Anorganisch-Chemisches Institut, Ruprecht-Karls-Universität Heidelberg, Im Neuenheimer Feld 270, 69120, Heidelberg.
Chemphyschem. 2021 May 17;22(10):935-943. doi: 10.1002/cphc.202100150. Epub 2021 May 4.
The computed fluoride ion affinity (FIA) is a widely applied descriptor to gauge Lewis acidity. Like every other single-parameter Lewis acidity scale, the FIA metric suffers from the one-dimensionality, that prohibits addressing Lewis acidity by the multidimensionality it inherently requires (i. e., reference Lewis base dependency). However, a systematic screening of computed affinities other than the FIA is much less developed. Herein, we extended our CCSD(T)/CBS benchmark of different density functionals and the DLPNO-CCSD(T) method for chloride (CIA), methide (MIA), hydride (HIA), water (WA), and ammonia (AA) affinities. The best performing methods are subsequently applied to yield nearly 800 affinities for 183 p-block element compounds of group 13-16 with an estimated accuracy of <10 kJ mol . The study's output serves as a consistent library for qualitative analyses and a training set for future statistical approaches. A first holistic correlation analysis underscores the need for a multidimensional description of Lewis acidity.
计算得到的氟离子亲合能（FIA）是一种广泛应用于衡量路易斯酸度的描述符。与其他单参数路易斯酸度标度一样，FIA 度量法存在一维性，这限制了通过其内在需要的多维性（即参考路易斯碱依赖性）来解决路易斯酸度问题。然而，对除 FIA 以外的计算得到的亲合力进行系统筛选的研究要少得多。在此，我们扩展了我们的 CCSD(T)/CBS 基准测试不同密度泛函和 DLPNO-CCSD(T)方法，用于测定氯化物（CIA）、甲化物（MIA）、氢化物（HIA）、水（WA）和氨（AA）的亲合力。随后，最好的方法被应用于生成近 800 种 13-16 族 p 区元素化合物的亲合力，估计准确度<10 kJ mol-1。该研究的结果为定性分析提供了一致的库，并为未来的统计方法提供了训练集。首次整体相关分析强调了需要对路易斯酸度进行多维描述。
