LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] nonperiodic boundary condition for wall

# Re: [lammps-users] nonperiodic boundary condition for wall

 From: Axel Kohlmeyer Date: Thu, 27 Jul 2017 07:16:17 -0400

On Thu, Jul 27, 2017 at 3:46 AM, SHABNAM GHAHREMANIAN wrote:
​Hi Dear all

I want to simulate a nano-channel with two Platinum walls in the Z direction (up and down) and argon gas between the walls.
I use periodic boundary condition in the x and y direction and non-periodic and fixed boundary in the z direction.
The input file is as follows:

units              metal
boundary        p p f
atom_style      full

# Atoms definition
lattice             fcc 3.920  orient x 1 0 0 orient y 0 1 0 orient z 0 0 1 &
origin 0 0 0

region             box  block -25 25   -25  25   -5  205  units box
create_box      2 box

region               up block -20 20   -20 20   201  204  units box
create_atoms    1 region up
group                up region up

region               down block -20 20   -20 20   -4  -1 units box
create_atoms    1 region down
group               down region down

lattice              fcc 20
region              ARGON block -20 20 -20 20  0 200  units box
create_atoms    2 region ARGON
group               ARGON region ARGON

mass              1 195.08
mass              2 39.95

velocity          down set 0.0 0.0 0.0
velocity          up set 0.0 0.0 0.0
velocity          ARGON create 114 4928459 dist uniform mom yes  rot yes

pair_style        hybrid eam  lj/cut 4.5
pair_coeff        * * lj/cut 0.0104 3.400
pair_coeff        1 1 eam Pt_u3.eam
pair_coeff        1 2 lj/cut 0.007611 3.2425

fix                  freeze1 up setforce 0.0 0.0 0.0
fix                  freeze2 down setforce 0.0 0.0 0.0

I have two questions:
1-
If the f boundary condition is used in the z direction, then the interaction between the particle walls must be considered together?(for example EAM potential)
In fact, do particles of platinum interact with each other?

​forces for the platinum atoms are computed, however, due to the use of fix setforce, they are ignored. in fact, for your specific setup there is no point in using eam, you could use pair style zero. the only force contributions that matter for time integration are between type 1 and either type 1 or type 2.​

2-
when f boundary condition is used this error occurs: Lost atom
What is the reason for this error?

​axel.​

```--
This email was Anti Virus checked by  Security Gateway.```

------------------------------------------------------------------------------
Check out the vibrant tech community on one of the world's most