Hello,
sorry i sent this request just to Axel, and not to the mailing list:
one more question about changing the values of the LJ-Potential during
a simulation.
In my data file i have fixed:
1 lj/charmm/coul/long 0.25 3.5635948725613571
2 lj/charmm/coul/long 0.10939999991572773 3.3996695084507409
3 lj/charmm/coul/long 0.015700000098461422 2.4713530426421655
4 lj/charmm/coul/long 0.01570000002623629 2.6495327872602221
5 lj/charmm/coul/long 0 0
then in the input file i call:
pair_coeff 1 1 lj/charmm/coul/long 0.275625 3.53473431107704
pair_coeff 2 2 lj/charmm/coul/long 0.12061349990709 3.37213653279448
pair_coeff 3 3 lj/charmm/coul/long 0.0173092501085537 2.45133824326477
pair_coeff 4 4 lj/charmm/coul/long 0.0173092500289255 2.62807495980059
pair_coeff 5 5 lj/charmm/coul/long 0 0
run 500000
is this the right way to do it?
I ran a simulation, where i increased the potential from +1% to +20%
(with 500000 steps in between) but i don't saw any changes. And now
i'm not sure if lammps acepted the changes. Is there a way to find
out? Or could you say me if i did it right
Thank you very much
If you are running pair_style hybrid, then you need the lj/ch/coul/long
in all the pair_coeff commands, otherwise not.
You don't need to set the coeffs in both the data file
and input script. One will just overwrite the other. Of
course you can run a first simulation using what was
in the data file, then change the values in the input
script, and run a 2nd simulation.
Steve
Yes i'm using pair stype hybrid
you said one will overwrite the other, does the input file overwrite
the data file or vice versa?
and, do i have to run a minimization or something else for changes
taking effect?
its because i don't see any changes by changing this values, and i
wana be sure its not me failing to do the script.
manual,
please set up and post a minimal example
for what you are trying to do that demonstrates
what is not working as expected (it doesn't have
to be realistic, just small and the absolute
minimum of commands so it is easy to look over).
Yes i'm using pair stype hybrid
you said one will overwrite the other, does the input file overwrite
the data file or vice versa?
later input will override the former.
any XXX_style command will generate an
instance of a class and the XXX_coeff statements
(including those embedded in a data file) will initialize
and then modify its settings.
defining a new XXX_style will delete the previous
class and all settings with it are lost.
and, do i have to run a minimization or something else for changes
taking effect?
its because i don't see any changes by changing this values, and i
wana be sure its not me failing to do the script.
it is hard to give advice on this without seeing a complete input.
axel.
Hi
This are the relevant parts of my input scripts:
units real
boundary p p p
atom_style full
pair_style hybrid buck 10.0 lj/charmm/coul/long 10.0 12.0
bond_style harmonic
angle_style harmonic
dihedral_style harmonic
special_bonds amber
kspace_style pppm 1.0e-6
neighbor 2.0 bin
compute Temp_layer layer temp
compute E_kin_layer layer ke
compute Eb metal group/group layer
thermo_style custom step spcpu c_Temp_layer c_Eb etotal pe
c_E_kin_layer epair evdwl ecoul elong emol ebond eangle edihed fmax
fnorm press
thermo 500
# Gold-Organic interactions
pair_coeff 1 5 buck 1470164.0 0.2235 6723.8
pair_coeff 2 5 buck 14166 0.2885 253.8
pair_coeff 3 5 buck 6742.1 0.2575 71.1
pair_coeff 4 5 buck 6742.1 0.2575 71.1
fix fshake layer shake 1e-4 20 0 t 3 4
fix 7 layer nvt temp 300 300 10
fix_modify 2 temp Temp_layer
#.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- Sigma: 0% Epsilon: 0%
-.-.-.-.-.-.-.-.-.-.-.-.-.-
pair_coeff 1 1 lj/charmm/coul/long 0.25 3.56359487256136
pair_coeff 2 2 lj/charmm/coul/long 0.109399999915728 3.39966950845074
pair_coeff 3 3 lj/charmm/coul/long 0.0157000000984614 2.47135304264217
pair_coeff 4 4 lj/charmm/coul/long 0.0157000000262363 2.64953278726022
pair_coeff 5 5 lj/charmm/coul/long 0 0
run 500000
#.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- Sigma: 0% Epsilon: 4%
-.-.-.-.-.-.-.-.-.-.-.-.-.-
pair_coeff 1 1 lj/charmm/coul/long 0.255 3.56359487256136
pair_coeff 2 2 lj/charmm/coul/long 0.111587999914042 3.39966950845074
pair_coeff 3 3 lj/charmm/coul/long 0.0160140001004307 2.47135304264217
pair_coeff 4 4 lj/charmm/coul/long 0.016014000026761 2.64953278726022
pair_coeff 5 5 lj/charmm/coul/long 0 0
run 500000
# and from here it goes on changing the values and running 0.5M steps
I'm simulating self assebled monolayers on a gold surface, and a
wanted to test the impact of changing the Van der Waals interactions.