Department of Chemistry, Johns Hopkins University, Baltimore, Maryland 21218, United States.
Institute of Modern Physics, Northwest University, Xi'an, Shaanxi 710069, People's Republic of China.
J Chem Theory Comput. 2021 Jul 13;17(7):4157-4168. doi: 10.1021/acs.jctc.1c00370. Epub 2021 Jun 16.
In our recent work, a diabatic Hamiltonian that couples the S and S states of formaldehyde was constructed using a robust fitting-and-diabatizing procedure with artificial neural networks, which is capable of representing adiabatic energies, energy gradients, and derivative couplings over a wide range of geometries including seams of conical intersection. In this work, based on the diabatization of S and S, the spin-orbit couplings between singlet states (S, S) and triplet state T are also determined in the same diabatic representation. The diabatized spin-orbit couplings are then fit with a symmetrized neural-network functional form. The spin-orbit couplings are well reproduced in large configuration space. Together with the neural-network-based potential energy surface for T, the full quasi-diabatic Hamiltonian for the S, S, and T states is completed, enabling a unified description of both internal conversion and intersystem crossing in formaldehyde. The vibrational levels on the three adiabatic states are found to be in good agreement with known experimental band origins.
在我们最近的工作中,使用带有人工神经网络的鲁棒拟合-绝热处理程序构建了一个耦合甲醛的 S 和 S 态的非绝热哈密顿量,该哈密顿量能够在包括锥形交叉缝在内的广泛几何范围内表示绝热能量、能量梯度和导数耦合。在这项工作中,基于 S 和 S 的绝热处理, singlet 态 (S, S) 和 triplet 态 T 之间的自旋轨道耦合也在相同的绝热表示中确定。然后,用对称化神经网络函数形式拟合绝热的自旋轨道耦合。自旋轨道耦合在大构型空间中得到了很好的再现。与基于神经网络的 T 势能面结合,完成了 S、S 和 T 态的全拟绝热哈密顿量,实现了甲醛中内转换和系间窜越的统一描述。三个绝热态的振动能级与已知的实验带起源很好地吻合。