Hello,
I am a beginner in this field so please bear with the following query:
Please advise whether the unit cell structure of SiO2 (or more complex molecules eg. borosilicate) be setup by the following approach:
I do not have access to a molecule drawing software.
Thanks
Tez
Please advise whether the unit cell structure of SiO2 (or more complex
molecules eg. borosilicate) be setup by the following approach:
1) introduce the constituent atoms in the whole number ratios of their molar
fractions within a box at random locations
Do you wish to create a "random" distribution of Si and O atoms with a
ratio of 1:2?
Note this will *not* auto-magically give you an amorphous SiO2.
2) apply potentials corresponding to the various atoms (tersoff in case of
SiO2)
Tersoff is designed for crystalline SiO2. I don't think it describes
amorphous, nor "random", SiO2 very well.
3) equilibrate the system.
No. Following the above steps, you will not get "the unit cell
structure of SiO2".
Ray
Hello,
I am a beginner in this field so please bear with the following query:Please advise whether the unit cell structure of SiO2 (or more complex
molecules eg. borosilicate) be setup by the following approach:
1) introduce the constituent atoms in the whole number ratios of their molar
fractions within a box at random locations
2) apply potentials corresponding to the various atoms (tersoff in case of
SiO2)
3) equilibrate the system.
this sounds more like you are trolling this mailing list.
what you should do is:
- look up the crystal structure
- build a corresponding unit cell
in a data file (can be done with
a text editor) read it into lammps
using the read_data command and
use the replicate command to build
a large enough structure to your liking.
- or use the lattice/crystal structure
building tools integrated into LAMMPS
- run the equilibration
I do not have access to a molecule drawing software.
this is *beyond* being ridiculous. apparently
you have access to the internet, so you *do*
have access. it may not be the one software
that you have in mind, but it may be sufficient
for your needs anyway.
axel.
Hi tezeswi,
Hello,
I am a beginner in this field so please bear with the following query:Please advise whether the unit cell structure of SiO2 (or more complex
molecules eg. borosilicate) be setup by the following approach:
1) introduce the constituent atoms in the whole number ratios of their molar
fractions within a box at random locations
2) apply potentials corresponding to the various atoms (tersoff in case of
SiO2)
3) equilibrate the system.
If you are after generating a crystalline structure (alpha quartz in
the case of SiO2 at ambient conditions)
then the answer is no. If that was your goal just considering the
above idea suggests that you need to go and read some solid stat
physics book. In addition, always keep in mind the application you are
intending to simulate before choosing your potential (cheap vs
expensive model in terms of computational demands). As to the drawing
software avogadro ( http://avogadro.openmolecules.net/wiki/Main_Page)
is free assuming you can manage to install it on your system.
Carlos
Dear Tez,
Ideally, you would get your hands on some software that may help you build coordinates in general and crystal structures in particular. Some options include moltemplate (comes bundled with lammps I believe). For your system I think you can use some of the built in lammps commands to create crystalline SiO2.
An alternative is aten which produces output that can be easily converted to work with lammps. All the software mentioned here is open source and therefore free. You should be able to obtain them.
Best of luck,
Salomon
Dear All,
Thanks so much for your input and for welcoming me to the lammps-users mailing list.
–Tez