MCTDH study on vibrational states of the CO/Cu(100) system.
MCTDH study on vibrational states of the CO/Cu(100) system.
复制标题
DOI：
10.1063/1.4826258
复制
发表时间：
2013-10
期刊：
影响因子：
--
通讯作者：
Qingyong Meng;H. Meyer
中科院分区：
文献类型：
--
作者：
Qingyong Meng;H. Meyer
利用最近报道的[R. 1]，对CO/Cu（100）体系的振动基本原理和泛音进行了全（6D）和降（4D和2D）维多构型时变Hartree （MCTDH）计算。马夸特，F. Cuvelier， R. A. Olsen， E. J. Baerends， J. C. Tremblay和P. Saalfrank， J. Chem。物理学报，32(4):555 - 556。为了使用海德堡封装（http://mctdh.uni-hd.de）有效地执行MCTDH计算，SAP-PES首先被改装为产品和形式。然后进行广泛的MCTDH计算，包括彻底的收敛检查，以确保我们的结果的准确性。然后进行全维改进弛豫和/或块改进弛豫以获得CO/Cu（100）的振动基态和激发态。此外，我们研究了受挫旋转（R模式）和垂直CO- cu拉伸（S模式），以及C-O拉伸，使用四维哈密顿量，其中包括CO与表面之间的距离，z， CO的键长，R， CO的取向角，θ，和φ。这个四维哈密顿量的补充，一个只包括x和y坐标的二维哈密顿量，被用来替代地计算受挫平移的状态（T模式）。降维结果与全维结果的比较表明，R或S模式与T模式之间的耦合不影响R和S模式的基本音和泛音，但对T模式的基本音和泛音有相当大的影响。对于T模的基元，6D计算得到25.09 cm（-1）的能量，接近31.8 cm（-1）的实验结果[a]。P. Graham， F. Hofmann， J. P. Toennies， G. P. Williams， C. J. Hirschmugl和J. Ellis， J. Chem。[j] .中国科学：自然科学进展，1998,11(2):1 - 2。马夸特，F. Cuvelier， R. A. Olsen， E. J. Baerends， J. C. Tremblay和P. Saalfrank， J. Chem。尽管这两项研究都是基于SAP-PES。另一方面，目前的计算对于S模态和C-O拉伸的基本原理给出了与以前相似的结果[R]。马夸特，F. Cuvelier， R. A. Olsen， E. J. Baerends， J. C. Tremblay和P. Saalfrank， J. Chem。物理学报，32(4):555 - 557。此外，对于横向T和R模式，本计算给出了相同的激励模式。我们的计算还表明，R模式与表面之间的耦合比T模式与表面之间的耦合弱。
Full (6D) and reduced (4D and 2D) dimensional multiconfiguration time-dependent Hartree (MCTDH) calculations for the vibrational fundamentals and overtones of the CO/Cu(100) system are carried out using the recently reported [R. Marquardt, F. Cuvelier, R. A. Olsen, E. J. Baerends, J. C. Tremblay, and P. Saalfrank, J. Chem. Phys. 132, 074108 (2010)] SAP potential energy surface (PES). To efficiently perform MCTDH calculations with the Heidelberg package (http://mctdh.uni-hd.de), the SAP-PES is first refitted in a sum-of-products form. Then extensive MCTDH calculations are carefully performed including thorough convergence checks to ensure the accuracy of our results. Full dimensional improved-relaxations and/or block-improved-relaxations are then performed to obtain vibrational ground and excited states of CO/Cu(100). In addition, we investigate the frustrated rotation (R mode) and vertical CO-Cu stretch (S mode), as well as C-O stretch, using a 4D Hamiltonian, which includes the distance between CO and the surface, z, bond length of CO, r, and orientation angles of CO, θ, and φ. The complement of this 4D Hamiltonian, a 2D Hamiltonian including only x and y coordinates, is used to alternatively calculate the states of frustrated translation (T mode). The comparison of reduced- with full-dimensional results show that the couplings between the R or S mode and the T mode do not influence the fundamentals and overtones of the R and S modes, however, do considerably influence those of the T mode. For the fundamental of the T mode, the 6D calculation yields an energy of 25.09 cm(-1), a value closer to the experimental result of 31.8 cm(-1) [A. P. Graham, F. Hofmann, J. P. Toennies, G. P. Williams, C. J. Hirschmugl, and J. Ellis, J. Chem. Phys. 108, 7825-7834 (1998)] than previously reported 19.6 cm(-1) [R. Marquardt, F. Cuvelier, R. A. Olsen, E. J. Baerends, J. C. Tremblay, and P. Saalfrank, J. Chem. Phys. 132, 074108 (2010)] although both investigations are based on the SAP-PES. On the other hand, the present calculations give similar results for the fundamentals of the S mode and C-O stretch to previous ones [R. Marquardt, F. Cuvelier, R. A. Olsen, E. J. Baerends, J. C. Tremblay, and P. Saalfrank, J. Chem. Phys. 132, 074108 (2010)]. Furthermore, for the lateral T and R modes, the present calculations give the same excitation pattern. Our calculations also show that the couplings between the R mode and surface are weaker than those between the T mode and surface.