# Parameterize a complex that contains Transition metal (Build Force Field)?

It is a general question. I know, but I need some help.

I am interested to parametrize a complex that contains transition metal like Ru and Pt etc. Furthermore, I will use this Force Field to do Molecular Dynamics with DNA and maybe later QM/MM. I saw some Literature, but they don't tell me how can I do it. So I am trying to find a Tutorial.

• If the size of the system is not very large (say < 5000 atoms), I guess the best way is to use an existing force field that supports metals, e.g. GFN-FF (onlinelibrary.wiley.com/doi/10.1002/anie.202004239). This will be more expensive than a user-parameterized force field, but saves you from the sheer effort to parameterize a new force field. Feb 17, 2022 at 12:35
• Thank you for your response, The complex itself around 150 atoms, but the system is more than 5000 atoms. I will take a look on this article Feb 17, 2022 at 12:42

## 1 Answer

I think that the parametrization process is straightforward (but very hardworking!):

• select and use an ab initio/DFT software that works well for your metal atoms,
• take a look at the force field equations that your selected molecular dynamic software uses,
• simulate the interactions between your metal with all the other atoms in your system using the ab initio/DFT (for example, for bonds/angles, you will need to run some scan of energy versus coordinates),
• fit the results from ab initio/DFT to the force field equations to get the corresponding parameters,
• add the parameters to the force field database files.
• Thank you for your answer. But I have two questions, What do you mean by scan of energy? Is it scan for the bond ? What do you mean by fit the results to force field equations? Feb 21, 2022 at 13:56
• In the output of DFT software, you will find the contribution to energy of different mechanism (van der Waals, Colulomb, etc.). You start with two atoms apart a distance $d_0$. Then you calculate the system energy for that distance ($E_0$). Then you change the distance to $d_1$ and calculate the energy, $E_1$, and so. At the end you will have energy terms vs d. Now you look for the force field expression for van der Waals, and fit the previous results with that expression to obtain the parameters related to both atoms. Take a look here.
– Camps
Feb 21, 2022 at 18:43
• This would work well for intramolecular terms, but it would be pretty hard to get good long range terms, wouldn't it? I am mainly thinking of the LJ terms. The charges are the first thing calculated, so they are fairly simple Jun 21, 2022 at 12:25