Laboratorium für Organische Chemie, ETH Zurich, Vladimir-Prelog-Weg 2, CH-8093 Zurich, Switzerland.
J Am Chem Soc. 2022 May 25;144(20):9007-9022. doi: 10.1021/jacs.2c01381. Epub 2022 May 12.
We report an evaluation of the importance of London dispersion in moderately large (up to 36 heavy atoms) organic molecules by means of a molecular torsion balance whose conformations "weigh" one interaction against another in the absence of solvents. The experimental study, with gas-phase cryogenic ion vibrational predissociation (CIVP) spectroscopy, solid-state Fourier transfer infrared (FT-IR), and single-crystal X-ray crystallography, is accompanied by density functional theory calculations, including an extensive search and analysis of accessible conformations. We begin with the unsubstituted molecular torsion balance, and then step up the complexity systematically by adding alkyl groups incrementally as dispersion energy donors (DEDs) to achieve a degree of chemical complexity comparable to what is typically found in transition states for many regio- and stereoselective reactions in organic and organometallic chemistry. We find clear evidence for the small attractive contribution by DEDs, as had been reported in other studies, but we also find that small individual contributions by London dispersion, when they operate in opposition to other weak noncovalent interactions, produce composite effects on the structure that are difficult to predict intuitively, or by modern quantum chemical calculations. The experimentally observed structures, together with a reasonable value for a reference cation-π interaction, indicate that the pairwise interaction between two -butyl groups, in the best case, is modest. Moreover, the visualization of the conformational space, and comparison to spectroscopic indicators of the structure, as one steps up the complexity of the manifold of noncovalent interactions, makes clear that in silico predictive ability for the structure of moderately large, flexible, organic molecules falters sooner than one might have expected.
我们通过分子扭秤评估了伦敦色散在较大(多达 36 个重原子)有机分子中的重要性,该扭秤的构象在没有溶剂的情况下“权衡”一种相互作用与另一种相互作用。实验研究结合气相低温离子振动预解离(CIVP)光谱、固态傅里叶变换红外(FT-IR)和单晶 X 射线晶体学,同时进行密度泛函理论计算,包括广泛搜索和分析可及构象。我们从未取代的分子扭秤开始,然后通过逐步增加烷基作为色散能供体(DED)来系统地增加复杂性,以达到与有机和有机金属化学中许多区域和立体选择性反应的过渡态相当的化学复杂性程度。我们发现了 DED 具有小的吸引力贡献的明确证据,这在其他研究中已有报道,但我们也发现,当伦敦色散的小个体贡献与其他弱非共价相互作用相反时,它们会对结构产生难以直观预测或通过现代量子化学计算预测的复合效应。实验观察到的结构,以及合理的参考阳离子-π 相互作用值,表明两个 -丁基基团之间的成对相互作用在最佳情况下是适度的。此外,随着非共价相互作用的复杂性不断增加,对构象空间的可视化以及与结构的光谱指标的比较,清楚地表明,对于较大、灵活的有机分子结构的计算预测能力比人们预期的更早失效。