I'm not a specialist for reactive MD, but since you're trying to model an etching process maybe something like ReaxFF could be suitable. But I don't know if there is a set of proper parameters for SiO2.

thanks Markus,I am beginners in molecular dynamics. I am using LAMMPS to simulate my model and I still looking the parameters for SiO2 using ReaxFF or Airebo. For now I try to use simpler potential.

In your opinion,is Morse potential suitable for etching process?

Thank You

well, since the Morse potential includes bond dissociation I guess it is a proper way to start. But in LAMMPS as far as I know, the Morse potential is a pair potential, i.e., it applies only to non-bonded atoms.

How do you model your silica? do you include covalent terms or is it purely non-bonded with pair- or manybody- interactions?

I still searching for the right way to model the silica.for interaction between argon and silica atoms i using lennard jones.i want to start with simple potential first before i using more complex potential.

Can you suggest potential that include the covalent term?

I think the covalent term would be available in either a ReaxFF potential or something similar to an EAM potential. However, I don't know if there are any parameters for SiO2 for an embedded atoms model.

Maybe you want to have a look at this thesis:

https://helda.helsinki.fi/bitstream/handle/10138/23267/molecula.pdf?sequence=1

As far as I could see it deals with some aspects you try to do. The Si/SiO potentials are described in chapter 4.3.3

I hope that helps!

Okay I try study the thesis,you have been really really helpful.

Thank You!

maybe you could try Tersoff potential.