I built a crosslinked membrane for using it in MD simulation. Since the crosslink is suppose to be via chemical reaction and the charges in reaction site supposed to be changed compared to the charges I assigned for the monomer before crosslink. So, I have decided to assign new charges for the whole crosslinked structure. There are some tools which can assign partial charges in different ways like gasteiger or bond increment method. If I would like use gasteiger partial charges with dredding forcefield, is it resonable? I mean is this kind of charges resonable for the MD sinulation?
As a general note: each force field is parametrized to work best with particular partial charges, and using different ones is usually a recipe for disaster.
If I remember correctly Dreiding works well with qeq charges or DFT-derived Mulliken charges. Gasteiger charges work with dreiding for some systems (e.g. http://pubs.acs.org/doi/abs/10.1021/ie800502m) but in general I would be cautios to use them.
Thank you very much Bartosz. I do see the paper suggested some system for Gasteiger charges with dredding. Any insight how it will workout for a system of Poly-amide membrane?
Whatever you do to generate the charges, you will have to validate your model. It depends on what the purpose of your MD simulation would be. If you are looking for geometrical features, you have to simulate a known structure and check its conformational stability by comparison with experimental data. Other characteristics may be more difficult to verify, because you will be confronted with a plethora of parameters to be adjusted. In one or other way, however, all static partial charges in mechanical force fields are wrong. The danger to over-interpret mechanical force fields is pretty big. You will have to verify that your model is able to represent the particular characteristics you are chasing for. There is no universal recommendation, nor a magic wand for generating partial charges for your problem.
- Badges
- Science topic