Using Lebedeva interlayer potential

LAMMPS LAMMPS Mailing List Mirror
1

Dear LAMMPS users,

I am trying to use the newly implemented Lebedeva interlayer potential to model two graphene layers (atomic types 1 and 2) with nitrogen defects (atomic type 3) in one of the layers. To do so, I considered C-N interlayer interactions equal to C-C, duplicating this entry in the potential file:

C C 14.558 21.204 1.8 3.198 4.16 -0.862 0.10049 0.6 0.4 1.0
C N 14.558 21.204 1.8 3.198 4.16 -0.862 0.10049 0.6 0.4 1.0

In my first attempt, I used the following commands to define the pair coefficients:

pair_style hybrid/overlay tersoff lebedeva/z 14
pair_coeff * * tersoff BNC.tersoff C C N
pair_coeff 1 2 lebedeva/z CN.Lebedeva C C
pair_coeff 2 3 lebedeva/z CN.Lebedeva C N

By running LAMMPS, I got the following error:

ERROR: Incorrect args for pair coefficients (…/pair_lebedeva_z.cpp:222)

Looking at the source file, I found at line 222:

if (narg != 3 + atom->ntypes)
error->all(FLERR,“Incorrect args for pair coefficients”);

Then I changed the input file:

pair_style hybrid/overlay tersoff lebedeva/z 14
pair_coeff * * tersoff BNC.tersoff C C N
pair_coeff 1 2*3 lebedeva/z CN.Lebedeva C C N

Now, when I run LAMMPS, the error is:

ERROR: Potential file is missing an entry (…/pair_lebedeva_z.cpp:431)

My simple question is: can I use Lebedeva’s potential to do this kind of simulation? If so, what am I doing wrong? The current documentation does not clarify the correct usage of this interatomic potential.

Thanks in advance,

Roberto

2

please try without modifying the potential file. if you want to use the C atom parameters, you just map your nitrogen atoms to the C entry for that potential. LAMMPS doesn’t know about chemistry. it just follows the input and doesn’t care otherwise.

second. following the potential file in the pair_coeff command line have to be as many labels as there are atom types, i.e. 3 in your case. atoms that you don’t want to map have to be labeled with NULL. so it has to be either:

pair_style hybrid/overlay tersoff lebedeva/z 20

pair_coeff * * tersoff BNC.tersoff C C N
pair_coeff 1 2 lebedeva/z CC.Lebedeva C C NULL
pair_coeff 2 3 lebedeva/z CC.Lebedeva NULL C C

or:

pair_style hybrid/overlay tersoff lebedeva/z 20

pair_coeff * * tersoff BNC.tersoff C C N
pair_coeff 1 2*3 lebedeva/z CC.Lebedeva C C C

axel.

3

Dear Axel,

thank you for your reply.

please try without modifying the potential file.

This was the first thing I tried. And, as expected, it works.

if you want to use the C atom parameters, you just map your nitrogen atoms to the C entry for that potential. LAMMPS doesn’t know about chemistry. it just follows the input and doesn’t care otherwise.

Yes, I know, but I am also thinking of the possibility of modifying the potential such that it will contain parameters specific for N-N and C-N interactions.

Indeed, soon after I sent the e-mail to the list, I think I found how to fix the last problem. Here is the potential file:

C C 14.558 21.204 1.8 3.198 4.16 -0.862 0.10049 0.6 0.4 1.0
C N 14.558 21.204 1.8 3.198 4.16 -0.862 0.10049 0.6 0.4 1.0
N N 14.558 21.204 1.8 3.198 4.16 -0.862 0.10049 0.6 0.4 1.0
N C 14.558 21.204 1.8 3.198 4.16 -0.862 0.10049 0.6 0.4 1.0

It seems that entries for C-N and N-C are required.

MD is running now…

Best regards,

Roberto