arXivDaily arXiv每日学术速递 周一至周五更新
arXiv physics/9808017physics.chem-ph
收藏

氢分子激发态的非球高斯基组从头计算

Ab initio calculations with a nonspherical Gaussian basis set: Excited states of the hydrogen molecule

T. Detmer, P. Schmelcher, L. S. Cederbaum

摘要:

提出并讨论了广义非球高斯函数的基组。作为第一个例子,我们报告玻恩奥本海默能量的氢分子。虽然已经得到了精确的结果,我们得出的结论是,H_2是太'简单',允许通过使用非球函数的实质性增益。我们更希望这些函数在大型系统的计算中特别有用。用一组GGTO基组同时计算了{1,3} n_{g,u}对称性各子空间内多个态的势能曲线。我们在此确定了核间距0.8 < R < 1000 a.u.的整个区域。特别地,与调用显式相关函数的计算相比,第四至第六电子状态的结果显示出高精度,例如,相对精度至少为10^{-5}a.u.改进了4^1 π_u^+和4- 6^3 π_u^+的能量,并首次给出了6^3 π_g^+、5^1 π_u^+和6^1 π_u^+态的精确数据。每个子空间内第七至第九电子态的能量数据估计误差为10^{-4}a.u.。发现7^1 <$_g^+和6^1 <$_u^+态在大的核间距处表现出非常宽的深外阱。

英文摘要:

A basis set of generalized nonspherical Gaussian functions (GGTOs) is presented and discussed. As a first example we report on Born-Oppenheimer energies of the hydrogen molecule. Although accurate results have been obtained, we conclude that H_2 is too 'simple' to allow for a substantial gain by using nonspherical functions. We rather expect that these functions may be particularly useful in calculations on large systems. A single basis set of GGTOs was used to simultaneously calculate the potential energy curves of several states within each subspace of {1,3}Σ_{g,u} symmetry. We hereby considerd the entire region of internuclear distances 0.8 < R < 1000 a.u. In particular the results for the fourth up to sixth electronic states show a high accuracy compared to calculations which invoke explicitely correlated functions, e.g. the relative accuracy is at least of the order of magnitude of 10^{-5}a.u. Energies for the 4 ^1Σ_u^+ and 4-6 ^3Σ_u^+ were improved and accurate data for the 6 ^3Σ_g^+, 5 ^1Σ_u^+, and 6 ^1Σ_u^+ state are, to the best of the authors knowledge, presented for the first time. Energy data for the seventh up to the nineth electronic state within each subspace were obtained with an estimated error of the order of magnitude of 10^{-4}a.u. The 7 ^1Σ_g^+ and the 6 ^1Σ_u^+ state were found to exhibit a very broad deep outer well at large internuclear distances.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑