Enantioselective desymmetrization of meso epoxides with anilines catalyzed by a niobium complex of a chiral multidentate binol derivative
Enantioselective desymmetrization of meso epoxides with anilines catalyzed by a niobium complex of a chiral multidentate binol derivative
复制标题
DOI：
10.1002/anie.200603787
复制
发表时间：
2007-01-01
影响因子：
16.6
通讯作者：
Kobayashi, Shu
中科院分区：
文献类型：
--
作者：
Arai, Kenzo;Salter, Matthew M.;Kobayashi, Shu
开发新型手性催化剂促进不对称反应是有机合成领域的重要课题之一。新的手性催化剂不仅必须以高产率和对映体选择性介导反应，而且越来越多地要求具有识别底物精确结构的能力。例如，可以识别底物中甲基和乙基之间差异的非酶催化剂将是该领域的理想催化剂。[1]我们现在描述的二元醇衍生物和铌的纯手性多齿配合物，严格区分底物结构和催化的desymmetrizaton的内消旋环氧化物与苯胺有效。我们以前报道了第一个基于铌（V）的高对映选择性刘易斯酸用于亚胺与烯醇化硅的Mannich型反应的例子。[2]该物种，从金属源和二醇衍生的三齿配体制备，被证明具有独特的双核结构，其中两个铌原子由两个分子的配体跨越。这种安排，其中的金属中心被固定在一个空间定义的阵列，但灵活的连接，似乎允许高度底物选择性反应。由于铌（V）对氧具有高亲和力，并且已经在我们的金属配体系统中大量配位，我们判断单齿物种将最有效地作为底物发挥作用，并因此专注于内消旋环氧化物[3]与苯胺的不对称开环。[4]一系列的初步实验表明，当络合物中的铌原子是四配位时，反应进行得最有效。这一发现导致了一类新的刘易斯酸配合物的发展，形成从铌和四齿配体1。进一步的研究表明，最有效的催化剂体系是Nb（OMe）5-1，在4-分子筛的存在下，以甲苯和CH 2Cl 2的3：2混合物作为溶剂。在优化了反应条件后，我们将注意力转向了反应的范围。我们用一系列不同的环氧化物和苯胺进行了反应，[5]得到了惊人的结果（表1）。首先，发现顺式-丁-2-烯氧化物（2a）与苯胺3a的反应非常顺利地进行，以非常高的对映选择性以定量产率得到相应的1，2-羟胺4aa（表1，条目1）。随后的实验，其中在10摩尔%催化剂存在下用一系列取代苯胺处理顺式-丁-2-烯氧化物（2a），进一步证实了该体系的实用性（表1，条目2-9）。这些结果表明，催化剂的活性是一般的广泛的苯胺亲核试剂，包括富电子和贫电子的例子，虽然在邻位取代的苯胺的情况下，反应的化学产率略低。此外，从合成的角度来看，我们可以将催化剂的量从10mol%降低到0.25mol%，而没有显着的活性损失。另一方面，当用密切相关的环氧化物底物顺式-己-3-烯氧化物（2b）和顺式-辛-4-烯氧化物（2c）进行反应时，几乎没有任何周转，并且仅形成痕量的产物（表1，条目10和11）。然而，当我们转换为衍生自环烯烃的环氧化物底物时，开环产物以良好至非常高的产率和通常高的对映选择性获得（表1，条目12-16）。据我们所知，这种高水平的分子识别和选择性在内消旋环氧化物的去对称化中是前所未有的，并且是...
The development of new chiral catalysts for the promotion of asymmetric reactions is one of the most important tasks in organic synthesis. New chiral catalysts must not only mediate reactions with high yields and enantioselectivities, but are also increasingly required to have the ability to recognize the precise structure of substrates. For example, a nonenzymatic catalyst that could recognize the difference between methyl and ethyl groups in a substrate would be an ideal catalyst in this field.[1] We now describe homochiral multidentate complexes of binol derivatives and niobium which distinguish substrate structures strictly and catalyze the desymmetrizaton of meso epoxides with anilines efficiently. We previously reported the first example of a highly enantioselective Lewis acid based on niobium (V) for Mannich-type reactions of imines with silicon enolates.[2] This species, prepared from a metal source and a binol-derived tridentate ligand, was shown to have a unique binuclear structure in which two niobium atoms were straddled by two molecules of the ligand. This arrangement, in which the metal centers are held in a spatially defined array by firm but flexible ligation, appeared to permit highly substrate-selective reactions. As niobium (V) has a high affinity for oxygen and is already heavily coordinated in our metal–ligand system, we judged that monodentate species would function most efficiently as substrates and focused accordingly on the asymmetric ring opening of meso epoxides [3] with anilines.[4] A series of initial experiments showed that the reaction proceeded most efficiently when the niobium atom in the complex was tetracoordinated. This finding led to the development of a new class of Lewis acid complexes formed from niobium and the tetradentate ligand 1. Further investigations revealed that the most effective catalyst system was Nb (OMe) 5–1 in the presence of 4-molecular sieves with a 3: 2 mixture of toluene and CH2Cl2 as the solvent. Following the optimization of the reaction conditions, attention was turned to the scope of the reaction.We conducted the reaction with a range of different epoxides and anilines,[5] with striking results (Table1). First, it was found that the reaction of cis-but-2-ene oxide (2a) with aniline 3a proceeded very smoothly to give the corresponding 1, 2-hydroxyamine 4 aa in quantitative yield with very high enantioselectivity (Table 1, entry 1). Subsequent experiments, in which cis-but-2-ene oxide (2a) was treated with a range of substituted anilines in the presence of 10 mol% of the catalyst, further confirmed the utility of the system (Table 1, entries 2–9). These results showed that the catalyst activity was general for a broad range of aniline nucleophiles, including both electron-rich and electron-poor examples, although in the case of ortho-substituted anilines the chemical yield of the reaction was slightly lower. Furthermore, from a synthetic point of view, we could decrease the amount of the catalyst from 10mol% to 0.25 mol% without significant loss of activity. On the other hand, when the reaction was conducted with the closely related epoxide substrates cis-hex-3-ene oxide (2b) and cis-oct-4-ene oxide (2c), there was hardly any turnover, and only traces of the products were formed (Table 1, entries 10 and 11). However, when we switched to epoxide substrates derived from cyclic alkenes, the ringopened products were obtained in good to very high yields and with generally high enantioselectivities (Table 1, entries 12–16). Such outstandingly high levels of molecular recognition and selectivity are, to the best of our knowledge, unprecedented in the desymmetrization of meso epoxides and are a unique characteristic of …