hi lammps users ,
I need your help. I am simulating a system of BN nanosheet using lammps and tersoff potential parameters. I have gone through some research papers like http://www.sciencedirect.com/science/article/pii/S092150931400985X in which cutoff function used in tersoff potential has been removed by setting large cutoff distances such that value of cutofff function always remains equal to 1. How can I change cutoff radius value in lammps tersoff such that an initial bond list is created using original cut off radius but value of cutoff R used in tersoff potential file is large enough than original cut off. Lammps takes the value from tersoff potential file only. So when I change this value above 2.5 system collapses.So I want ot create bonds using original cut off but at the same times wants lammps to use large cutoff value in its calculations. Is there any way to do this?
Tersoff potential is a bond order potential with no predefined bonds so you should not worry about cutoff values when creating your structure file. You can change the 4th number from last in the tersoff potential file (BNC.tersoff) to a larger value, but please be aware that changing the cutoff effectively changes the potential function. Chances are likely you will include the 2NN or even 3NN. No guarantee there that any arbitrary number would work.
Thanks for your response Mr. Ray
I know how to change this value in BNC.tersoff. but I want to make potential function always equal to one. How can do that?
hi lammps users ,
I need your help. I am simulating a system of BN nanosheet using lammps and
tersoff potential parameters. I have gone through some research papers like http://www.sciencedirect.com/science/article/pii/S092150931400985X in which
cutoff function used in tersoff potential has been removed by setting large
cutoff distances such that value of cutofff function always remains equal to
1. How can I change cutoff radius value in lammps tersoff such that an
initial bond list is created using original cut off radius but value of
cutoff R used in tersoff potential file is large enough than original cut
off. Lammps takes the value from tersoff potential file only. So when I
change this value above 2.5 system collapses.So I want ot create bonds using
original cut off but at the same times wants lammps to use large cutoff
value in its calculations. Is there any way to do this?
the only way i see this happening is through writing your own C++ code
for a custom tersoff pair style.
since you have a reference paper, i would suggest you check the paper
or contact the authors to inquire which simulation code they used and
how this modification was applied. it is difficult to give good advice
on these things through indirect knowledge.
Thanks for your response Mr. Ray
I know how to change this value in BNC.tersoff. but I want to make
potential function always equal to one. How can do that?
This does not make sense to me. What do you mean by "potential function
always to one?" Do you mean the cutoff function?
OK. But when I when use large value of cutoff radius above 2.5 my structure collapses. The reference paper I have mentioned has done this but I don’t know how they did. It has said that this can be done by creating a bond list using original cutoff value of 2 but giving large enough value to cutoff value in LAMMPS potential file. And by not updating the neighbor list through while simulation. But I don’t know how to do that? Can I use two cutoff values one for creating bond and another large value in LAMMPS potential file?
No doubt, since they “fixed” or “froze” the bond order between atoms to their ideal values - an approach against the philosophy of Tersoff the potential.
My two cents here…
I agree with Ray. Yet, if you want to pursue it, you can ask the authors the Tersoff potential file that they used in LAMMPS and a sample input script to carry on your simulations.