Is CP2K a good choice for AIMD simualtion for iron oxide reduction by hydorogen using CP2K

Kejiang Li

unread,

Apr 15, 2022, 7:45:11 AM4/15/22

to cp2k

Dear CP2K users,

I am currently doing atomistic simulation to study the reduction process of iron oxides (Fe2O3, Fe3O4 and FeO) by hydrogen, which is the key reaction in hydrogen metallurgy and catalysis process using iron/iron oxides.

I have done some ReaxFF simulations using LAMMPS and the results looks good, but I need to do AIMD simulations to validate the results and improve the ReaxFF potential parameters.

I have tried a lot to do BOMD and CPMD to simulate the reduction reaction using Quantum Espresso 7.0. For BOMD, the computation speed is two slow, for CPMD there seems to be some bugs to consider magnetism and U for iron oxides in cp.x. I learned that CP2K can do AIMD very efficiently, but I have never used CP2K before.

Therefore, I have two questions to consult before I started to devote a lot of effort/time to learn CP2K:

1. Is it possible or a good choice to use CP2K to conduct AIMD for the hydrogen reduction of iron oxides. I think either BOMD or CPMD should be OK for my study.

2. Can spin, magnetism and U be easily handled with CP2K? It seems to be a common practice to consider spin, magnetism and U in any DFT-level studies about iron oxides.

3. If it is a good choice to use CP2K for my study, could anyone please provide me some sample input scripts to do AIMD simulation for chemical reactions using CP2K.

Any suggestions or comments will be much appreciated.

Thanks a lot.

All the best,

Kejiang

Dr. Kejiang Li

School of Metallurgical and Ecological Engineering, University of Science and Technology Beijing Xueyuan Rd., Haidian District, Beijing 100083, P. R. China

https://kejiangli.com/