I am using REAXFF and as I understood from GULP - Manual that REAXFF has two approach to calculate the charge. The default method and the alternative one, I just want to know what is the difference between the two approaches and dose the alternative one is the (QEQ). One more thing, dose both methods lead to a neutral-system?
Thank you,

The two methods in ReaxFF are algorithms for computing the charges via either 1) matrix diagonalisation, or 2) iterative minimisation. Both give the same answer (if the latter converges), but they differ in computational cost, scaling with system size and parallel performance.

Thank you Prof julian.
Dose the iterative minimization is the (QEQ) method?

I’m afraid your question doesn’t make sense to me. QEq is different to ReaxFF, but both can use iterative methods or diagonalisation.

Dear Prof Julian,
Thank you again, I really appreciated your help.
I just got a little bit confused because in LAMMPS we should to use QEQ method to run REAXFF. Now I understood that GULP using different approach.


Dear prof Julian,
I am running MD using REAXFF. in the restart file all the charges are zero. I am wondering if I have something wrong in the input file which lead to zero charge or if I should to mention any keyword to write the charge in the restart file.

the second question is that in the energy component the only energy I got is the ReaxFF force field
dose this energy include the coulomb interaction or because I have zero charge I have no separate coulomb interaction component. (60.0 KB)
ReaxFF_T14.out (138.1 KB)
T14_REAXFF_1.restart (197.6 KB)

I attached three files to this topic, the input, output and the restart file.
Thank you!

The important things here are:

  1. You do have charges since they are printed in your output
  2. It doesn’t really matter whether the charges are in the restart file since ReaxFF will re-generate them at every structure/timestep & so the restart will work just fine.
  3. At the start of your input there is a warning about parameters being missing for the elements you have and so make sure that everything is correct and you are sure that the missing parameters aren’t critical to your run.

Dear Prof Julian,
Thank you for such a kind assistant.