Lozynski Marek, Rusinska-Roszak Danuta
Faculty of Chemical Technology, Poznan University of Technology, Berdychowo 4, 60-965, Poznan, Poland.
Faculty of Chemical Technology, Poznan University of Technology, Berdychowo 4, 60-965, Poznan, Poland.
J Mol Graph Model. 2021 Jun;105:107884. doi: 10.1016/j.jmgm.2021.107884. Epub 2021 Mar 3.
A predictive model for intramolecular hydrogen bond energy (E) calculation of polyaromatic ortho-hydroxyaldehydes based on a set of small, functionalized hydrocarbons is developed. The complete data set of 18 compounds was used for this study. The model is based on one of four optional categories of molecular descriptors: geometric, spectroscopic, bond order and topological indices. The model of Wiberg bond indices (WBIs) as descriptors of the CC involved bond based on stepwise regression has acceptable prediction abilities for 14 structures of ortho-hydroxyformylobenzo[a]pyrene derivatives already at the semiempirical level. The presented correlation enables a significantly more rapid and quantitative description of the hydrogen bonding strength than the much more time-consuming MTA method. Thus, WBIs are shown to provide a reliable means for fast prescreening of the energy of chelate hydrogen bonds potentially for any polyaromatic derivatives.
基于一组小型功能化烃类开发了一种用于计算多环芳族邻羟基醛分子内氢键能（E）的预测模型。本研究使用了包含18种化合物的完整数据集。该模型基于分子描述符的四个可选类别之一：几何、光谱、键序和拓扑指数。基于逐步回归的Wiberg键指数（WBI）作为CC相关键描述符的模型，对半经验水平下的14种邻羟基甲酰基苯并[a]芘衍生物结构具有可接受的预测能力。与耗时得多的MTA方法相比，所提出的相关性能够更快速、定量地描述氢键强度。因此，WBI被证明是一种可靠的手段，可用于快速预筛选任何多环芳族衍生物中螯合氢键的能量。
