|
|
本帖最后由 X教授 于 2020-10-21 03:53 编辑
最近想研究用高斯计算酞菁镍电催化CO2还原过程(之前看到的文献都是用第一性原理计算的),看到一篇Nature Energy的文章(https://doi.org/10.1038/s41560-020-0667-9),是用高斯计算的,所以想先重复一下文献的工作。酞菁镍电催化CO2的还原过程主要有三步过程, 第一步是CO2在酞菁镍(*代替)表面转变成COOH(*+CO2+(H+)+(e-)=*COOH), 第二步COOH转变成CO(*COOH+(H+)+(e-)=*CO+H2O), 第三步CO脱附(*CO=*+CO),然后是计算方法,这是原文献使用的方法:Computational details. All DFT calculations on metal phthalocyanine molecules in this study were performed using the Gaussian 09 program. A PBE0 functional with D3 correction (Becke–Johnson damping) was adopted for its robustness and dispersion corrections, which make it widely accepted as the proper functional to study the reactions of transition metal complexes. The Stuttgart–Dresden pseudopotential and double-ξ valence basis set were used for transition metal atoms (nickel). For all other main group elements (H, C, N, O), the all-electron 6–31G* basis set was used. The geometric structures of all species were fully optimized. Harmonic vibrational frequencies were also computed, whose result showed that all reaction intermediates have no imaginary frequency. The gas-phase Gibbs free energies (G) of high- and low-spin forms of all intermediates, G, were calculated with the harmonic potential approximation at optimized structures at 298.15 K and 1 atm to determine the ground states in realistic conditions. These ground-state structures were then used in the mechanism study。
我的计算流程是:使用高斯16 A01优化酞菁镍(*),酞菁镍COOH(*COOH)以及酞菁镍CO(*CO)的结构,泛函使用pbe1pbe, C, H, O基组6–31G*,Ni使用SDD基组和赝势,同时使用empiricaldispersion=gd3bj,然后进行频率计算。结构优化过程中我发现如果不对酞菁镍骨架进行固定,酞菁镍COOH以及酞菁镍CO的结构会严重变形,因此我选择冻结酞菁镍骨架,最后得到优化后的结构以及Sum of electronic and thermal Free Energies:
E(*) = -1835.956069 Hartree
E(*COOH) = -2025.29839 Hartree
E(*CO) = -1949.5489 Hartree
E(CO2) = -188.39227 Hartree
E(H2O) = -76.32206 Hartree
E(H+) = -0.01 Hartree
E(CO) = -113.19593 Hartree
因此,起始物的能量E1 = E(*) + E(CO2) + 2E(H+) = -2025.47406 Hartree, 第一步能量E2 = E(*COOH) + E(H+) = -2025.28839 Hartree, 第三步能量E3 = E(*CO) + E(H2O) = -2025.87096 Hartree, 最后一步能量E4 = E(*) + E(CO) + E(H2O) = -2025.47406 Hartree, 最后所有能量减去第一步能量再转换成eV得到能量变化图如下,结果非常奇怪,理论上E1应该是能量最低状态,而且第一步能量差应该很大(几十eV,太夸张)。
这是文献计算值,每一步能量差不到2eV, 而且E1与E4相同。所以想求教到底是什么原因导致差异如此之大的结果?
|
-
Graph2.jpg
(284.41 KB, 下载次数 Times of downloads: 61)
我计算的能量变化图
-
Untitled.jpg
(155.15 KB, 下载次数 Times of downloads: 64)
|