Questions about LAMMPS

Hi Sammaneh,

Firstly, you should also include the lammps mailing list if you have problems, both so others with the same problem are aware, and others with more experience/the same issue can help. I’ve cc’d it in this email.

As to your issues:
1 - Yes you should use fix shake with TIP3P water, it is parameterized for that. The ideal settings can be found here in the lammps manual.
2 - Missing atoms error can come from a variety of issues, have you properly minimized your system and made sure there are no overlapping atoms? Also have you double checked that your input file is correct. Are the missing atoms from water or from the CNT? It might be worth testing on individual small sized systems to make sure your system is properly parameterized.
3 - I have not had a great deal of experience with fix wall, however (and please someone correct me if I am wrong) for wall/reflect you need a non-periodic direction that your wall is being positioned, and for NPT dynamics you need periodicity in all xyz directions, so this is probably where your error is coming from. Maybe try using a soft wall (fix wall/lj93 etc.) and keeping your system periodic. Again I have not used fix/wall much so I’m not 100% sure about this.

Regards,
Kamron

Hi Sammaneh,

Firstly, you should also include the lammps mailing list if you have
problems, both so others with the same problem are aware, and others with
more experience/the same issue can help. I've cc'd it in this email.

​more importantly: find somebody local that knows MD and can advise you.
there is more to learn about MD than what can be explained with an e-mail
and a few days is not a long time. it will take *much* longer and a lot of
experimenting and practicing to really "get" how to do proper simulations​.

As to your issues:
1 - Yes you should use fix shake with TIP3P water, it is parameterized for
that. The ideal settings can be found here
<http://lammps.sandia.gov/doc/Section_howto.html#howto_7>in the lammps
manual.

​...and always check the literature, if you don't know. don't trust some
random person you don't know.

2 - Missing atoms error can come from a variety of issues, have you
properly minimized your system and made sure there are no overlapping
atoms? Also have you double checked that your input file is correct. Are
the missing atoms from water or from the CNT? It might be worth testing on
individual small sized systems to make sure your system is properly
parameterized.

3 - I have not had a great deal of experience with fix wall, however (and
please someone correct me if I am wrong) for wall/reflect you need a
non-periodic direction that your wall is being positioned, and for NPT
dynamics you need periodicity in all xyz directions, so this is probably
where your error is coming from. Maybe try using a soft wall (fix
wall/lj93 etc. <http://lammps.sandia.gov/doc/fix_wall.html>) and keeping
your system periodic. Again I have not used fix/wall much so I'm not 100%
sure about this.

​i don't think that fix npt is a meaningful choice for such a system. in
general, this sounds like somebody trying to set up a simulation without
the proper background knowledge and there is no route more certain to waste
you and everybody else's time than doing simulations without knowing what
you are doing.​ like the song says: there ain't no escape from the blues.

axel.