[lammps-users] WRONG water density with Pair style reax?

Dear all,

I was simulating water (just O and H in the system) with pair style reax (with ffield.reax.cho) and kspace none. There are about 2000 water molecules in the system, first I do NPT ensemble for 10ps, then NVT ensemble for another 10ps, with timestep 0.25fs. I found that the final density of water, which is also stable at the last steps of NPT ensemble, is about twice as much as bulk water.

Did anyone else noticed this?

Regards,Xiong Wei

Dear Xiong,

This force field was not designed for bulk-water. It described the water molecule
correctly (as a combustion gas-phase product) but it indeed is not correct for
the bulk. Please see Chenoweth et al (JPC-A 2008) to check what this force
field was trained for. More recently published ReaxFF parameter sets
(e.g. van Duin et al., JPC-A 2010) are suitable for bulk water (although these
underestimate the bulk density by about 5%).

Best regards,


Adri C.T. van Duin
Associate Professor
Department of Mechanical and Nuclear Engineering
Pennsylvania State University
136 Research East Building
Bigler Road
University Park PA 16802 USA
phone: (814)-863-6277