Modeling spheres

Hi LAMMPS users,

I want to make a system having two types of sphere. The type 1 sphere will have a radii ®, r=L. The type 2 will have r=2L. Which is the simple way to make this model? I though that using atom_style sphere and use set to set the diameter, but I don’t know if this is the right way.

Hi LAMMPS users,

I want to make a system having two types of sphere. The type 1 sphere will
have a radii (r), r=L. The type 2 will have r=2L. Which is the simple way to
make this model? I though that using atom_style sphere and use set to set
the diameter, but I don't know if this is the right way.

why don't you just make a simple test and see if it works the way you need?

axel.

HI,all,
I also concern this question,actually, I have a proposal, you can define the two types of sphere as two types of atom(such as 1 and 2), and the interaction potential between them(11,12,22) are known(which may not exist in lammps). and you must be careful with the range of interaction which should be out of the range of diameter of sphere.

Liu

If the original post means the diameter of atom_style sphere,
then that is set on a per-atom basis, in the data file. It is
not a function of the “type” of the atom.

Steve

Hi Steve,

That means if I have 10 sphere of type 1, using the atom_style sphere, I have to specify the radii of each one using set?

yes, or if you input the geometry thru a data file, then the diameter

must be included for each atom.

Steve

Thanks Steve

after re-reading the thread, i have the suspicion, that there is
perhaps a bit of a misdirection here.

alexa, what kind of interaction model do you want to have for your
"spheres". do you want particles that do have friction on the surface
(like granular particles), or just "oversize" atoms of a given
diameter? you didn't specify what kind of interaction you want to have
between those spheres and how "long" the interaction is supposed to be
relative to the diameter of the spheres. these are very important
factors that determine the approach. please keep in mind that almost
all potentials in LAMMPS represent spherical particles. they are just
more or less "sticky" and have a different length and strength of
interactions and are softer or harder. if you want "extended"
particles, you need atom style sphere, but you can also just stick
with point particles. there also are the colloid package models that
represent large clusters of small particles with a single interaction
potential (through integration over the small potentials).

so please be more specific on your model, then people can be more
specific in their advice.

axel.

Actually I want model oversize atoms interacting via LJ potential. Now, How this change what you have been said before?

Actually I want model oversize atoms interacting via LJ potential. Now, How
this change what you have been said before?

for that you could simply use a lennard jones potential with a large
enough sigma or switch to the colloid pair style.

axel.