Run stops without error - Reax/c


I was trying to run ReaxFF with TiO2 system that including 1000 H2O molecules.
The input and data used are as attached. And program ended with no error printed.
The program is ended after the line, (1.36 KB) (348 KB)

fail.log (348 KB) (1.36 KB)

successful.log (1.16 MB)

I tried different cases to gather some info and tell it here.
(As a test) the simulations that stop with Reax/c can easily and fast run with a Class II FFs. So, I guess there shouldn’t be a fundamental problem with the system.

The problem can happen in interactive runs as well and no error was shown.

Generally two methods could help the system run with less chance of stopping:

1- Using high values in this command:
pair_style reax/c lmp_control checkqeq yes safezone 3 mincap 500

2- Using small time steps such as 0.25 fs

I guess this should be related to check bondings during running :slight_smile:


If you are certain that:

(1) error output is not being eaten by the system (less likely with interactive runs)
(2) the behavior is reasonably reproducible

I would like to take a look. If the answer to both (1) and (2) is yes, please post an example that stops quickly e.g. small values of mincap and safezone.