Zeyuan
Zhang
a,
Zhen
Dai
a,
Xinkun
Ma
a,
Yihan
Liu
a,
Xiaojun
Ma
a,
Wanli
Li
a and
Chen
Ma
*ab
aSchool of Chemistry and Chemical Engineering, Shandong University, Jinan, 250100, P R China
bState Key Laboratory of Natural and Biomimetic Drugs, Peking University, Beijing, 100191, P R China
First published on 19th April 2016
An efficient Cu-catalyzed cascade reaction protocol was developed for the synthesis of fused oxazepinone derivatives via sp2 C–H and O–H cross-dehydrogenative coupling. A readily available catalyst, Cu2O, was used in this modular and convergent approach. An unusual Smiles rearrangement reaction was involved in this synthesis. Various reaction parameters were evaluated and optimized, and the target compounds were obtained in good-to-excellent yields.
The functionalization of carbon–hydrogen bonds by bidentate auxiliary-directed cross-coupling approaches has become increasingly popular in recent decades.9–13 While these coupling approaches have been demonstrated to be effective in single-bond constructions, there is much interest in improving their efficiency by using one-pot procedures that can bring about multiple transformations in a cascade manner. Success in this area has come from a few impressive examples10d,e,j,12g,13a,c of exploitation of established multicomponent reactions as far as we know.
The Smiles rearrangement (SR) is an intramolecular nucleophilic aromatic substitution reaction which can be easily included in cascade reactions.14–18 Based on our previous work14 and experience with the Smiles rearrangement and construction of fused oxazepinone derivatives, we envisaged that a copper-catalyzed one-pot annulation between N-(quinolin-8-yl)benzamides 1 and 2-bromophenols 2 would allow access to dibenzoxazepinones 3 by a cross-dehydrogenative coupling (C–O construction)/Smiles rearrangement (SR, C–N construction)/O-arylation (C–O construction) domino approach (Scheme 2).
To confirm the feasibility of this approach, it was tested using N-(quinolin-8-yl)benzamide 1a and 2-bromophenol 2a as the model substrates, with Cu(OAc)2 as the catalyst, K2CO3 or Na2CO3 as the base, and DMF as the solvent. Surprisingly, we found that the SR product 4a was obtained when K2CO3 was used as the base, while Na2CO3 as the base gave a direct coupling product 3a (Scheme 3c). This result was contrary to our expectations, as the SR product is often obtained by a weaker base-mediated process, due to the slightly lower potential energy of the SR metastable intermediate.
In this work, we report the development of an efficient and selective copper-catalyzed domino synthesis of dibenzoxazepinones. This domino synthesis includes a C–N coupling reaction of secondary benzamides via an unusual Smiles rearrangement reaction.19 Several methods for the synthesis of fused oxazepinone scaffolds have been reported.14i,17a,20 In 2011, we developed a metal-free methodology for the synthesis of fused oxazepinone derivatives via the Smiles rearrangement by the reaction of N-substituted salicylamides with substituted benzenes or pyridines (Scheme 3a).14i In 2012, Sniechus reported a highly regioselective CuI-initiated domino reaction which provides a general route to dibenzoxazepinones from 2-iodobenzamides and 2-bromophenols using dbm as the ligand (Scheme 3b).17a This method provides an efficient, scalable, and modular approach for the rapid construction of dibenzoxazepinones. However, these methods are not ideal because they either involve multiple steps or use 2-functionalized benzamide as the substrate. To overcome these drawbacks, the synthesis reported here is a modular and convergent approach using commercially available or readily prepared starting materials. Moreover, this method involves a directed ortho-metalation strategy, which enables the rapid formation of fused oxazepinone derivatives.
The reaction conditions were optimized using the model reaction between N-(quinolin-8-yl)benzamide 1a and 2-bromo-4-chlorophenol 2a, and the results are summarized in Table 1. First, several bases were investigated under atmospheric conditions, and we found that only 3a was obtained when Na2CO3 was used (Table 1, entry 2). As the strength of the base increased, the proportion of product 4a increased but the total yield gradually decreased (Table 1, entries 1 and 3–5). To obtain the single compound 3a, different reaction conditions were optimized using Na2CO3 as the base. Then, several different copper catalysts such as CuBr, Cu(OH)2·CuCO3, Cu(OTf)2, CuO, CuI, and Cu2O were screened. The yield catalyzed by 20% Cu(OTf)2 (91%) was little higher than that realized by 10% Cu2O (90%), while no desired product was obtained when CuO was used (Table 1, entries 6–11 and 20). Given that Cu2O was much cheaper than Cu(OTf)2 (see the ESI†), Cu2O was selected for further experiments, and the effects of Cu2O loading were also investigated. It was found that addition of 10 mol% Cu2O resulted in 90% yield, while higher catalyst loadings did not increase the yield any further (Table 1, entries 6 and 11–15). Furthermore, several different reaction solvents were investigated, and there was no increase in the yield upon changing the solvent (Table 1, entries 16–19). The other relevant reaction conditions are summarized in the ESI.†
Entry | Catalyst (mol%) | Base | Solvent | Yieldb (%) | |
---|---|---|---|---|---|
3a | 4a | ||||
DMF = N,N-dimethylformamide, DMA = N,N-dimethylacetamide, NMP = N-methylpyrrolidone, DMSO = dimethyl sulfoxide.a Reaction conditions: 1a (0.2 mmol), 2a (0.3 mmol), catalyst (10 mol%), base (3.5 equiv.), solvent (2 mL), in a sealed tube at 150 °C under air for 6 h.b Yield of isolated products. | |||||
1 | Cu(OAc)2 (10) | K2CO3 | DMF | 35 | 22 |
2 | Cu(OAc)2 (10) | Na2CO3 | DMF | 77 | Trace |
3 | Cu(OAc)2 (10) | K3PO4·H2O | DMF | 25 | 30 |
4 | Cu(OAc)2 (10) | Cs2CO3 | DMF | 16 | 30 |
5 | Cu(OAc)2 (10) | NaOH | DMF | Trace | Trace |
6 | CuBr (10) | Na2CO3 | DMF | 71 | Trace |
7 | Cu(OH)2CuCO3 (10) | Na2CO3 | DMF | 69 | Trace |
8 | Cu(OTf)2 (10) | Na2CO3 | DMF | 85 | Trace |
9 | CuO (10) | Na2CO3 | DMF | N.D. | Trace |
10 | CuI (10) | Na2CO3 | DMF | 69 | Trace |
11 | Cu 2 O (10) | Na 2 CO 3 | DMF | 90 | Trace |
12 | Cu2O (5) | Na2CO3 | DMF | 72 | Trace |
13 | Cu2O (7) | Na2CO3 | DMF | 81 | Trace |
14 | Cu2O (12) | Na2CO3 | DMF | 88 | Trace |
15 | Cu2O (15) | Na2CO3 | DMF | 86 | Trace |
16 | Cu2O (10) | Na2CO3 | DMSO | 86 | Trace |
17 | Cu2O (10) | Na2CO3 | DMA | 82 | Trace |
18 | Cu2O (10) | Na2CO3 | NMP | 75 | Trace |
19 | Cu2O (10) | Na2CO3 | Toluene | 40 | Trace |
20 | Cu(OTf) 2 (20) | Na 2 CO 3 | DMF | 91 | Trace |
With the optimized reaction conditions, the substrate scope of this method was investigated. Firstly, a variety of 2-halogenophenols 2a–l were investigated in the reaction with N-(quinolin-8-yl)benzamide 1a (Table 2). The corresponding products were obtained in good-to-excellent yields, and the yields were found to improve in the presence of electron-withdrawing substituents R2 on the aryl ring. Low yields were obtained when electron-donating substituents R2 were present on the aryl ring. Notably, having Br and I as the 2-substituent in the 2-halogenophenols gave good yields, while no desired product was obtained when F and Cl were the 2-substituents.
a Reaction conditions: 1a (0.2 mmol), 2a (0.3 mmol), catalyst (10 mol%), base (3.5 equiv.), solvent (2 mL), in a sealed tube at 150 °C under air for 6 h. |
---|
Next, several different quinolinyl benzamide derivatives, N-(quinolin-8-yl)benzamides 1b–j, were evaluated in the reaction and the results are shown in Table 3. The target compounds were obtained in good yields. The presence of electron-withdrawing groups (–CF3, –F, and Cl) on the substrate slightly increased the yields of the target products 3k–3p and 3r–3v compared to the yields of the derivatives 3g and 3q with an election-donating group (–Me). The halide (fluoride and chloride) derivatives 3k–3m and 3r–3v were also obtained in good yields. With regard to the meta-substituted benzamides, cleavage of the C–H bond occurred para to the substituent to yield single regioisomers 3n and 3q.
a Reaction conditions: 1a (0.2 mmol), 2a (0.3 mmol), catalyst (10 mol%), base (3.5 equiv.), solvent (2 mL), in a sealed tube at 150 °C under air for 6 h. |
---|
To elucidate the mechanism of the reaction, several control experiments were performed (Scheme 4). The ortho-blocked benzamide 1l failed to generate the Goldberg amidation product (Scheme 4a), supporting the hypothesis that the cross-dehydrogenative coupling (CDC) should take place first in this reaction. To allow the reaction to remain in the first step, the amount of the base was reduced to 2.0 equiv. and the temperature was lowered to 75 °C. After the amide starting material was fully consumed, the mixture was cooled to room temperature. Then, 1.5 equiv. strong base t-BuOK was added to the mixture. Finally, the desired product 4a was obtained in 65% yield (Scheme 4b). It was further evidenced that the SR step requires higher energy to overcome the resistance from sterically hindered ligands.
Based on the above control experiments and our previous work on Smiles rearrangement chemistry, a possible mechanism for this reaction has been proposed, as shown in Scheme 5. Firstly, the reaction of N-(quinolin-8-yl)benzamide 1a and 2-bromo-4-chlorophenol 2a affords the intermediate 5. Then, a carboxamide anion 6 is formed. Subsequently, intermediate 6 can react via two pathways (a and b). Pathway a leads to 3a by direct intramolecular cyclization. In contrast, pathway b forms the intermediate 8via the Smiles rearrangement, which then undergoes another intramolecular cyclization, leading to the corresponding product 4a.
In summary, an efficient method has been developed for the synthesis of fused oxazepinone derivatives via sp2 C–H and O–H cross-dehydrogenative coupling, catalysed by the readily available Cu2O. This method is applicable for a wide variety of substrates, and the target compounds were obtained in good-to-excellent yields. Further studies on the application of this method towards the synthesis of pharmaceutical compounds are in progress.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 1449079 and 1449087. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6qo00040a |
This journal is © the Partner Organisations 2016 |