Hi all,
I would like to perform a simulation of Styrene-Butadiene rubber (SBR) crosslinked with sulphur. For only SBR I used reaxff but here sulphur is added now.
So, which potential should I use? Any help would be appreciated.
Thank you.
Pinky
You should use ReaxFF parameters which are parameterized/optimized for the material system that you are studying. You could search for such parameters set in literature.
Sanjib
Hello Dr. Chowdhury,
I tried with pair_style reax/c NULL lgvdw yes
pair_coeff * * ffield.reax.lg C H S.
But it is not able to calculate the energy of the S-S bond. Would you please guide me on how to solve this problem?
Thank you.
Sincerely,
Pinky
Well, obviously, the force field file you are using was not parameterized for the system you are studying.
The comment in the file says: Reactive MD-force field: nitramines (RDX/HMX/TATB/PETN))+lg: L. Liu, Y. Liu, S. V. Zybin, H. Sun and W. A. Goddard, Journal of Physical Chemistry A, 115, 11016-11022 (2011). That is very different from what you are looking at.
There are many ReaxFF parameterizations in the published literature. You need one that is suitable and not just any file with the elements you need, since the parameterizations for ReaxFF are not very transferable. After all, this is not quantum chemistry, but still an empirical force force field, and a very complex one at that. Complex force fields have many more adjustable parameters, but they are also much less general.
Looks like you need to do a thorough search of the literature…
Axel.
But it is not able to calculate the energy of the S-S bond.
Do you mean that the ‘ffield.reax.lg’ force field doesn’t have S-S bond interaction parameters.
Sanjib
Thank you Dr. Kohlmeyer. I will recheck the literature.
Hello Dr. Chowdhury,
Please find the attached trajectory, There is a S-S bond in the main (.xyz) file. But, from the trajectory it is not visible. That’s why I am confused.
Thank you.
Pinky
Thank you Dr. Kohlmeyer. I will recheck the literature.
Hello Dr. Chowdhury,
Please find the attached trajectory, There is a S-S bond in the main (.xyz) file. But, from the trajectory it is not visible. That’s why I am confused.
You cannot judge bonding patterns from .xyz trajectory, especially one that does not contain proper element names. The file only contains the position information. How should the visualization software know what it was that you have simulated and thus correctly decide where to draw a bond and where not?
The bonding pattern will be generated by visualization programs, usually based on simple heuristics e.g. based on covalent radii. However, those are rather inflexible and arbitrary and don’t always represent the real chemistry. Just look at how complicated this can be with quantum chemistry calculations. Given that sulphur has a significantly larger covalent radius than carbon or nitrogen, the bonds are likely not as expected when the visualization software cannot correctly identify the elements. The color coding of your visualization also indicates that exactly that happened. If you use a tool like VMD, then it also simply cannot take into account how bonds are broken or formed during a trajectory since that is not part of the data model. That is except you use a “Dynamic Bonds” representation where you have to specify he bond cutoff explicitly (so it is an even worse and less flexible heuristic than using covalent radii).
The only reliable way to judge bonding patterns from ReaxFF calculations is to look at the bond order values and use those to analyze which atoms are bound to which others and thus what molecules are present. See, e.g. fix reax/c/species and fix reax/c/bonds.
This entire discussion suggests that you need to spend more time reading about ReaxFF in general and learning how ReaxFF calculations work.