Xiao-Feng Xia*,
Su-Li Zhu,
Zhen Gu and
Haijun Wang
The Key Laboratory of Food Colloids and Biotechnology, Ministry of Education, School of Chemical and Material Engineering, Jiangnan University, Wuxi, Jiangsu 214122, China. E-mail: xiaxf@jiangnan.edu.cn; Fax: +86-510-85917763
First published on 18th March 2015
A copper-catalyzed 8-amide chelation-assisted C–H alkylation of 8-aminoquinolines is described. The secondary and tertiary alkyl carboxylic acids can be used as alkylating agents. This reaction allows for the highly regioselective preparation of C2-adamantyl 8-aminoquinoline scaffolds by the decarboxylative alkylation catalytic system.
Substituted 8-aminoquinolines are significant synthetic intermediates which are widely used in dye, medicine and material chemistry.4 Since the pioneering works by Daugulis,5 8-aminoquinolines as auxiliary chelating directing-groups have been used to assist C–H functionalization, which provides a powerful tool toward diverse molecule synthesis. Current researches in this field typically focus on the C–H functionalization on the carboxamide scaffolds.6 Few studies were reported on C–H transformations of the 8-aminoquinoline frameworks. At same time, duing to the difficulty in the control of the regioselectivity, the C–H functionalization of 8-aminoquinolines meet a significant challenge.7 Recently, the Zeng group successfully realized C5- or the C4-allylation of 8-aminoquinoline (Scheme 1-1).8 However, C2 functionalization of 8-aminoquinoline was rarely reported. Alkyl carboxylic acids can easily undergo decarbonylation to generate alkyl radicals which can be used in many synthetic transformations.9 Herein, we successfully realized a copper-catalyzed C2-alkylation of 8-aminoquinolines via 8-amide chelation assistance using alkyl carboxylic acids as alkylating agent (Schemes 1 and 2). In our opinion, firstly, the copper catalyst coordinates with the substrate 1 and 2 to form the chelation complex B. The amido group in complex B can adjust the distribution of electron density on the quinoline. Then the complex B was oxidized to give the radical-cation amidoquinoline C via an SET mechanism. Last, the carbon at the 2-position went through nucleophilic attack of alkyl radicals to give the C2-alkyl 8-aminoquinoline through an oxidative decarboxylation, deprotonation and aromatization processes (Scheme 2).
Our investigation began with the reaction of 1a with adamantanecarboxylic acid 2a using Cu(OAc)2·H2O as catalyst and Ag2CO3 as oxidant. To our delight, a decarboxylative coupling product 3a was obtained (Table 1, entry 1). With AgNO3 as an oxidant, a 20% yield was obtained (entry 2). When K2S2O8 was used as oxidant and AgNO3 as co-catalyst, a 40% product was separated. In order to further improve the yield, water was added as co-solvent according to previous report,9g and the yield can increase to 64% (entry 5). The loading of the catalysts were also screened (entries 6–8), and 30% AgNO3 gave a better result. When (NH4)2S2O8 was instead of K2S2O8, a lower yield was got. When 10% Pd(OAc)2 was employed as catalyst, no target product was achieved. The oxidant was proved to be crucial for the reaction because no product was obtained in the absence of K2S2O8. Meanwhile, when the catalyst was omitted, no reaction occurred (entry 12).
Entry | Catalyst (x% mmol) | Oxidant (y eq.) | Solvent | Yieldb |
---|---|---|---|---|
a Reaction conditions: 1a (0.30 mmol), 2a (2.0 equiv.), 80 °C, 12 h.b Isolated yields. | ||||
1 | Cu(OAc)2·H2O (20) | Ag2CO3 (2) | CH3CN | 15% |
2 | Cu(OAc)2·H2O (20) | AgNO3 (2) | CH3CN | 20% |
3 | Cu(OAc)2·H2O (20) + AgNO3 (30) | K2S2O8 (2) | CH3CN | 40% |
4 | Cu(OAc)2·H2O (20) + AgNO3 (30) | K2S2O8 (2) | CH3CN/H2O (9![]() ![]() |
50% |
5 | Cu(OAc)2·H2O (20) + AgNO3 (30) | K2S2O8 (2) | CH3CN/H2O (1![]() ![]() |
64% |
6 | Cu(OAc)2·H2O (10) + AgNO3 (30) | K2S2O8 (2) | CH3CN/H2O (1![]() ![]() |
65% |
7 | Cu(OAc)2·H2O (10) + AgNO3 (20) | K2S2O8 (2) | CH3CN/H2O (1![]() ![]() |
60% |
8 | Cu(OAc)2·H2O (10) + AgNO3 (10) | K2S2O8 (2) | CH3CN/H2O (1![]() ![]() |
58% |
9 | Cu(OAc)2·H2O (10) + AgNO3 (30) | (NH4)2S2O8 (2) | CH3CN/H2O (1![]() ![]() |
55% |
10 | Pd(OAc)2 (10) + AgNO3 (30) | K2S2O8 (2) | CH3CN/H2O (1![]() ![]() |
<5% |
11 | Cu(OAc)2·H2O (10) + AgNO3 (30) | — | CH3CN/H2O (1![]() ![]() |
0 |
12 | — | K2S2O8 (2) | CH3CN/H2O (1![]() ![]() |
0 |
Under the optimized reaction conditions (Table 1, entry 6), firstly, the influence of 8-amide scaffolds on the C–H alkylation of quinolines was probed (Scheme 3). Benzoyl protected 8-aminoquinoline can deliver a 65% yield of the product. An alkyl-substituted acyl-quinoline gave a lower yield (3b). When 8-aminoquinoline without any protection was subjected to the standard reaction conditions, no product was detected. According to these phenomenons, the acyl protecting group was significant to the reaction, which is in accordance with the proposed mechanism. Then, a series of functional groups were screened in different positions of benzoyl groups in this reaction, including chloro, bromo, fluoro, trifluoromethyl and ether substituents. An electron-withdrawing substituent favored product formation (3f, 3h, 3j, 3k), whereas an electron-donating group slightly hindered the reaction (3d, 3e). Trifluoromethyl substituted substrate gave the best result (3h). When a sterically demanding ortho substituent was used, a low yield was obtained (3i). To confirm further the structural assignment of products in the present decarboxylative alkylation, the structure of the product 3f was unambiguously assigned by X-ray crystallography (see Fig. 1).10
![]() | ||
Scheme 3 Reaction conditions: 1 (0.3 mmol), 2a (0.6 mmol), Cu(OAc)2·H2O (10% mmol), AgNO3 (30% mmol), K2S2O8 (0.6 mmol), CH3CN/H2O (1![]() ![]() |
Next, we tune our attention on the alkyl carboxylic acids, the adamantanecarboxylic acids with carbonyl and hydroxyl can be well-tolerated to give the corresponding products in moderate yields (Scheme 4, 3m and 3n). Substituted methoxy 8-aminoquinoline can deliver the products 3o and 3p in moderate yields. 3-Noradamantanecarboxylic acid can be also used in the decarboxylative coupling reaction to give the product 3q in 35% yield. Substituted pyridine-2-ylmethanamine can participate in the reaction to give the alkylation product 3r in moderate yield. To our disappointment, other non-cyclic tertiary acids such as pivalic acid and 3-hydroxy-2,2-dimethylpropanoic acid did not give any products in the reaction conditions, and the starting materials can be quantitatively recovered. When secondary aliphatic carboxylic acids such as cyclohexanecarboxylic acid and cyclopentanecarboxylic acid were subjected to the standard conditions, mixed isomers were obtained in moderate to good yields (3s, r.r. = 3:
5; 3t, r.r. = 5
:
3). 2,3-Dihydrobenzo[b][1,4]dioxine-2-carboxylic acid was also tolerated in this reaction to give a mixed isomer 3u (78%, r.r. = 5
:
3). Unfortunately, when primary aliphatic carboxylic acid such as 2-(p-tolyloxy)-acetic acid was employed, no reaction occurred duing to the poor reactivity.
To gain insight into the mechanism, 2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO) as a radical quencher was added into the reaction system, and the reaction was completely inhibited, suggesting that the reaction involved a radical pathway, which is in accordance with classical Minisci reaction (Scheme 5).
The benzoyl protecting group can be easily removed by base hydrolysis. For example, treatment of 3h with 15 equivalent of NaOH in EtOH at 80 °C for 8 h affords 2-adamantyl-8-aminoquinoline 4 in moderate yield (64%), which meant that the synthetic method was useable (Scheme 6).11
Footnote |
† Electronic supplementary information (ESI) available. CCDC 1029795. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c4ra16896h |
This journal is © The Royal Society of Chemistry 2015 |