An MO-based identification of charge-shift bonds.
An MO-based identification of charge-shift bonds.
复制标题
DOI：
10.1002/cphc.201200147
复制
发表时间：
2012-06
期刊：
影响因子：
--
通讯作者：
Renana Gershoni‐Poranne;A. Stanger
中科院分区：
文献类型：
--
作者：
Renana Gershoni‐Poranne;A. Stanger
化学键是化学中的基本概念之一。在超过世纪的时间里，化学家们认识到两种类型的分子内（即同一分子中的原子之间，而不是分子之间）化学键[1]-共价键和离子键。最近，Shaik，Hiberty和他们的合作者[2]介绍了一种新型的化学键-电荷位移（CS）键。这种类型的键可以在许多化合物中找到，正如许多论文所证明的那样。[3，4]当键合电子由于核心电子（例如F2）的排斥而具有高动能时发生。它也发生在异常键合的情况下（如螺桨烷）。在价键（VB）术语中，这种类型的键表现为给定键的共价和离子正离子形式之间的异常大的共振能。在分子中原子的量子理论（QTAIM）术语中，它也表现为在键临界点处的正拉普拉斯（正拉普拉斯）。[2b对于第一行元素，该标准似乎与VB计算很好地相关。然而，对于第2行元素，QTAIM将正LAP分配给正常共价键，例如在H3 SiO2 SiH 3和六硅环己烷中。[5]再加上大多数计算化学家使用基于分子轨道（MO）而不是基于VB的计算工具，这使得我们寻找基于MO的CS键识别标准。等键和同键反应首先由Pople等人提出。[6]大约四十年前，作为克服计算水平缺陷的一种手段。使用这些反应背后的假设是，不是薛定谔方程的精确解的每个计算水平（因此，所有计算）以不同的精度描述不同的键。因此，如果每种类型的键的数量在反应的两侧保持相等，则不足应该相互抵消，并且方程的结果应该与计算水平无关（并且与现实非常相似）。事实上，这种方法已被用于数百篇论文，并显示出良好的效果。在这里，我们以相反的方式运用这一原则。也就是说，设计了同键反应，使得所讨论的键与纯共价键（例如H3 C3 CH 3）相比。如果所讨论的键是纯共价键，那么等键（或同键）反应的结果将不依赖于计算水平。然而，如果所讨论的键是不同类型的，例如CS键，则预计结果将显示出对计算水平的很大依赖性。[七]《中国日报》
The chemical bond is one of the fundamental concepts in chemistry. For over a century chemists recognized two types of intramolecular (ie between atoms in the same molecule, not between molecules) chemical bonds [1]—covalent and ionic bonds. Recently, Shaik, Hiberty, and their collaborators [2] introduced a new type of chemical bond—the charge-shift (CS) bond. This type of bond can be found in many compounds, as has been demonstrated in numerous papers.[3, 4] It occurs when the bonding electrons have high kinetic energy due to repulsion by core electrons (eg F2). It also occurs in abnormal bonding situations (eg propellanes). In valence bond (VB) terms, this type of bond is manifested by an unusually large resonance energy between the covalent and ionic canonic forms of the given bond. It is also manifested, in quantum theory of atoms in molecules (QTAIM) terminology, by a positive Laplacian (LAP) at the bond-critical point.[2b, 4] For 1st row elements, this criterion seems to correlate well with VB calculations. However, for 2nd row elements, QTAIM assigns positive LAPs to normal covalent bonds, such as in H3SiÀSiH3 and hexasilacyclohexane.[5] This, coupled with the fact that most computational chemists use molecular orbital (MO)-based rather than VB-based computational tools, led us to search for MO-based criterions for the identification of CS bonds. Isodesmic and homodesmic reactions were first suggested by Pople et al.[6] about four decades ago, as a means to overcome computational level deficiencies. The assumption behind the use of these reactions is that each computational level that is not the exact solution of the Schrçdinger equation (therefore, all computations) describes different bonds with different accuracies. Consequently, if the number of bonds of each type is kept equal on both sides of the reactions, the deficiencies should cancel each other out and the result of the equation should be independent of the computational level (and very similar to reality). Indeed, this approach has been used in hundreds of papers and was shown to yield good results. Here we employ this principle in the opposite manner. Namely, homodesmic reactions are devised such that the bonds in question are compared to pure covalent bonds, eg H3CÀCH3. It is expected that if the bond in question is of a purely covalent nature, the results of the isodesmic (or homodesmic) reactions will not depend on the computational level. However, if the bond in question is of a different type, such as a CS bond, it is expected that the results will show a large dependence on the computational level.[7]