We have developed a novel code for non-periodic ab initio molecular dynamics based on density functional method employing real space grids and simulated the SN2 reaction of CH3Cl and OH- in gas phase. In particular in this paper, we describe the details of the code. Furthermore, we have parallelized the code by using MPI (Message Passing Interface) and illustrated the efficiency of the real space grid method for the parallel computing.
The stacking interaction of 5, 10, 15, 20-tetramethylporphyrinatonickel (II) was calculated as a trimer by the molecular mechanics version 2 using both the point-charge model and the three-center charge model. The stacking structure for the complex was reproduced fairly well by both methods; the obtained structures are roughly in agreement with the crystal structure. By using the three-center charge model, the minimum displaced distance between each stacking molecule was consistent with the X-ray structures. On the other hand, the distance obtained from the method using the point-charge model was slightly longer. The obtained Ni—N bonds by both methods were considerably longer compared with the X-ray results. The factor for this elongation is discussed.
Four-Leaf molecular devices were studied by the MOPAC-PM3 semi-empirical calculation method. The systems were made only from some hydrogen atoms and some carbon atoms. Therefore, excellent environment conformity in the systems could be expected. Furthermore, because it had strange structure, the systems were interesting target compounds in organic chemistry.
In order to analyze the interaction between proteins, smoothed atomic fractal dimension (SAFD) score which was defined by Petit and Bowie for each atom of a protein was mapped on to the protein surface and visualized. This method was applied to about forty protein dimers (or DNA and protein). It was found that the interaction sites of proteins tend to include regions with relatively large SAFD, especially for heterodimers and spherical-shaped proteins with contact area less than around 2000Å2.
Colored geological column charts based on petrographic observations of well cuttings and geophysical logging are often used for predicting the developing direction of productive zones in an oil or gas reservoir. However, interpreting a column chart is not simple, and is generally based on geophysical information. Here, we propose the use of colored charts based on the chemical composition of minerals in well cuttings. When analyzed by principal component analysis, these charts allow productive zones in an oil or gas reservoir to be determined rapidly.