was not any obvious defect observed (Table 2, entry 5). It was noted that the hetero aryl methyl ketones could be transformed into the corresponding products in good yields (Table 2, entry 10–11).
Entry | 1-Arylethanol | Product | Yieldb (%) |
---|---|---|---|
a Store solution A: 0.667 M of alcohol in the DMF, flow rate 0.327 mL min−1; store solution B: 1.333 M of hydrogen peroxide, 5 mol% sulphuric acid (98%) and 5 mol% potassium iodide in DMF and H2O (29![]() ![]() |
|||
1 | ![]() |
![]() |
90(88)c |
2 | ![]() |
![]() |
89 |
3 | ![]() |
![]() |
NO |
4 | ![]() |
![]() |
50 |
5 | ![]() |
![]() |
90 |
6 | ![]() |
![]() |
92 |
7 | ![]() |
![]() |
91 |
8 | ![]() |
![]() |
88 |
9 | ![]() |
![]() |
90 |
10 | ![]() |
![]() |
91 |
11 | ![]() |
![]() |
87 |
Different amines were tested under the optimized conditions. The secondary amines could be transformed to afford the corresponding products with moderate to good yields (Table 3, entries 1–5). Furthermore, cyclic secondary amines showed higher activity compared with acyclic secondary amines. As for primary amines, the weak nucleophiles, only trace amount of products were detected, which was similar to many conventional method.
Entry | Amines | Product | Yieldb (%) |
---|---|---|---|
a Store solution A: 0.667 M of DL-1-phenethylalcohol in the DMF, flow rate 0.327 mL min−1; store solution B: 1.333 M of hydrogen peroxide, 5 mol% sulphuric acid (98%) and 5 mol% potassium iodide in DMF and H2O (29![]() ![]() |
|||
1 | ![]() |
![]() |
90 |
2 | ![]() |
![]() |
89(84)c |
3 | ![]() |
![]() |
86 |
4 | ![]() |
![]() |
89 |
5 | ![]() |
![]() |
73 |
6 | ![]() |
![]() |
NO |
7 | ![]() |
![]() |
NO |
8 | ![]() |
![]() |
NO |
After the establishment of the one-pot preparation of α-ketoamides, we believed that a cross dehydrogenative coupling (CDC) reaction between aryl propanols and functional amines could undergo successfully under the established reaction conditions before.
1-Phenyl-1-propanol and 1-oxa-4-azacyclohexane were chosen to test our conjecture as shown in Scheme 1.
To our delight, 5a, the desired product was generated smoothly with good yield (91%). Moreover, only a slight decrease of isolated yield in gram scale was observed with no further optimization (Table 4, entry 1). These structural elements of α-amino ketones are present in many drugs55–57 and photoinitiators.58 Moreover, they could be used as precursors to synthetize β-aminoalcohols59 and some heterocyclic compounds.60 Now, only a few literatures were reported to one-pot synthetize α-amino ketones from direct α-C–H amination of ketones. For example, MacMillan61 reported a O2/CuBr2-promoted direct coupling of α-carbonyls with functionalized amines. Meanwhile, Guo62 developed a oxidation system of NH4I/NaCO3·1.5H2O2 and Prabhu35 reported a NIS/TBHP system to do the similar work. There has been no report on the synthesis α-amino ketones using H2O2 as the oxidant. As far as we know, this is the first report of synthesis α-amino ketones from aryl propanols. Later, the amount of H2O2 was screened. 3.5 equiv. of H2O2 was proper, which was shown in Table S4 (ESI†). Next, the scope of aryl propanols and amines was investigated (Table 4).
Entry | Aryl propanols | Amines | Product | Yieldb (%) |
---|---|---|---|---|
a Store solution A: 0.667 M of alcohol in the DMF, flow rate 0.327 mL min−1; store solution B: 1.333 M of hydrogen peroxide, 5 mol% sulphuric acid (98%) and 5 mol% potassium iodide in DMF and H2O (29![]() ![]() |
||||
1 | ![]() |
![]() |
![]() |
91(86)c |
2 | ![]() |
![]() |
88 | |
3 | ![]() |
![]() |
NO | |
4 | ![]() |
![]() |
87 | |
5 | ![]() |
![]() |
88 | |
6 | ![]() |
![]() |
85 | |
7 | ![]() |
![]() |
84 | |
8 | ![]() |
![]() |
![]() |
89 |
9 | ![]() |
![]() |
90 | |
10 | ![]() |
![]() |
41 | |
11 | ![]() |
![]() |
NO | |
12 | ![]() |
![]() |
NO |
The reaction between 1-oxa-4-azacyclohexane and aryl propanols containing electron-withdrawing and electron-donating groups could be underwent smoothly to generate the desired products with good yields (Table 4, entries 1, 2, 4 and 5). Specially, there was no obvious decrease of yield observed when 4-(1-hydroxypropyl)benzotrifluoride, a strong electron-deficient aryl propanol, was involved (Table 4, entry 5). However, 4-(1-hydroxypropyl)phenol failed to afford the desired product, which was similar to many previous methods reported (Table 4, entry 3). Inspiringly, hetero aromatic propanols reacted with 1-oxa-4-azacyclohexane to yield the correspond products with satisfactory yields (Table 4, entries 6 and 7). As for the amines, both cyclic and acyclic secondary amines could be transformed to the α-amino ketones in moderate to good yields (Table 4, entries 1, 8–10). Interestingly, cyclic secondary amines showed higher reactivity than acyclic secondary amines. Unfortunately, almost no products were obtained when primary amines were employed (Table 4, entries 11 and 12). The optimization of coupling with primary amines is ongoing in our laboratory.
As for reaction mechanism, we believed that in was similar to Wan's report (TBHP/I2),38 which was shown in Scheme 2. Detailed discussion was summarized in ESI.† A control experiment was performed to validate this mechanism. When the temperature was decreased to 30 °C, the intermediate C could be obtained with 32% isolated yield. Moreover, intermediate E could be detected with the help of high resolution mass spectrometry. These supported this mechanism strongly.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra27653e |
This journal is © The Royal Society of Chemistry 2016 |