Development of generation of large-male potential energy surfaces num local interpolation and molecular mechanics

大雄势能面num局部插值和分子力学生成的发展

基本信息

批准号：
16550025
负责人：
ISHIDA Toshimasa
金额：
$ 2.16万
依托单位：
Kyoto University
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (C)
财政年份：
2004
资助国家：
日本
起止时间：
2004 至 2007
项目状态：
已结题

项目摘要

In large scale systems, so called QM/MM methods have been employed in the quantum chemistry society and tens of trajectories (even only one trajectory) sometimes have been employed to extract the picture of reaction mechanism. However, it is still difficult for us to obtain the number of trajectories which is statistically significant because potential energy calculations are time consuming. The statistically significant result may be drastically different from the picture that was obtained from tens of trajectory calculations. Thus, it is desirable to construct a method to evaluate potential energy surface fast. We attempted to combine a local interpolation scheme, IMLS/Shepard scheme, and molecular mechanics, but the attempt is still under way.Thus, we have undertaken a direct dynamics for a biomolecule, retinal, instead. We described the dynamics of retinal using Zhu-Nakamura formula for non-adiabatic transitions from the first (photo) exited state to the ground state. We treated all the degrees of freedom in our direct dynamics although we originally planned to describe small number of degrees of freedom icing IMLS/Shepard scheme and the other degrees of freedom using molecular mechanics. We have found that retinal (at least in gas phase) experiences two-step relaxation after the deexcitation through non-adiabatic transition.Also, we have applied our IMLS/Shepard scheme to the O+HCl system. Large scale trajectory calculations using three potential energy surfaces have been carried out to obtain reactive differential cross sections and the results are in agreement with experiment We have found that the dynamics is determined by that in the ground state although the two excited states are in a relatively low energy region.

在大规模系统中，量子化学学会已采用所谓的QM/MM方法，有时采用数十条轨迹（甚至只有一条轨迹）来提取反应机理的图像。然而，由于势能计算非常耗时，我们仍然很难获得具有统计意义的轨迹数量。统计上显着的结果可能与通过数十次轨迹计算获得的图片有很大不同。因此，需要构建一种快速评估势能面的方法。我们尝试将局部插值方案、IMLS/Shepard 方案和分子力学结合起来，但该尝试仍在进行中。因此，我们对生物分子（视网膜）进行了直接动力学研究。我们使用 Zhu-Nakamura 公式描述了从第一（照片）激发态到基态的非绝热跃迁的视网膜动力学。尽管我们最初计划使用分子力学描述少量的结冰 IMLS/Shepard 方案自由度和其他自由度，但我们还是处理了直接动力学中的所有自由度。我们发现视网膜（至少在气相中）在通过非绝热转变去激发后经历两步弛豫。此外，我们已将我们的 IMLS/Shepard 方案应用于 O+HCl 系统。利用三个势能面进行大规模轨迹计算，得到电抗微分截面，结果与实验一致。我们发现，尽管两个激发态处于相对较弱的状态，但动力学是由基态决定的。低能量区。