# "boundary" and "box"

Dear Steve and all LAMPPS users,I want to ask some questions about "boundary" and "box".
1) When I use f boundary with enough vacumm, the result is the same as the p boundary condition. However, the condition with s boundary or m boundary is different. I wonder where the s or m boundary is used.
2) A box contains two groups placed up and down with boudary "p p f", When I change the zhi, the force interaction between two groups is different. I don't how the boundary except p boundary influences the result and how to determine the box dimensions.
The manual says: If the system is non-periodic (in a dimension), then you should not make the lo/hi box dimensions (as defined in your region command) radically smaller/larger than the extent of the atoms you eventually plan to create, e.g. via the create_atoms command. For example, if your atoms extend from 0 to 50,you should not specify the box bounds as -10000 and 10000.

In the example,the bounds as -10 and 60 or -10 and 70 result in different result.
That's all. Thank you very much for explanation.
Regards,

Erchia

Dear Steve and all LAMPPS users,

1) When I use f boundary with enough vacumm, the result is the same as the
p boundary condition. However, the condition with s boundary or m boundary
is different. I wonder where the s or m boundary is used.

*what* result for *what* is different?
there can be lots of things different that have nothing to do with
what you claim.

2) A box contains two groups placed up and down with boudary "p p f", When I
change the zhi, the force interaction between two groups is different. I
don't how the boundary except p boundary influences the result and how to
determine the box dimensions.

this is all far to vague to comment on.

The manual says: If the system is non-periodic (in a dimension), then you
should not make the lo/hi box dimensions (as defined in your region command)
radically smaller/larger than the extent of the atoms you eventually plan to
create, e.g. via the create_atoms command. For example, if your atoms extend
from 0 to 50,you should not specify the box bounds as -10000 and 10000.

yes, because LAMMPS uses the box as the basis for its domain
decomposition and if your box extends far beyond the positions of the
atoms, you create a load imbalance.

In the example,the bounds as -10 and 60 or -10 and 70 result in different
result.

in *what* example?

The data document creates a box as follows:
0.000000000 85.2168 xlo xhi
0.000000000 34.4332 ylo yhi
-8.500000000 18.257102924 zlo zhi
This box contains a carbon nanotube above three layers of graphene. The boudary is set as boudary “p p f”. The ensemble is NVT. Then the simulation is that the carbon nanotube moves with an initial vx. When I change the zhi as 20 (a little larger) or 200 (some vacuum), the movement of the carbon nanotube is different. I don’t know what box can describe a carbon nanotube above three layers of graphene.

The data document creates a box as follows:
0.000000000 85.2168 xlo xhi
0.000000000 34.4332 ylo yhi
-8.500000000 18.257102924 zlo zhi
This box contains a carbon nanotube above three layers of graphene. The
boudary is set as boudary "p p f". The ensemble is NVT. Then the simulation
is that the carbon nanotube moves with an initial vx. When I change the zhi
as 20 (a little larger) or 200 (some vacuum), the movement of the carbon
nanotube is different. I don't know what box can describe a carbon nanotube
above three layers of graphene.

this is useless information, since nobody can reproduce this. i
seriously doubt that the box dimensions or choice of boundary
conditions are the culprit.

axel.

I don’t understand the boundary f, s and m through the manual. Are there some explanation about what f, s and m physically mean?
Regards,
Erchia

I don't understand the boundary f, s and m through the manual. Are there
some explanation about what f, s and m physically mean?

the explanation *is* in the manual. physically they are all them same,
i.e. there are no interaction beyond the box boundary. only with "f"
this boundary is fixed (i.e. an atom that is outside is ignored and
"lost") whereas with "s" the boundary is adjusted to that atoms do not
leave and "m" is the same as "s" but with a minimum dimension (with
"s" the box would be shrunk to the extent of the particle
distribution)

axel.

If it’s certain that the atoms will not be outside,will the “f”,“s” and “m” be the same?

``````
``````
``````Regards,
``````
``````Erchia
``````
``````
``````
``````
``````

If it's certain that the atoms will not be outside,will the "f","s" and "m"
be the same?

have you read what i just wrote?

axel.

If it's certain that the atoms will not be outside,will the "f","s" and "m"
be the same?

have you read what i just wrote?

i consider it extremely rude and offensive to ask exactly what i just
explained. congratulate yourself, that i managed to not return that
favor. if you don't believe me, make some simple tests. and if you
believe that LAMMPS is not behaving the way it should (or rather the
way it is documented), then make these (simple) tests available here.

axel.