Dear altruists
I have a system for bi-layer graphene and 100 CO2. I want to observe the adsorption phenomenon of the CO2 on graphene. I have build the data file by VMD and here is my input file-
units real
atom_style full
boundary p p p
box tilt large
read_data data.graphene-100-CO2
set atom 2508 type 3
set atom 25095016 type 4
set type 2 charge -0.3256
set type 1 charge +0.6512
set type 3 charge 0.0000
set type 4 charge 0.0000
special_bonds amber
dihedral_style none
pair_style lj/cut/coul/cut 10
pair_coeff 1 1 0.0558971 2.757 10.0
pair_coeff 2 2 0.1599810 3.033 10.0
pair_coeff 1 2 0.0945654 2.895 10.0
pair_coeff 3 3 0.0556407 3.4 10.0
pair_coeff 4 4 0.0556407 3.4 10.0
bond_style harmonic
bond_coeff 1 480.00 1.420
bond_coeff 2 450.00 1.149
angle_style harmonic
angle_coeff 2 147.7 180
angle_coeff 1 90.00 120
group graphene1 id 1:2508
group graphene2 id 2509:5016
group carbondioxide id 5017:5316
neighbor 2.0 bin
neigh_modify delay 0
compute myRDF all rdf 1000 34 12
fix 6 carbondioxide ave/time 100 1 100 c_RDF file graphene.rdf mode vector
dump 1 all custom 2000 graphene-CO2.dump id type xs ys zs ix iy iz
dump 3 all dcd 2000 graphene.dcd
dump 4 all xyz 1000 graphene.xyz
dump_modify 1 pbc yes
dump_modify 3 pbc yes
thermo_style multi
thermo 100
fix 1 all nvt temp 273.0 273.0 100.
fix 2 graphene1 rigid/nvt single temp 273.0 273.0 100
fix 3 graphene2 rigid/nvt single temp 273.0 273.0 100
fix 4 graphene1 move linear 0 0 0
fix 5 graphene2 move linear 0 0 0
minimize 1e-4 1.0e-8 5000 100000
timestep 1
run 5000000
I am facing some problems like -
-
in output file there is a warning one or more atoms are time integrated more than once (…/modify.cpp:278) , I suppose this warning for the use of fix rigid and move linear comments together, if so then how much it will influence my calculation? could anyone help me to resolve this problem?
-
In case of RDF calculation I want to calculate the distribution of the CO2 with the respect of graphene. So in the compute command I have separated them according to the types of the atom. Is it the right way for calculate the phenomena which one I want?
I am not sure about my mistakes, It will be great helpful for me if anyone mention me about my mistakes.
Thanks in advance
Dear altruists
dear <insert popular brand name of american manufacturer of recliners
!
I have a system for bi-layer graphene and 100 CO2. I want to observe the
adsorption phenomenon of the CO2 on graphene. I have build the data file by
VMD and here is my input file-
[...]
fix 1 all nvt temp 273.0 273.0 100.
fix 2 graphene1 rigid/nvt single temp 273.0 273.0 100
fix 3 graphene2 rigid/nvt single temp 273.0 273.0 100
fix 4 graphene1 move linear 0 0 0
fix 5 graphene2 move linear 0 0 0
minimize 1e-4 1.0e-8 5000 100000
timestep 1
run 5000000
I am facing some problems like -
1. in output file there is a *warning one or more atoms are time
integrated more than once (../modify.cpp:278) , *I suppose this warning
for the use of fix rigid and move linear comments together, if so then how
much it will influence my calculation? could anyone help me to resolve
this problem?
there is another overlap,too. and yes, it will ruin your simulation,
destroy your hard drive and give you serious headaches.
seriously, this kind of issue is discussed in the manual and also has been
discussed many,many times on this mailing list. i suggest you help yourself
and re-read the relevant parts of the manual and search the mailing list
archive and then think for yourself. what you have here in your input for
time integration makes no sense.
2. In case of RDF calculation I want to calculate the distribution of the
CO2 with the respect of graphene. So in the compute command I have
separated them according to the types of the atom. Is it the right way for
calculate the phenomena which one I want?
more importantly, you should ask yourself, whether a radial distribution
function, a descriptor devised for (homogeneous) bulk system, is a good
choice to describe the "distribution of CO2 with respect to graphene" in
the first place. in fact, your formulation of what you want to compute is
quite vague, so you might want to take a step back and do some basic
research and more clearly define what you really want. i.e. look at what
kind of distributions are possible and then consider which of them is
meaningful for the information that you want to gather from your
simulation(s). your system geometry is not exactly new (there are probably
more studies with water instead of CO2, but as far as analysis goes, they
are equivalent) and thus there should be a wealth of existing publications
from which you can learn. hell, people have looked at the distribution of
molecules on (structured) surfaces before i entered grad school...
axel.