[lammps-users] annealing a grain boundary using the fix npt command

LAMMPS LAMMPS Mailing List Mirror
1

Hi all,

I'm trying to anneal a grain boundary in silicon. My annealing
"protocol" is as follows: using a constant (zero) pressure ensemble
that allows the x, y, and z direction allow to change independently, I
equilibrate the system at T = 300 K, 600 K, 900 K, and 1200 K for
50,000 md steps each. I then decrease the temperature continuously
from 1200 to 300 K over 1 million MD steps

For step 1, I use the following npt fix:
fix 1 all npt temp 300.0 300.0 .1 aniso 0.0 0.0 1000.0

and it returns this error:
ERROR: Illegal fix npt command

What am I doing wrong in my fix?

Thanks,
Zack Cordero

1st half of my input script:
#Annealing grain boundaries in Silicon

#initialization
boundary p p p
units metal
echo screen
newton on
log log.si_anneal

#create atoms
atom_style atomic
read_data position1.xyz

group all type 1
pair_style tersoff
pair_coeff * * Si.tersoff Si
neighbor 2.0 bin
neigh_modify every 1
dump 1 all xyz 100 position.xyz
thermo 100

velocity all create 400 429349 dist gaussian
fix 1 all npt temp 300.0 300.0 0.1 aniso 0.0 0.0 1000.0

timestep .01

run 50000

2

Dear Zach,
fix npt syntax has changed recently. Please see the doc page.

3

Thanks for replying. I triple-checked here
(http://lammps.sandia.gov/doc/fix_nh.html) before I e-mailed out to
the list.

According to the example on the doc page the input should go like
fix 1 water npt temp 300.0 300.0 100.0 iso 0.0 0.0 1000.0

The only difference between that one and mine is that I changed the
Tdamp to .1 and the iso to aniso.

Am I missing something?

Thanks,
Zack

4

Dear Zack, when you use aniso, you have to to starting and ending pressure in all three directions as they are being barostated independently. So, try this out
fix 1 water npt temp 300.0 300.0 100.0 aniso 0.0 0.0 0.0 0.0 0.0 0.0 1000.0

5

The WWW page docs refer to the most current version
of LAMMPS. Are you running the most current version
of LAMMPS? The syntax for fix npt changed in April,
as I recall (see the bug page).

Steve