the about uniaxial tension

Dear all users:
In LAMMPS website, some tutorials are presented in LAMMPS tutorials page ( The LAMMPS Beginner Help 3 shows how to deform a three-dimensional periodic simulation cell in uniaxial tension for aluminum and the simulation was performed in the NPT ensemble. however, the LAMMPS Beginner Help 9 shows how to deform a nanowire but the simulation was performed in the NVT ensemble, I read the related commands about fix npt and nvt , and did not find the reason that why they used different ensemble. Actually they both used periodic conditions in three dimensions, and similar pair_style of metal atoms. Are their real loading condition different? can someone give me some advice? thanks very much!
best wishes!

Xuepeng Liu

  1. That’s not the LAMMPS website. It is the wiki website from CAVS/HPC at Mississippi State University
  2. They are different structures and therefore different simulations. One is a nanowire while the other is a aluminum crystal.
  3. In the case using NPT during deformation, it is not periodic in the direction being deformed. More accurately, it can’t be.

By using NPT, it will allow the sides not being deformed to change with the structure (I’m not saying that’s why it is NPT, just making a comment).
I would guess Dr. Tschopp (the individual who made those pages) used NVT with the nanowire so that the box would not collapse in on the empty box around the nanowire.


Dear Michael:
Thank you very much for your advice! one more question, In general, we had better deform a crystal under uniaxial tension in npt ensemble? and the result between npt and nvt will not have a large gap? All i want to know is if i want to calculate the elastic modulus of atom crystal, which ensemble could be better? thanks very much!
best wishes
Xuepeng Liu

That seems like something you need to figure out from literature… This is a mailer list to get help with LAMMPS and specific questions on MD related topics as they pertain to LAMMPS-not a how to on how to set up your project.


Dear Nichael:
Thanks for your help!
Xuepeng liu