Synthesis of Aziridines by Palladium-Catalyzed Reactions of Allylamines with Aryl and Alkenyl Halides: Evidence of a syn-Carboamination Pathway

Synthesis of Aziridines by Palladium-Catalyzed Reactions of Allylamines with Aryl and Alkenyl Halides: Evidence of a syn-Carboamination Pathway
复制标题

DOI:
10.1002/anie.200903178
复制
发表时间:
2009-01-01
影响因子:
16.6
通讯作者:
Oshima, Koichiro
Oshima, Koichiro
中科院分区:
化学1区
文献类型:
--
作者:
Hayashi, Sayuri;Yorimitsu, Hideki;Oshima, Koichiro

文献摘要

被引文献

相似文献

钯催化的烯烃与有机卤化物的分子内碳醚化或碳胺化反应是一种有吸引力的构建杂环的方法,在单一操作中形成碳-杂原子键和碳-碳键。[1]通过这种方法已经合成了许多五元杂环[2,3],然而,应变三元杂环的制备仍然是一个挑战。我们以前曾报道过钯催化的叔烯丙醇与芳基或烯基卤化物的碳醚化反应,得到多取代的环氧化物。[4]我们期望该反应可以扩展到碳胺化反应,以烯丙胺为原料合成氮杂环丙烷。在这里,我们提出了我们的碳胺化沿着与一个合理的反应机理的证据的初步结果。我们的研究开始于N-苯基烯丙基胺1a在烯丙基位置带有两个苯基的反应(表1)。在钯催化下,在叔丁醇钠的存在下用溴苯处理1a导致氮杂环丙烷化和C12 C键的形成,以98%的产率提供相应的芳基化氮杂环丙烷2a(表1,条目1)。与我们以前的环氧化反应相比,氮丙啶化反应没有受到竞争性Mizoroki-Heck反应的影响。[4]广泛的芳基溴化物和氯化物以优异的产率掺入相应的氮杂环丙烷2a-2 h中(表1,条目2 -9)。烯基氯也被证明是用于氮丙啶化反应的合适底物(表1,条目10和11)。然后检查几种N-芳基烯丙基胺与溴苯的反应(表2)。氮原子上芳基取代基的电子性质不影响反应效率(表2,条目1和2)。令人满意的是,烷基取代的烯丙基胺1d和1 e也以令人满意的产率进行了反应(表2,条目3和4)。结果发现,较大的取代基,RL,和苄基部分取向的顺式构型中的主要非对映异构体。[5]有趣的是,在氮丙啶化反应中获得的主要非对映异构体具有与环氧化物相反的构型。[4]例如,与If的反应以90%的产率得到作为单一非对映异构体的4a(表3,条目1)。此外,带有三氟甲基的烯丙胺Ig也以高非对映选择性参与反应(表3,条目2)。然而,产率和非对映体均
Palladium-catalyzed intramolecular carboetherification or carboamination reactions of alkenes with organic halides emerged as an attractive method to construct heterocycles, forming both carbon–heteroatom and carbon–carbon bonds in a single operation.[1] A number of five-membered heterocycles have been synthesized by this methodology,[2, 3] however, the preparation of strained three-membered heterocycles has remained a challenge. We have previously reported palladium-catalyzed carboetherification reactions of tertiary allyl alcohols with aryl or alkenyl halides, which provide multisubstituted epoxides.[4] We expected that the reaction could be extended to carboamination for the synthesis of aziridines starting from allylamines. Herein we present our preliminary results of the carboamination along with evidence for a plausible reaction mechanism. Our investigation began with a reaction of N-phenylallylamine 1a bearing two phenyl groups at the allylic position (Table1). Treatment of 1a with bromobenzene in the presence of sodium tert-butoxide under palladium catalysis led to aziridination and CÀC bond formation, providing the corresponding arylated aziridine 2a in 98% yield (Table 1, entry 1). In contrast to our previous epoxidation reactions, the aziridination reaction did not suffer from a competitive Mizoroki–Heck reaction.[4] A wide range of aryl bromides and chlorides were incorporated into the corresponding aziridines 2a–2h in excellent yields (Table1, entries2–9). Alkenyl chlorides also proved to be suitable substrates for the aziridination reactions (Table 1, entries 10 and 11). The reactions of several N-arylallylamines with bromobenzene were then examined (Table 2). The electronic nature of aryl substituents on the nitrogen atom did not affect the efficiency of the reaction (Table 2, entries 1 and 2). Gratifyingly, alkyl-substituted allylamines 1d and 1e also underwent the reaction in satisfactory yields (Table 2, entries 3 and 4).Next, we turned our attention to reactions of allylamines bearing a stereogenic center at the aminated carbon atom, in which two diastereomers could be obtained (Table 3). It was found that the larger substituent, RL, and the benzyl moiety were oriented in a cis configuration in the major diastereomer.[5] Interestingly, the major diastereomer obtained in the aziridination reaction possessed a configuration opposite to that of the epoxides.[4] For instance, the reaction with 1 f afforded 4a as a single diastereomer in 90% yield (Table 3, entry 1). Moreover, allylamine 1g bearing a trifluoromethyl group also took part in the reaction with high diastereoselectivity (Table 3, entry 2). However, both yield and diastereo-