Stable cyclic silenes from reaction of disilenides with carboxylic acid chlorides
Stable cyclic silenes from reaction of disilenides with carboxylic acid chlorides
复制标题
DOI：
10.1002/anie.200700067
复制
发表时间：
2007-01-01
影响因子：
16.6
通讯作者：
Scheschkewitz, David
中科院分区：
文献类型：
--
作者：
Bejan, Iulia;Gueclue, Denis;Scheschkewitz, David
自从40年前由Gusel Nikow和Flowers首次证明是中间体以来，[1]Silenes，即具有Si=C键的化合物，一直吸引着人们的兴趣。[2]Brook等人报道了稳定的Silene 1(方案1)。[3]它出人意料的长的Si=C键引起了激烈的讨论，并通过计算得到了解决。[4]创造了“反极性”一词：1中的π-给体硅氧基显著降低了Si=C键的极性，从而增强了它的稳定性，增加了键长。此后不久，Wiberg等人。报道了Silene 2，它没有π供体取代基，因此显示出预期的较短的Si=C键。[5]其他可分离的无给体Silenes随后被Apeloig和Kira小组报道。[6]Ottosson等人最近的研究。对硅烯酸酯3和其他具有比硅氧基更强的给π能力的取代基的硅烯的聚焦表明，随着硅中心的锥化，Si=C的键级降低。[7]通过适当的酰基硅烷的热和光化学重排，很容易从酸性氯化物和硅基阴离子反应中获得第一类硅烯。[8]在过去的几年里，通过我们两个基团的努力，一些二硅烯，即乙烯基阴离子的二硅烷类似物出现。[9]鉴于4a作为一种不饱和的亲核试剂的许多成功应用，[10]我们尝试了4 a，b与酸性氯化物反应制备酰基二硅烯(方案2)。然而，在4a与1-金刚烷基和甲氧丙酰氯的反应中，即使在193K也没有检测到预期的(和可能的中间体)酰基二硅烯5a，b。这种不稳定性与碳类似物α，β不饱和酮形成鲜明对比。相反，根据核磁共振和UV/VIS光谱数据(图式2，见支持信息)，定量地形成了四元环状硅烯6a，b。因此，4b很容易与1-金刚烷基氯反应生成环状硅烯衍生物6c。~(29)Si核磁共振谱显示，6a和6b的三配位和四配位硅核的化学位移几乎相同。奇怪的是，在6a的情况下，高场(d=17.5ppm)的信号被分配给双键的Si原子。与相应的信号1(d=41.4ppm)相比，它被相当大的屏蔽。[3C，12]环碳原子表现出与1(d=214.2 ppm[3c])非常相似的化学位移(6a：d=213.4 ppm，6b：d=214.6 ppm)。6b中双键的Si-C耦合明显小于1(6b：1j(C，Si)=67.3 Hz)，1：1j(C，Si)=84.4 Hz[3c]。这些观察结果可以通过环应变引起的内环键p特征的增加而得到合理的解释。因此，在硅上承载剩余电子密度的轨道应该会变得更高
Since first evidenced as intermediates by Gusel nikow and Flowers 40 years ago,[1] silenes, that is, compounds with Si= C bonds, have attracted continuous interest.[2] The stable silene 1 (Scheme 1) reported by Brook et al. gave further impetus to the field.[3] Its unexpectedly long Si= C bond prompted an intense discussion, which was resolved by calculations.[4] The term “reversed polarity” was coined: the π-donating siloxy group in 1 reduces the polarity of the Si= C bond significantly, thereby enhancing its stability and increasing the bond length. Shortly thereafter Wiberg et al. reported silene 2, which is free of π-donating substituents and accordingly exhibited the expected shorter Si= C bond.[5] Other isolable donor-free silenes were subsequently reported by the groups of Apeloig and Kira.[6] Recent studies by Ottosson et al. focusing on silenolate 3 and other silenes bearing substituents with stronger π-donating ability than that of the siloxy group in 1 revealed a decreasing bond order for Si= C accompanied by increased pyramidalization of the silicon center.[7] Silenes of type 1 had been prepared by thermal and photochemical rearrangement of suitable acyl silanes, easily accessible from reactions of acid chlorides with silyl anions.[8] In the last years a few disilenides, that is, disila analogues of vinyl anions became available by efforts of both of our groups.[9] In light of a number of successful applications of 4a as an unsaturated, nucleophilic reagent,[10] we attempted the preparation of acyl disilenes by reaction of 4 a, b (Scheme 2) with acid chlorides.In reactions of 4a with 1-adamantoyl and pivaloyl chloride, however, the expected (and likely intermediate) acyl disilenes 5a, b were not detected even at 193 K. This instability is in marked contrast to the carbon analogues, α, βunsaturated ketones. Instead, the four-membered cyclic silenes 6a, b were formed quantitatively as indicated by NMR and UV/Vis spectroscopic data (Scheme2, see the Supporting information).[11] The reaction of disilenides with acid chlorides turns out to be quite general. Thus, 4b readily reacts with 1-adamantoyl chloride to yield the cyclic silene derivative 6c. 29Si NMR spectroscopy reveals almost identical chemical shifts for the tri-and the tetracoordinated silicon nuclei of both 6a and 6b. Curiously, in the case of 6a the signal at higher field (d= 17.5 ppm) is assigned to the Si atom of the double bond. It is considerably shielded compared to the corresponding signal of 1 (d= 41.4 ppm).[3c, 12] The ring carbon atoms exhibit chemical shifts (6a: d= 213.4 ppm, 6b: d= 214.6 ppm) that are very similar to that found for 1 (d= 214.2 ppm [3c]). The Si–C coupling of the double bond in 6b is distinctly smaller than that in 1 (6b: 1J (C, Si)= 67.3 Hz), 1: 1J (C, Si)= 84.4 Hz [3c]). These observations can be rationalized by the increase in p character of the endocyclic bonds due to ring strain. As a consequence the orbitals hosting the remaining electron density at silicon should get higher