Can someone help point me in the direction on the best way to model metallic oxides? Can this be achieved by simply using the lattice command or does a data file need to be read?
Ben
You mean create a metal oxide structure. Both will do.
Ray
Can someone help point me in the direction on the best way to model metallic
oxides?
Can this be achieved by simply using the lattice command or does a
data file need to be read?
hmm... looks to me you are asking the second question before the
first. generating the structure is a straightforward procedure and
which of the two methods is preferable, depends on the "shape" you
need.
however, the major questions is: what type of model (i.e. potential)
do you intend to use and have you found suitable parameters for that.
as far as LAMMPS is concerned, you are just feeding it point particles
with some math that allows it to compute forces between them and thus
compute their motions. it doesn't know whether it is an oxide, a
metal, an isolator, or whatever else.
there is little use to worry about geometries, if you don't have an
answer to this question.
axel.
I intend on using the BKS potential such as:
pair_coeff 1 1 silica.tabulated BKS_1_1
pair_coeff 1 2 silica.tabulated BKS_1_2
pair_coeff 2 2 silica.tabulated BKS_2_2
I am able to model it by reading in the data file sio2.data.
I was not sure if modeling silica was even possible with the lattice command but if it is pretty straightforward, perhaps I need to look at the doc page again.
Thanks for the help,
Ben
I intend on using the BKS potential such as:
pair_coeff 1 1 silica.tabulated BKS_1_1
pair_coeff 1 2 silica.tabulated BKS_1_2
pair_coeff 2 2 silica.tabulated BKS_2_2I am able to model it by reading in the data file sio2.data.
I was not sure if modeling silica was even possible with the lattice command
but if it is pretty straightforward, perhaps I need to look at the doc page
again.
sure.
you define a lattice with the Si positions only. create a box
reserving space for two atom types and create the Si atoms. then you
define a (compatible) second lattice with the O positions only and
create the O atoms. done.
but then again, using the "inorganic builder" plugin in VMD and
writing out a data file with topotools may be just as straightforward.
once LAMMPS has the system info (provided it is correct), there is no
difference.
axel.
I was not sure if modeling silica was even possible with the lattice command
You’re confusing two operations here. All the lattice command would
do for you is create a perfect lattice (if used with create_atoms) with
SI on some sites and O on others. The use of a pair style
has nothing to do with the lattice command. You can use a pair
style with a data file of input coords that are created w/out the
lattice command.
Steve
Axel,
Thanks for the advice on the inorganic builder - it works really well. When converting from pdb to lammps data file, the “q value” was missing. Is there a way to convert the file to include that value or do I need to edit it in txt?
Ben
Axel,
Thanks for the advice on the inorganic builder - it works really well. When
converting from pdb to lammps data file, the "q value" was missing. Is there
a way to convert the file to include that value or do I need to edit it in
txt?
a pdb does not contain this value. you have to load the .psf file as well.
but why even write out a pdb. you can just write out the data file
right away. all the information should be internally stored in VMD.
and with more accuracy. the pdb standard massively truncates the
coordinate resolution.
axel.
I was following these instructions at the bottom of the page from this archive:
[http://sourceforge.net/mailarchive/message.php?msg_id=30514701](http://sourceforge.net/mailarchive/message.php?msg_id=30514701)
I loaded the pdb and psf file and then converted it using your software, topotools. It then wrote out a data file that can be used for LAMMPS. I am not sure if I can specify the q-value as 0 or if I need to calculate partial charges and implement those.
Ben
I was following these instructions at the bottom of the page from this
archive:
but you don't need to do that. just look at the topotools docs.
blindly following some instructions without comparing them to the
relevant documentation will make you miss a lot of good opportunities.
...and you - by far - won't learn as much about the tools that you are
using and that you have to learn when to trust and when not to.
axel.
I checked the doc pages and figured it out.
However, I took the amorphous silica form VMD, assigned it a tersoff potential in LAMMPS and implemented simulation.
For some reason, the amorphous silica is vibrating quite a bit (more than the atomic radius of the molecule).
My assumptions were that it would be more rigid. Is tersoff not an appropriate potential?
I checked the doc pages and figured it out.
However, I took the amorphous silica form VMD, assigned it a tersoff
potential in LAMMPS and implemented simulation.
For some reason, the amorphous silica is vibrating quite a bit (more than
the atomic radius of the molecule).My assumptions were that it would be more rigid. Is tersoff not an
appropriate potential?
is your input appropriate? is the system properly equilibrated?
axel.
My input is below. The data file was taken directly from the VMD data for amorphous silica.
units real
atom_style full
#read_restart restart.silica
read_data practice.data
pair_style tersoff
pair_coeff * * …/…/potentials/SiO.tersoff O Si
fix 1 all box/relax iso 1
minimize 1e-6 1e-6 300 300
velocity all create 300.0 25484595 dist gaussian
write_restart restart.silica
timestep 1.0
thermo 100
fix 2 all nve
fix 3 all langevin 300.0 300.0 100. 699483
dump 1 all xyz 100 dump.xyz
dump 2 all image 100 image.*.jpg type type &
zoom 1.2 adiam 1.2
dump_modify 2 pad 6
run 30000
write_restart restart2.silica
My input is below. The data file was taken directly from the VMD data for
amorphous silica.units real
do you have a custom tersoff potential file made for "real" units? if
not, then you probably need to use "metal" units and change the rest
of your input accordingly (especially the timestep and thermostat
setting). have a look at the tersoff docs and the units command.
axel.
Thanks Axel,
That was the problem. Units strike again
Ben
Thanks Axel,
That was the problem. Units strike again
.... and did you also see, that when you're asking better questions
and provide suitable information, you get better (and friendlier)
answers? 
axel.
Yep! Thank you for your patience - LAMMPS is a steep learning curve