羰基镍簇Ni（CO）n（n=1～4）的结构和Ni—Co键解离性质的密度泛函理论研究

Density Functional Calculations on Structures and Ni—CO Bond Dissociation Energies of Ni(CO)_(n)(n=1—4)

在密度泛函理论框架下，应用不同泛函计算了配合物Ni（CO）n（tn=41～4）的平衡几何构型和振动频率．考察了泛函和基组重叠误差对预测Ni—CO键解离能的影响．计算结果表明，用杂化泛函能得到与实验一致的优化几何构型和较合理的振动频率．对Ni（CO）n（n=2～4）体系，用“纯”泛函，如BP86和BPW91，可得到与CCSD（T）更符合、并与实验值接近的解离能．当解离产物出现单个金属原子或离子（如金属羰基配合物的完全解离）时，BSSE校正项的计算中应保持金属部分的电子结构一致．只有考虑配体基组和不考虑配体基组两种情况下金属的电子构型与配合物中金属的构型一致时，才能得到合理的BSSE校正，从而预测合弹的懈离能．

Density tries and harmonic functional theory（DFT） methods with several functionals were used to determine geomefrequencies of Ni（CO）n （n = 1-4） complexes. In consideration of the zero point energy （ZPE） and basis set superposition error（BSSE） corrections, the nickel carbonyl bond dissociation energies of these complexes have been evaluated. The obtained results show that hybrid DFT methods, such as B3LYP, yield reasonable geometries and vibrational frequencies. For Ni （CO）n （n = 2--4）, the bond dissociation energies predicted by ＂pure＂ DFT methods, such as BP86 and BPW91, are in good agreement with those from the sophisticated coupled-cluster method, CCSD（T）, and the experimental measurements. It was shown that the validity of the calculated bond dissociation energies for NiCO complex or the cumulative binding energies for other complexes strongly depends on an adequate description of BSSE. The calculated BSSE corrections are reasonable only if the electronic structures of the metal with and without ghost basis sets on the ligand positions are identical with that of the metal in metal complexes. The reasonable BSSE correction is the precondition for obtaining valid bond dissociation energies.