A new "spectroscopic" potential energy surface for formaldehyde in its ground electronic state.

A new "spectroscopic" potential energy surface for formaldehyde in its ground electronic state.
复制标题

甲醛基电子态的新“光谱”势能表面。

DOI:
--
复制
发表时间:
2011
影响因子:
4.4
通讯作者:
W. Thiel
W. Thiel
中科院分区:
化学2区
文献类型:
--
作者:
A. Yachmenev;S. Yurchenko;P. Jensen;W. Thiel

文献摘要

被引文献

相似文献

我们报道了甲醛(H(2)(12)C(16)O)在基电子状态下的一个新的“光谱”势能面(PES),通过对文献中现有的甲醛实验光谱数据进行最小二乘拟合,改进了从头算的PES。从头算PES使用CCSD(T)/aug-cc-pVQZ方法在30840个几何形状下计算,覆盖能量范围高达44000 cm(-1)以上的平衡。利用程序TROVE[理论旋转振动能;李建军,刘建军,李建军,等。光谱学学报,2004,26(6)。在J = 0、1、2和5的情况下,将PES解析表达式中的参数值优化到319个反振动能。拟合中的初始参数值为从头算PES的初始参数值,由该PES获得的反振动特征函数作为拟合过程中的基集,并施加约束以确保改进后的PES在实验数据未采样的位形空间区域不会与从头算PES产生非物理偏差。得到的精细化PES称为H(2)CO-2011,再现了现有实验J≤5的数据,均方根误差为0.04 cm(-1)。
We report a new "spectroscopic" potential energy surface (PES) of formaldehyde (H(2)(12)C(16)O) in its ground electronic state, obtained by refining an ab initio PES in a least-squares fitting to the experimental spectroscopic data for formaldehyde currently available in the literature. The ab initio PES was computed using the CCSD(T)/aug-cc-pVQZ method at 30 840 geometries that cover the energy range up to 44 000 cm(-1) above equilibrium. Ro-vibrational energies of formaldehyde were determined variationally for this ab initio PES by means of the program TROVE [Theoretical ROtation-Vibration Energies; S. N. Yurchenko, W. Thiel, and P. Jensen, J. Mol. Spectrosc. 245, 126 (2007)]. The parameter values in the analytical representation of the PES were optimized in fittings to 319 ro-vibrational energies with J = 0, 1, 2, and 5. The initial parameter values in the fittings were those of the ab initio PES, the ro-vibrational eigenfunctions obtained from this PES served as a basis set during the fitting process, and constraints were imposed to ensure that the refined PES does not deviate unphysically from the ab initio one in regions of configuration space not sampled by the experimental data. The resulting refined PES, referred to as H(2)CO-2011, reproduces the available experimental J ≤ 5 data with a root-mean-square error of 0.04 cm(-1).