I would echo the suggestion to contact the author, who has posted a tutorial but it lacks a lot of detail, starting from a pre-constructed topology. Note that this force field and its intended use is very different from “normal” GROMACS workflows and you cannot use pdb2gmx for it. Rather, topologies of surfaces and materials should be generated with x2top, which allows for the processing of generic species based solely on distances between the atoms.
We try to simulate nanocoposite comprising the substrate of aluminum oxide (Al2O3) and binder (different polymers (PS, PC, PMMC, etc.)). However, the CLAYFF force field has a limited number of bonded parameters to define the interactions for polymer binder. Is it possible to combine (without any significant problems) the CLAYFF force field with any other atomistic force field to simulate polymer-metal oxide composites in Gromacs?
It is doable to combine Clayff with other ff’s, but it has not really been designed for it. It is typically used with GAFF or OPLS as far as I know. As an alternative, the INTERFACE ff by Hendrik Heinz is supposed to be compatible with CHARMM/CGenFF, but overall it is likely very system dependent, and hence the best approach in mixing inorganic and organic ff’s is likely to try out a few different combinations of them, (like Clayff or Interface ff mixed with GAFF, OPLS, CGenFF), and then see how sensitive the results are to the choice of combination of ff’s.
What forcefield did you have in mind for the polymers? You can contact me directly if you need to see some examples of Clayff/Interface ff’s mixed with organic ff’s.
Thank you for your response. Sorry for the delay in responding. Last year, we attempted to implement Clayff for Al2O3 and OPLS FF for the polymer. We should continue our work. Perhaps you could provide a suggestion: is it possible to implement INTERFACE or the Clayff force fields for computer simulation of ZrO2 or TiO2 surfaces?
Hi,
Neither ZrO2 or TiO2 is supported by Interface or Clayff, but one can try to derive something like a Interface-like or Clayff-like ff parameters for ZrO2 and TiO2. Ther question would be what type of properties would you primarily want to reproduce for ZrO2 and TiO2? Perhaps its best generating an entirely tailor-made ff for them?
Hello @mholmboe,
I will conduct simulations to model the interactions between polymer chains and the surfaces of ZrO2 or TiO2. In addition, I need these parameters to simulate the interaction between polymers and the MOF UiO-66, which consists of Zr-type atoms. I aim to investigate the thermal, mechanical, and transport properties. The polymers are mostly generated utilizing the GAFF or OPLS FF methodologies. I lacked sufficient expertise in calculating force field parameters and verifying their accuracy. Could you provide an example where parametrization was executed?
Hi,
There should already be some forcefields for these oxides in the literature I guess. Else worst case I could try to parameterize (email me). The problem is to verify the compatibility of the combination of forcefields. Do you have any experimental reference data for this coombination of inorganic/organic components/system?
/Michael
Can you please share Clayff data with me? I can only find the files on the previously posted link (GitHub - thomasunderwood/ClayFF: A GROMACS implementation of the ClayFF force field) which is lacking in a lot of files and doesn’t work.
I am trying to simulate electroreduction of CO2 on Cu surface in water solution. It seems to me that building a Clayff would be the best way to go about that in Gromacs, so it would mean a lot if you can send the ff data. And if you have any advice for my system I would be really grateful.
As you may be aware of Clayff does not have parameters for Cu or CO2, so you will need to find and verify those yourself. If you are new to Gromacs, this could be a bit challenging to set up… I know CO2 has been modelled in Clayff mineral based systems in a few prior studies, which could be a good starting point… for Cu Im not so sure but there are likely some cases in the literature. You could also look into the Interface FF forcefield which atleast have support for some metals.
Anyhow, I have a Github repo where I share some Clayff cases, slides, and MATLAB script library that we use to setup input files for mainly Gromacs. See the [ATOM_scripts_lecture] (atom/ATOM_scripts_lecture at master · mholmboe/atom · GitHub) folder. Note that our examples/scripts called CLAYFF_2004 has the original atomtypes from Cygan et al 2004, whereas the examples or scripts simply called CLAYFF has an extended number of atomtypes, with also modified atomtype names. There is some documentation in the form of a slides and html pages, also available here [Atomistic Topology Operations in Matlab – Molecular Geochemistry at Umeå University], and even a link to a OpenAi GPTs tutor we recently have played around with… if that could be to any help…
I have downloaded your Clayff files and added Cu and CO2 parameters to atomtypes, atommass and ffnonbonded.itp. First I wanted to simulate just the copper slab structure I built using Avogadro- I have .pdb file of this copper slab. I also decided to simulate the interactions between copper atoms with LJ potential (it has been used as a good approximation for embedded atom potential which is not available in Gromacs). So there are no bonded interactions in my slab. The problem occurs here:
when I ran the gmx pdb2gmx command to create .gro and .top from the .pdb of Cu slab I encountered the problem which said “there is no .rtp file in Clay.ff”. So I figured I have to create .rtp file for Cu slab which I did but it does not work. For bonded interactions I just put nulls because there is none (and it says the bondedtypes section is mandatory so I had to write smth there). I also specified again the LJ potential between Cu atoms but this line had a problem so I had to delete it. It doesn’t work so I don’t know what I’m doing wrong.
I read about .rtp file in the Gromacs manual and it just says you have to specify the bonded interactions which I don’t have. I really don’t know how should this file look like for a metal slab with no bonded interactions. Any help would be great!