I used MS to build the alkane system model and used the COMPASSII force field. But the result is not good, may I ask how to realize the combined atomic force field like TraPPE-UA in lammps？Look forward to your answer！
for any force field you can do the following:
- look up the original publication(s) describing the force field
- compare the potential functions in that publications to the one that LAMMPS has to offer. that will tell you what pair, bond, angle, dihedral style to use. the paper will also provide the necessary parameters (or coefficients in LAMMPS-language).
- if the style is not available, you can often emulate it through tabulation (which requires some experimenting and care to get it to work well and sufficiently accurate)
- check of other requirements. TraPPE for example, as a monte carlo force field has requirements that are difficult to implement in a parallel code (all bonds between carbons are supposed to be rigid)
- make test simulations trying to reproduce results in the original publications to make sure you are implementing that force field of your choice correctly.