Zn–Cl–C–H–O ReaxFF parameterization

LAMMPS
1

Subject: Looking for a ReaxFF parameter file for Zn/Cl/C/H/O systems (ZnCl₂–lactic acid DES)

Dear all,

I am preparing LAMMPS ReaxFF simulations to investigate the interactions between a ZnCl₂-based deep eutectic solvent (DES) and lignin. The DES consists of ZnCl₂ and lactic acid, and my objective is to study the reactive interactions between the solvent and lignin during biomass pretreatment.

The reactive elements required in the force field are:

  • Zn

  • Cl

  • C

  • H

  • O

I searched the available ReaxFF parameter files distributed with LAMMPS and the SCM ReaxFF database. While I found parameterizations for systems such as CHOZn, ZnOH, CHON, and CHONSMgPNaCuCl, I could not find a single validated ReaxFF parameter file that simultaneously includes Zn, Cl, C, H, and O for organic systems.

Could anyone advise whether there is a publicly available LAMMPS-compatible ReaxFF parameter file suitable for Zn/Cl/C/H/O chemistry? In particular, I am interested in parameterizations that have been used for:

  • ZnCl₂-based deep eutectic solvents

  • ZnCl₂ interacting with oxygenated organic molecules

  • ZnCl₂–biomass or ZnCl₂–lignin systems

  • Molten ZnCl₂ or ZnCl₂-containing reactive simulations

I understand that simply matching the element set is insufficient and that any candidate force field must be validated against the chemistry of interest. Before production simulations, I plan to validate the force field by examining Zn–Cl coordination, Zn–O interactions with lactic acid, and available structural or thermodynamic data.

If anyone knows of a published parameter file, repository, or research group that has developed a Zn–Cl–C–H–O ReaxFF parameterization, I would greatly appreciate your guidance.

Thank you for your time and assistance.

2

@bhaswati_bharadwaj thanks for writing a lucid and well written post where you properly describe your problem, provide sufficient detail and context, and explain what you have done so far. I wish more people would follow your example.

Sadly, there are two problems with this:

  1. It is not really a LAMMPS question even though you plan to use LAMMPS. This is about the science of your research and not LAMMPS itself and thus it is - strictly speaking - off-topic for this category. Something like Science Talk would be more suitable.
  2. Experience from many years of posting to mailing lists and forums tells me that this kind of inquiry yields a helpful response only in extremely rare cases. Your best chance is to do a thorough search of the published literature (just checking databases of potentials may not be sufficient) but most likely you will need to identify a suitable collaborator that you can work with to create a suitable parameter set.
    Instead of ReaxFF you may want to explore how well machine learning potentials could be trained and applied to your project. By now there is a wide variety of machine learning potential schemes available with different levels of accuracy and computational cost. It may not be your first choice, but worth considering if you come to the conclusion that you will need a new ReaxFF parameterization so the effort of creating training data and validation would be the same.
3

Should I drop the plan to use reaxff for this project ?

4

Only you can answer this question.