| 研究実績の概要 |
In this year, we have succeeded in the improvements of parameters for the density-functional tight-binding (DFTB) method for metal nanoparticles containing Fe, Ni, Co, and Cu in combination with C, H, and N. At first, Mr. Aulia Hutama learned both the automatic repulsive parameterization utility described by Bodrog et al. in the Journal of Chemical Theory and Computation, 2011, 7, 2654. He also learned to perform manual parameterization based on the Birch-Murnaghan equation of state. In early 2015, Dr. Liubov Antipina visited us and performed extensive benchmarks on model clusters of different sizes and compositions from DFTB-based molecular dynamics simulations and compared the resulting geometries and energetics/forces by performing first-principles GGA DFT calculations using VASP. We found excellent agreement between DFTB and DFT results. In addition, we located transition states for combination reactions such as HC--CH on the metal nanoparticles using nudged elastic band calculations in both DFT and DFTB approaches, and found satisfactory agreement between them. Thus, the validation stage is now completed for performing on-the-fly, direct kinetic Monte Carlo (flyKMC) simulations of carbon cap nucleation on metal nanoparticles.
The purchase of additional 2 compute nodes with 16 CPU cores using this grant-in-aid was essential to perform these extensive benchmark calculations.
|
| 今後の研究の推進方策 |
We already wrote a preliminary GRRM-DFTB interface, which however requires tuning and modifications for flyKMC. In particular, we wish to replace the current GRRM code by Ohno and Maeda by our own software that can perform limited transition state searches for reactions of interest, such as carbon-carbon bond formation or breaking on metal nanoparticles. Our goals for the year 2015 are to (i) the flyKMC method using at first the existing GRRM tool for testing, and (ii) implement our own GRRM code. In the year 2016 we will then apply our code in new simulations of metal-catalyzed carbon nanotube and graphene growth simulations. The new studies will shed light on the role of defect healing processes previously largely ignored in our QM/MD simulations, and allow the realistic simulations of nanostructure growth.
|
