Suppr超能文献

通过缩放从头算吉布斯自由能的振动贡献估算羧酸的精确气相酸度。

Accurate gas phase acidities of carboxylic acids estimated by scaling the vibrational contribution of ab initio Gibbs free energies.

作者信息

Cysewski Piotr

机构信息

Physical Chemistry Department, Collegium Medicum, Nicolaus Copernicus University, Kurpińskiego 5, 85-950, Bydgoszcz, Poland.

出版信息

J Mol Model. 2007 Jul;13(6-7):801-8. doi: 10.1007/s00894-007-0200-0. Epub 2007 Apr 3.

Abstract

The gas phase Gibbs free energies deltaG(T) of dissociation reaction of 14 carboxylic acids were calculated on the SCF, as well as G3 and CBS-Q levels. Corresponding accuracies were critically compared with experimental data. Since all of the results suffer from systematic errors, the procedure of scaling of thermal contribution to Gibbs free energy was applied for minimizing differences between theoretical and experimental values of deltaG(T). Two parameters were adjusted, namely the scaling of thermal contribution to Gibbs free energy of neutral and anionic forms. The presented results suggest the great effectiveness of such a procedure since for all applied basis sets within the SCF framework the achieved accuracy was below the experimental error. Besides, the proposed low-cost approximation method leads to precision comparable to or even exceeding the quality offered by more sophisticated composite quantum chemistry methods. The extension of the set of training molecules up to 82 has an insignificant impact on the overall quality of deltaG(T) estimation, which suggests that a wisely chosen set of reference data may be used for the characteristics of the whole class of compounds. There is a straightforward way for the analysis of acidities/basicities of other classes of chemicals such as DNA bases, alcohols, phenols, amines, amino acids, etc.

摘要

在SCF以及G3和CBS-Q水平上计算了14种羧酸解离反应的气相吉布斯自由能ΔG(T)。将相应的精度与实验数据进行了严格比较。由于所有结果都存在系统误差,因此采用了对吉布斯自由能热贡献进行缩放的程序,以最小化ΔG(T)理论值与实验值之间的差异。调整了两个参数,即中性和阴离子形式的吉布斯自由能热贡献的缩放。给出的结果表明了该程序的巨大有效性,因为在SCF框架内对于所有应用的基组,所达到的精度低于实验误差。此外,所提出的低成本近似方法导致的精度与更复杂的复合量子化学方法相当甚至超过其质量。将训练分子集扩展到82个对ΔG(T)估计的整体质量影响不大,这表明明智选择的一组参考数据可用于整个化合物类别的特征描述。对于分析其他类别的化学物质(如DNA碱基、醇、酚、胺、氨基酸等)的酸度/碱度,有一种直接的方法。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验