Dear Aidan
Thank you for detailed reply. I really appreciate it!! Thanks.
I am still wondering is there a citation for the ffield.reax.mattsson parameterization. I guess that those parameters must be comes from somewhere. If there is a citation for that, can you please let me know. Thanks!
Also, I did a quick calculation on an arbitrary material consists of CHO elements by using the CHO force field. Here is the energy comparison
reax/c (with hbondcutoff = 6.0)
reax[1] reax[2] reax[3] reax[4] reax[5] reax[6] reax[7] reax[8] reax[9] reax[10] reax[11] reax[12] reax[13] reax[14]
-43272.22 -1509.8776 3.2248825 0 2474.4126 4.7127079 -0.080915062 -26.499655 758.516 -753.23029 12328.68 -615.63608 0 304.6614
reax (default, hbnewflag=tripflag=1)
reax[1] reax[2] reax[3] reax[4] reax[5] reax[6] reax[7] reax[8] reax[9] reax[10] reax[11] reax[12] reax[13] reax[14]
-43274.042 -1509.8776 3.2248825 0 2473.0463 4.7127079 -0.080914551 -26.499655 758.516 -753.23029 12328.68 -615.63442 0 304.6587
reax (required, hbnewflag = 1 and tripflag =0, this setting is suggested in the README.reax)
reax[1] reax[2] reax[3] reax[4] reax[5] reax[6] reax[7] reax[8] reax[9] reax[10] reax[11] reax[12] reax[13] reax[14]
-43232.103 -1509.8776 3.2248825 0 2473.0463 4.7127079 -0.080914551 -26.499655 758.516 -753.23029 12328.68 -615.63442 0 304.6587
I noticed that the reax with default setting gives a very similar result to that of reax/c. However, the reax with the suggested setting is quite different. At this point, it might be important to confirm that tripflag =0 is a required setting while using CHO force field. If you have any comments on this, please let me know. Thanks!
Thanks,
Sincerely,
Steve