Vinyl azides are a class of unique functionalized alkenes with high intrinsic reactivity.[1 ] The numerous transformations of vinyl azides provide reliable synthetic approaches to diverse, structurally distinct molecular frameworks, including hetero/carbocycles,[2 ] amides,[3 ] ketones,[4 ] and 2H -azirines.[5 ] Recently, Wu and Liu[6 ] described an efficient and mild method for the synthesis of various α-fluoroketones from the corresponding vinyl azides by using Selectfluor® as the fluorine source (Scheme [1a ]). Additionally Chiba[7 ] and Liu,[4a ] respectively, disclosed radical trifluoromethylation of vinyl azides with Me3 SiCF3 , to allow construction of α-trifluoromethyl azines, which could be further converted into α-trifluoromethyl ketones (Scheme [1b ]). Bi and co-workers presented a radical-induced enamination of vinyl azides that proceeded with electron-withdrawing substrates, and a variety of β-functionalized primary enamines, including β-nitro, acyl, and sulfonyl derivatives, could be prepared, which were successfully transformed into a series of α-functionalized ketones (Scheme [1c ]).[8 ] However, with respect to the construction of α-functionalized ketones, these methods usually suffer from disadvantages such as tedious multistep operations or harsh conditions. In a continuation of our efforts on the construction of C–N bond reactions,[9 ] we herein report a novel radical amination of vinyl azides using a copper catalyst, to afford a series of α-ketoamides (Scheme [1 ]). To our knowledge, this is the first example of the conversion of vinyl azides into α-ketoamides, which are important units in biologically active molecules, synthetic drugs, and drug candidates.[10 ]
Scheme 1 Transformations of vinyl azides
For the optimization of the reaction conditions, we carried out the reaction of vinyl azide 1a and N ,N -dimethylformamide (DMF; 2a ), as model substrates for the amination reaction, to screen different catalysts, solvents and additives (Table [1 ]). Firstly, when CuI was used as the catalyst in DMF as the solvent at 100 °C for 5 h under an oxygen atmosphere, the corresponding α-ketoamide 3a was obtained in 67% yield (entry 1). Using CuBr as the catalyst also gave a comparable yield under similar conditions (entry 2). Other metal catalysts, such as CuCl2 , NiCl2 , or Ag2 CO3 , gave either low yields or trace amounts of the desired product 3a (entries 3–5). Subsequent solvent screening demonstrated that DMF was the best choice; solvents, such as toluene, ethylene glycol, and nitromethane resulted in no desired product (entries 6–8). The use of additives, such as m -CPBA, TBHP or BPO, did not improve the efficiency of the reaction (entries 9–11). However, the addition of acids (HBF4 or benzoic acid) did improve the efficiency of the reaction, affording 3a in 70% and 88% yields, respectively (entries 12 and 13). The reaction failed to occur in the absence of either the catalyst or additive (entries 14 and 15).
Table 1 Optimization of the Reaction Conditionsa
Entry
Catalyst
Solvent
Additive
Yield (%)b
1
CuI
DMF
t -BuOCl
67
2
CuBr
DMF
t -BuOCl
56
3
CuCl2
DMF
t -BuOCl
trace
4
NiCl2
DMF
t -BuOCl
19
5
Ag2 CO3
DMF
t -BuOCl
20
6
CuI
toluene
t -BuOCl
0
7
CuI
EG
t -BuOCl
0
8
CuI
MeNO2
t -BuOCl
0
9
CuI
DMF
m -CPBA
26
10
CuI
DMF
TBHP
53
11
CuI
DMF
BPO
38
12
CuI
DMF
t -BuOCl + HBF4
c
70
13
CuI
DMF
t -BuOCl + PhCO2 Hc
88
14
–
DMF
t -BuOCl + PhCO2 Hc
0
15
CuI
DMF
–
0
a Reaction conditions: 1a (0.5 mmol), 2a (0.75 mmol), catalyst (0.15 mmol), solvent (1.0 mL), additive (1.0 mmol), O2 atmosphere, 100 °C, 5 h.
b Isolated yield.
c Acid (1.0 mmol).
With the optimized reaction conditions in hand (Table [1 ], entry 13), the generality of this α-ketoamide synthesis was examined by using a series of vinyl azides to react with DMF (Scheme [2 ]). α-Aryl-substituted vinyl azides were readily transformed into the corresponding α-ketoamides 3a –n in moderate to excellent yields. Electron-donating groups in the para -position of the aromatic ring were well tolerated, leading to the corresponding products 3a –e in good to excellent yields. Likewise, vinyl azides possessing aryl substituents bearing electron-withdrawing groups, such as -F, -Cl, -Br, -CHO, and -CO2 Me, afforded the desired α-ketoamides 3f –j in 79–90% yield. Note that vinyl azides bearing heteroaromatic substituents such as thienyl could be smoothly incorporated onto α-ketoamide 3k in 88% yield. Moreover, vinyl azide derivatives bearing a variety of substitutions in the meta -position of the aryl ring, including both electron-donating and electron-withdrawing groups, were well tolerated; thereby affording the functionalized α-ketoamides 3l –n in 69−87% yields.
Scheme 2 Substrate scope for the reaction of vinyl azides with DMF (used as both solvent and reactant)
To investigate the applicability of this protocol further, N ,N -diethyl-α-ketoamide products 4a –f were prepared using vinyl azides 1 and N ,N -diethylformamide (2b ) as the substrates under the optimized conditions (Scheme [3 ]). Vinyl azides 1a –f with aromatic substituents, including aryl and heteroaryl groups, underwent the reaction smoothly to give the desired products 4a –f in 77−93% yield.
Scheme 3 Substrate scope for the reaction of vinyl azides with N ,N -diethylformamide. Reaction conditions: 1 (0.5 mmol), 2 (1.0 mL), CuI (0.15 mmol), PhCO2 H (1.0 mmol), t -BuOCl (1.0 mmol), O2 atmosphere, 100 °C, 5 h.
Encouraged by these results, we next turned our attention to other representative nitrogen sources (Table [2 ]). Indeed, not only DMF (2a ) and N ,N -diethylformamide (2b ) but also N ,N -dimethylacetamide (DMA) and N ,N -dimethylpropylamide (DMP) were tolerated under the standard conditions, with the resultant α-ketoamide products being obtained in slightly lower yields (entries 1–4). In addition, the long chain N ,N -dimethylbutylamide (DMB) was also examined, although no desired product was detected (entry 5).
Table 2 Substrate Scope for the Reaction of Vinyl Azides with N ,N -Dimethyl Substratesa
Entry
Substrate
Product
Yield (%)b
1
3a
51
2
3b
44
3
3g
55
4
DMPA
3a
20
5
DMB
3a
0
a Reaction conditions: 1 (0.5 mmol), 2 (1.0 mL), CuI (0.15 mmol), PhCO2 H (1.0 mmol), t -BuOCl (1.0 mmol), O2 atmosphere, 100 °C, 5 h.
b Isolated yield.
To elucidate a plausible reaction mechanism for this transformation, some potential intermediates were prepared and tested under the standard conditions (Table [1 ], entry 13). Vinyl azide 1a′ did not afford benzamide 3a under the standard conditions (Scheme [4a ]), although 3-phenyl-2H -azirine (1b′ ), N -(1-phenylvinyl)acetamide (1c′ ), and 2-chloro-1-phenylethan-1-one (1d′ ) all afforded 3a in good yields (Scheme [4b–d ]). 2-(Diethylamino)-1-phenylethan-1-one (1e′ ) did not react under the standard conditions (Scheme [4e ]),[11 ] nor did dimethylamine (2a′ ; Scheme [4f ]). Interestingly, the addition of the radical scavengers 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) or 2,6-di-tert -butyl-p -cresol (BHT) suppressed the reaction, which indicates that free radical intermediates could be involved in this transformation (Scheme [4g ]). Notably, only 20% yield of 3a was observed under an argon atmosphere, suggesting that oxygen is required to oxidize the catalyst in preparation for another cycle (Scheme [4g ]). These control experiments indicate that the vinyl azide substrate possibly reacts with the t -BuOCl as the electrophile and then undergoes oxidation catalyzed by the Cu/O2 system.
Scheme 4 Control experiments
According to previous reports and considering the above results,[2e ]
[12 ] a possible mechanism for this amination reaction is proposed in Scheme [5 ]. Initially, the vinyl azide 1a undergoes thermal decomposition into a highly strained three-membered cyclic imine (2H -azirine A ) that reacts with the Cu(I) catalyst and t -BuOCl to form Cu(II) imine B , which is then converted into Cu(II) imine C .[2e ]
[13 ] Radical intermediate E would be formed via Cu(III) peroxide radical D , which could be generated by a combination of molecular oxygen with Cu(II) imine C . Subsequent reductive elimination of radical intermediate E gives intermediate F and regenerates the active Cu(I) species. Finally, α-ketoamide 3a is obtained through acid hydrolysis of intermediate F .
Scheme 5 Proposed mechanism
The potential synthetic applicability of this method was investigated on a gram scale by using the model reaction. As shown in Scheme [6, 1 ].54 g of α-ketoamide 3a was isolated in 87% yield without any significant loss of efficiency, demonstrating the potential of this methodology for large-scale synthesis of α-ketoamide derivatives.
Scheme 6 Gram-scale experiment
In conclusion, we have developed a general and efficient method for the synthesis of α-ketoamides from vinyl azides with N ,N -dialkylacylamide as the nitrogen source.[14 ] The key to success is not only the introduction of a Cu/O2 catalytic system but also the use of t -BuOCl and benzoic acid as additives. This method features readily accessible substrates, commercially available and inexpensive reagents, and mild conditions. This mild catalytic reaction demonstrates a broad substrate scope and high functional group tolerance. A possible mechanism involving copper-catalyzed oxidative generation of peroxide radicals is proposed. The reaction can be effectively scaled up and the product conveniently obtained in a one-pot process. Efforts to further clarify the mechanism and to expand the application of vinyl azides are under way in our laboratory.
