Dear Haofan,

I have checked your input script and edited some parts of it.

First, I think the name of the ReaxFF potential has changed in the new version and its syntax is "pair_style reaxff", for my version (29 Oct 2020) I use "pair_style reax/c, I recommend checking the Reax syntax for your lamps version.

Secondly, Reax needs a charge attribute for atoms so your atom_style must also have a charge attribute, like full or charge atom_style.

I have converted your data file to full, used atom_style full, and changed the syntax of pair_style to reax/c.

Remember, Reax has its own function to define bonds and you don't need to add bonds in your data file.

I have attached all files, hope it helps.

With best wishes.

Nader Malih Thanks for your explaining, Nader! The model runs well after adding fixed boundary conditions. I'm more clear now.

Haofan Sun

You're welcome.

I'm glad to hear that you solved it.

Dear Haofan Sun and Nader Malih ,

I am also going to use REAXFF for my model, but I have a problem at the very beginning step. I have installed the USER-REAXC package using make-yes-reaxc, and with make ps, it is shown "yes", and then I also recompiled the lamps with make serial. However when I am tryin the lip_serial -h command to check the packages I see that it is not included yet. Could please advice me how I can tackle this problem?

Hi Afsaneh Khajeh ,

I hope you have already solved the problem.

I didn't encounter this problem, but I recommend using CMake to build LAMMPS (as I do myself), it's much easier to use. Sometimes removing all directories and extracting source code files and compiling again also helps to solve the problem. and another way is to use the precompiled versions.

All the best

Hi Afsaneh Khajeh , I guess the problem is "make serial" part, you can try with a parallel build see if it works.