Hello LAMMPS community,
I need to study the sintering of metal nanoparticles on surface. I am using EAM potential for the M-M interaction, and Morse potential for M-Surface interaction, and none for the surface using Pair_style hybrid.
The issue is that I should remove the bonds of the nanoparticles from the data file to simulate the sintering, however, from the first steps, the metal atoms are scattered everywhere in the box.
The EAM potential is describing well the particle alone, withoyt bonds and the particle is stable during the simulation.
Can someone help me to figure out what can be the reason for this behavior ?
Thank you
Hello,
Your post is quite unclear, why do you have bonds at all? You entire message is quite confusing because it seems that you have identified the origin of the issue, but are also asking for “the reason for this behavior”.
Simon
Hello,
Thank you for the reply
As a first step, and in order to verify the stability of the nanoparticle on the surface I had to define the bonds and consider the nanoparticle as rigid (otherwise I had the atoms scattering, as if the EAM potential cannot hold the atoms together).
And to enable the sintering process (coalescence of two particles), the bonds should not be defined.
OK, well there can be many reasons for that particular issue: badly defined potential, bad topology with overlapping atoms, bad dynamics (such as having two commands integrating the equations of motion at the same time). But it is impossible to guess without details about your input and LAMMPS version (c.f. forum guidelines).
Thank you for the reply.
The input file I am using (for the nanoparticle equilibration) is the following : (I am considering the surface as fixed slab)
units metal
atom_style full
dimension 3
boundary p p p
read_data sys1.data
**pair_style hybrid eam/alloy morse 8.0 **
pair_coeff * * eam/alloy Co.eam.alloy Co NULL NULL
pair_coeff 1 2*3 morse 0.5 1.4 2.714 8.0
pair_coeff 23 23 none
group TiO2 type 2 3
group Cob type 1
neighbor 2.0 bin
neigh_modify every 1 delay 0 check yes
velocity TiO2 set 0.0 0.0 0.0
velocity all create 300 12345 dist gaussian
run 0
velocity all scale 300.0
fix freeze TiO2 setforce NULL 0.0 0.0
compute inter Cob group/group TiO2 pair yes
timestep 0.0001
thermo 100
thermo_modify flush yes
**thermo_style custom step temp c_inter pe ke ecoul etotal elapsed dt time press density vol **
fix 1 Cob nvt temp 300.0 300.0 100.0
dump d1 all custom 1000 dump.300 id type x y z
dump_modify d1 sort id
##
restart 1000 nvt300-1.rst nvt300-2.rst
run 300000
write_data nvt300.data
LAMMPS reads input from top to bottom, so the command " velocity TiO2 set 0.0 0.0 0.0" is erased by velocity all create 300 12345 dist gaussian. Therefore since you also cancel the force on atoms with initial velocities, it explodes.
Edit: careful, with the metal unit system, time is in pico second, therefore your time constant for the fix nvt is 100 ps !!!
You mean it will explodes the Co atoms ?
The system must explode, because atoms are moving like imperturbable bullets.
In my simulation, I have the TiO2 surface stable and only the cobalt atoms are exploding
From what I see, this is surprising. However there seems to be many issues with that input so you should take a step back an simulate simpler system before dealing with a relatively advanced system like this one.
Thank you for taking the time and reply my messages. I appreciate it