LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
[lammps-users] about a model
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[lammps-users] about a model

From: "htc" <619771202@...1204...>
Date: Mon, 6 Nov 2017 20:38:56 +0800

if i model all particles as ellipsoids ,and take the sphere as special case of ellipsoid.
that means i should use"hybrid dipole ellispoid" , and that means i cannot use"fix nve/sphere update dipole" to update the dipole with the"sphere"?
if i also wanna to update the dipole,what should i do?


On Mon, Nov 6, 2017 at 4:08 AM, htc <619771202@...1204...> wrote:
i wanna a model that contain sphere with dipole and ellipsoid without dipole,how do i use the command"atom style"?
Can i write this "atom_style  hybrid dipole sphere ellipsoid  " ?
Because i have to use the "fix nve/sphere update dipole" to describe sphere with dipole.
But for "hybrid",when you set the arguments of each atom,you have to list them as follows "atom-ID  atom-type x y z sub-style1 sub-style2...".
So i just donnot how to set the args,if i have to use "fix nve/sphere update dipole" to update the dipole set onto sphere within a system which is a mixture of sphere with dipole and ellipsoid without dipole.

​you have to provide all properties for all atoms, even if they are not used by the individual pair style or fix.
in your particular case, the *real* worrying question is: how are you going to describe the interactions between the spheres and the ellipsoids?
the only consistent solution that i can imagine would be to model all particles as ellipsoids, as a sphere is just a special case of an ellipsoid.​ same for the dipole, you can just make the dipole vector (0,0,0) for particles without dipole.