Ab-Initio Identification of Atoms in Molecules

理论和计算化学项目正在支持爱荷华州立大学的K.Ruedenberg教授。未来两年将开展的研究将集中于开发一种新的量子化学理论，这将导致相应的新的计算化学代码。其目的是在严格的从头计算理论框架内定量确定分子中的原子，并根据这些原子之间的相互作用来定量表达结合能。形变原子通过其最优波函数确定，其轨道受限于全价空间MCSCF分子轨道空间。通过密度矩阵分析确定了定向的准原子轨道，最大限度地参与了特定的键，并且具有明确的电子布居。结合能表现为原子内推进和原子间非键排斥、电子共享和电荷转移的贡献之和。通过确定无电荷转移的最佳分子MCSCF波函数，确定了电荷转移所产生的能量增益。共价能量增益来自代表原子间电子迁移的组态。该分析不仅适用于平衡几何，也适用于非平衡几何，并将扩展到全价空间之外。尽管两个世纪以来，化学实验一直有效地受到分子由原子组成的概念的指导，但关于这个建立得很好的经验模型的理论基础，主要悬而未决的问题一直持续到今天，而且它还没有在严格的量子力学计算的背景下进行量化。这是因为分子中的原子概念实际上不是基础理论中的一个基本概念。最近，鲁登堡研究小组开发了一种新的分析方法，能够在严格的分子计算中识别原子亚单位。在这种理论方法的帮助下，原子以及原子之间的相互作用将在分子中得到定量测定。即将开发的方法有望通过将理论计算的结果与实验者的直觉推理更直接地联系起来，使理论计算的结果对实际化学工作更有用。

The Theoretical and Computational Chemistry program is supporting Prof. K. Ruedenberg at Iowa State University. The research to be carried out over the next two years will focus on the development of a new quantum chemical theory that will result in corresponding new computational chemistry codes. The objective is the quantitative determination of atoms in molecules within the framework of rigorous ab-initio theory, and the quantitative expression of binding energies in terms of interactions between such atoms. The deformed atoms are determined through their optimal wave functions in terms of orbitals restricted to the full-valence-space MCSCF molecular orbital space. Oriented quasiatomic orbitals, maximally involved in specific bonds and with unambiguous electron populations, are determined through a density matrix analysis. The binding energy appears as the sum of contributions due to intra-atomic promotion and inter-atomic non-bonded repulsion, electron sharing and charge transfer. The energy gain due to charge transfer is identified by determining the optimal molecular MCSCF wave function without charge transfer. The covalent energy gain is derived from the configurations representing electron migrations between atoms. The analysis applies to non-equilibrium as well as to equilibrium geometries and will be extended beyond the full valence space. Although for two centuries chemical experimentation has been effectively guided by the concept of molecules being built from atoms, major unsolved problems have persisted to this day regarding the theoretical foundation of this well established empirical model and it has not yet been quantified within the context of rigorous quantum mechanical calculations. This is because the atoms-in-molecules concept is, in fact, not a fundamental one in basic theory. Recently, the Ruedenberg research group has developed a new analysis that is able to identify atomic sub-units in rigorous molecular calculations. With the help of this theoretical approach, atoms, as well as the interactions between them, will be quantitatively determined in molecules. The method to be developed is expected to make the results of theoretical computations more useful for practical chemical work by relating them more directly to the intuitive reasoning of experimentalists.