[ESPResSo] Problems with lennard-jones and lj-gen commands

[Nazar Ileri]

Dear all,

I am getting different results using exactly the same system but changing the way of defining lennard-jones interaction. 

To clarify, I used two copies of sample script "lj_liquid.tcl". For the second script, just changed the interaction command:

"inter 0 0 lennard-jones $lj1_eps $lj1_sig $lj1_cut $lj1_shift $off"

 to

"inter 0 0 lj-gen  $lj1_eps $lj1_sig $lj1_cut $lj1_shift $off  $e1 $e2 $b1 $b2".

All parameters are kept the same, i.e.,

lj1_eps=1.0,
lj1_sig=1.0,
lj1_cut=1.12246, 
off=0, 
e1=12, 
e2=6, 
b2=b1=1, 
and lj1_shift is calculated by [calc lj_shift $lj1_sig $lj1_cut].

After running the scripts for 1000 time steps I have had different particle configurations. Attached please find a plot showing non-bonded energies for the two cases. It seems like there is an off-set (the energy output of regular lennard-jones is slightly higher than the generic lennard-jones). Any ideas on the problem?

Thanks in advance,

nazar

PS: I am using version 2.1.2j.

non-bonded-comp.docx
Description: application/vnd.openxmlformats-officedocument.wordprocessingml.document