Introducing DDEC6 atomic population analysis: part 3. Comprehensive method to compute bond orders
Introducing DDEC6 atomic population analysis: part 3. Comprehensive method to compute bond orders
复制标题
DOI：
10.1039/c7ra07400j
复制
发表时间：
2017-01-01
期刊：
影响因子：
3.9
通讯作者：
Manz, Thomas A.
中科院分区：
文献类型：
--
作者：
Manz, Thomas A.
自从世纪前刘易斯对化学键的开创性工作以来，发展一种综合的方法来计算键级一直是化学家们所回避的问题。在这里，一个计算效率高的方法来解决这个问题，并证明了不同的材料，包括从每个化学组和周期的元素。该方法适用于非磁性，共线磁性，和非共线磁性材料与本地或离域键合电子。研究的例子包括拉伸的O-2分子，26个双原子分子，3d和5d过渡金属固体，每个晶胞具有1至8748个原子的周期性材料，生物分子，超配位分子，缺电子分子，氢键系统，过渡态，刘易斯酸碱络合物，芳香族化合物，磁性系统，离子材料，分散结合系统，纳米结构和其他材料。从近零键到高阶键进行了研究。每个原子的键级和键级之和在各种键合类型中都是准确的：金属键合、共价键合、极性共价键合、离子键合、芳香键合、配位键合、超配位键合、缺电子多中心键合、无规键合和氢键合。该方法产生类似的结果相关的波函数和密度泛函理论的输入和不同的SZ值的自旋多重态。该方法只需要电子和自旋磁化强度密度分布作为输入，并具有计算成本线性缩放的原子数量的单位细胞。没有一种方法是通用的。该方法不适用于超临界态、高度依赖时间的态、某些极高能量的激发态和核反应。
Developing a comprehensive method to compute bond orders is a problem that has eluded chemists since Lewis's pioneering work on chemical bonding a century ago. Here, a computationally efficient method solving this problem is introduced and demonstrated for diverse materials including elements from each chemical group and period. The method is applied to non-magnetic, collinear magnetic, and noncollinear magnetic materials with localized or delocalized bonding electrons. Examples studied include the stretched O-2 molecule, 26 diatomic molecules, 3d and 5d transition metal solids, periodic materials with 1 to 8748 atoms per unit cell, a biomolecule, a hypercoordinate molecule, an electron deficient molecule, hydrogen bound systems, transition states, Lewis acid-base complexes, aromatic compounds, magnetic systems, ionic materials, dispersion bound systems, nanostructures,and other materials. From near-zero to high-order bonds were studied. Both the bond orders and the sum of bond orders for each atom are accurate across various bonding types: metallic, covalent, polar-covalent, ionic, aromatic, dative, hypercoordinate, electron deficient multi-centered, agostic, and hydrogen bonding. The method yields similar results for correlated wavefunction and density functional theory inputs and for different SZ values of a spin multiplet. The method requires only the electron and spin magnetization density distributions as input and has a computational cost scaling linearly with increasing number of atoms in the unit cell. No prior approach is as general. The method does not apply to electrides, highly timedependent states, some extremely high-energy excited states, and nuclear reactions.