ERROR: Pair hybrid sub-style is not used

Dear all,

I am testing my input script for a system which I recently created. I have amorphous carbon (modeled as tersoff), interacting with a polymer through LJ interactions. I plan to use hybrid style for the simulations. Atomtypes 1-12 interact using lj/cut/coul/long + bonded interactions while atomtype 13 (amorphous carbon) is modeled with tersoff. The amorphous carbon/polymer interactions are to be modeled as lj/cut.

First, I switched tersoff potential to real units from metal by modifying the SiCGe.tersoff files and rescaling eV (A,B) parameters.

When I am trying to run following script, I am getting following error.

“Pair hybrid sub-style is not used”. In the below script, both sub-style are used. I am wondering if it needs to be done differently. Any guidance will be highly appreciated. Two related question?

  1. Do I need to specify the lj interactions specifically within 1 to 12 atom types explicitly as well when I use hybrid pairstyle.

  2. Do I need to specify the pair_style command before reading the file?

testout is the file which is being read. It has Pair Coeffs for 13 atom types. (13th pair coeff is just a dummy).

Dear all,
I got it to work (but have to check if that makes sense) after

  1. removing pair_coeffs from the data file,

  2. specifying pair_style none before reading the file.

  3. specifying pair_style hybrid tersoff lj/cut/coul/long 12 12 (after reading the file)

  4. removing pair_coeff * * lj/cut/coul/long 12 12 line

  5. Adding all possible combinations of pair_coeffs (total 90) seperately.

Is there a better way to do this.

Best Regards,
Vikas

Dear all,

I am testing my input script for a system which I recently created. I have
amorphous carbon (modeled as tersoff), interacting with a polymer through LJ
interactions. I plan to use hybrid style for the simulations. Atomtypes 1-12
interact using lj/cut/coul/long + bonded interactions while atomtype 13
(amorphous carbon) is modeled with tersoff. The amorphous carbon/polymer
interactions are to be modeled as lj/cut.

First, I switched tersoff potential to real units from metal by modifying
the SiCGe.tersoff files and rescaling eV (A,B) parameters.

When I am trying to run following script, I am getting following error.

"Pair hybrid sub-style is not used". In the below script, both sub-style are
used. I am wondering if it needs to be done differently. Any guidance will
be highly appreciated. Two related question?
1. Do I need to specify the lj interactions specifically within 1 to 12 atom
types explicitly as well when I use hybrid pairstyle.

no. mixing rules will be applied, if mixing is supported by a pair style.

2. Do I need to specify the pair_style command before reading the file?

if it contains a PairCoeff segment, yes. if not, no.

Dear all,
I got it to work (but have to check if that makes sense) after
1. removing pair_coeffs from the data file,
2. specifying pair_style none before reading the file.
3. specifying pair_style hybrid tersoff lj/cut/coul/long 12 12 (after
reading the file)
4. removing pair_coeff * * lj/cut/coul/long 12 12 line
5. Adding all possible combinations of pair_coeffs (total 90) seperately.

Is there a better way to do this.

yes. put pair_style hybrid *before* reading the data file
keep your PairCoeff statements but prefix epsilon and sigma with
lj/cut and use "none" for type 13
then *after* reading the data file add the missing pair_coeff lines
including tersoff.

that should pretty much do what you wanted initially.

axel.