[lammps-users] How to define the potential of SiO2 and graphene

LAMMPS LAMMPS Mailing List Mirror
1

Dear all,

I’m a beginner because I recently started studying lammps.
I would like to place graphene on the SiO2 substrate(α-quartz) for mitigation.
The following is my setting method, is there any mistake?
If you make a mistake, please tell me a good way.

Type 1 is C
Type 2 is O
Type 3 is Si.

pair_style hybrid airebo 3.0 tersoff lj/cut 2.5
pair_coeff * * CH.airebo C NULL NULL
pair_coeff * * SiO.tersoff NULL O Si
pair_coeff 1 2 lj/cut 0.006734 3.275
pair_coeff 1 3 lj/cut 0.008890 3.628

Thanks,

Rusty

2

please note that there are two kinds of “correctness” that need to be considered.

  1. if you input has the correct syntax
  2. if your input represents a meaningful model

point 1) is easy to confirm from the documentation and just running it, because LAMMPS will refuse incorrect syntax. but an input with a correct syntax can still descriobe a very unphysical and meaningless model.
point 2) is much more difficult to determine, definitely not in a binary yes/no way, and even less so from remote.

your input looks correct syntactically, but that is easy to confirm for you as well. the rest is tricky. your cutoff of the lj/cut interaction looks very short for the choice of parameters (it cuts of before the minimum) and more if you used cut and paste from an input using reduced units instead of metal units.
getting a meaningful model with pair style hybrid and from mixing 3 types of potentials is particularly difficult. you cannot just treat pair styles like lego bricks assuming they are all compatible and fit together. the question of what is a correct model is something that is not really what this mailing is about and rather something you need to discuss with your adviser/supervisor/tutor and learn about from text books and published literature.

axel.

3

Thanks, Axel.
I want to model an AFM experiment, so I’ll experiment with your advice.

Another question is, when simulating the contact-AFM model and considering the frictional force, do you consider the interaction between the AFM tip and the SiO2 substrate? Or can it be ignored?

2020年12月24日(木) 20:34 Axel Kohlmeyer <[email protected]>:

4

representing an AFM experiment with classical MD simulations is a very tricky business because you are connecting time and length scales of macroscopic objects with their microscopic behavior.

while it is straightforward to represent the surface and part of the tip by with atoms, you also have to consider that the entire cantilever, on which the tip is mounted, is huge compared to the typical MD system size and thus it has a giant moment of inertia. this will make any motion of the tip - however gentle on the macroscopic level - become some kind of wild slashing with very high force. not to mention that your system will usually only be large enough to encompass a small part of the tip. also, you need to consider the time scale of the motion of the AFM tip versus the time scale of the simulation.

Thanks, Axel.
I want to model an AFM experiment, so I’ll experiment with your advice.

Another question is, when simulating the contact-AFM model and considering the frictional force, do you consider the interaction between the AFM tip and the SiO2 substrate? Or can it be ignored?

all interactions matter unless they would be outside the cutoff or have tapered off so much that they are insignificant.

axel.