Dear Lammps-Users,
I am now trying to use two cutoffs for "coul/dsf",
in the framework of "hybrid/overlay".
pair_style hybrid/overlay buck 25.0 coul/dsf 0.1 25.0
pair_coeff 1 2 buck 18003.7572 0.205205 133.5381 5.0
pair_coeff 2 2 buck 1388.7730 0.362319 175 25.0
pair_coeff 1 1 buck 0 0.205205 0 25.0
pair_coeff 1 2 coul/dsf 5.0
pair_coeff 2 2 coul/dsf 25.0
pair_coeff 1 1 coul/dsf 25.0
This gives an error
"ERROR: Incorrect args for pair coefficients (../pair_coul_dsf.cpp:184)"
at the position of
"pair_coeff 1 2 coul/dsf 5.0"
This means that I could set two different cutoffs for "buck",
but the syntax for "pair_coeff coul/dsf" is somewhat different.
I have changed the syntax as
"pair_coeff 1 2 coul/dsf 0.1 5.0"
to accept the value of alpha, but the error message is the same.
Please help me if you know the solution.
Thanks in advance.
Kota Ito
The doc page for the coul/dsf pair style says
it does not take any arguments (other than itype jtype)
for its pair_coeff command. Hence the errors.
The cutoff must be specified with the pair_style command.
However you can use coul/dsf multiple times (with different
cutoffs) in the pair hybrid/overlay command. Ditto for
any substyle. See the pair hybrid doc page for details.
Whether that is a good idea and will be a valid model
for pairwise Coulombic interactions is a different Q.
Steve
Dear Lammps-Users,
I am now trying to use two cutoffs for "coul/dsf",
in the framework of "hybrid/overlay".
pair_style hybrid/overlay buck 25.0 coul/dsf 0.1 25.0
pair_coeff 1 2 buck 18003.7572 0.205205 133.5381 5.0
pair_coeff 2 2 buck 1388.7730 0.362319 175 25.0
pair_coeff 1 1 buck 0 0.205205 0 25.0
pair_coeff 1 2 coul/dsf 5.0
pair_coeff 2 2 coul/dsf 25.0
pair_coeff 1 1 coul/dsf 25.0
This gives an error
"ERROR: Incorrect args for pair coefficients (../pair_coul_dsf.cpp:184)"
at the position of
"pair_coeff 1 2 coul/dsf 5.0"
This means that I could set two different cutoffs for "buck",
but the syntax for "pair_coeff coul/dsf" is somewhat different.
yes, and that is actually documented in the manual.
I have changed the syntax as
"pair_coeff 1 2 coul/dsf 0.1 5.0"
to accept the value of alpha, but the error message is the same.
it is not a good idea to do this. please recall, that coul/dsf is an
approximation to long-range coulomb interactions, only that you don't
have an explicit kspace style.
for that to be consistent, *all* coulomb interactions must have the
*same* damping and the damping determines the cutoff. so the correct
choice here is to define one global cutoff for *all* coulomb
interactions based on the choice of alpha and then just use
pair_coeff * * coul/dsf
axel.
Hi Steve and Axel,
Thank you for your kind reply.
I understood that the same cutoff should be used for coul/dsf in general,
but there could be some rooms if I use "individual pair style multiple times"
in hybrid command.
Thank you again for your kind suggestions.
Best regards,
Kota Ito
----- e$B85$N%a%C%;!<%8e(B -----
e$B:9=P?Me(B: "Axel Kohlmeyer" <[email protected]>
[email protected](B: "Kota Ito" <[email protected]...>
e$BAw?.:Q$_e(B: 2017e$BG/e(B7e$B7ne(B7e$BF|e(B, e$B6bMKF|e(B e$B8a8ee(B 11:42:21
e$B7oL>e(B: Re: [lammps-users] Different cutoff for hybrid/overly coul/dsf
Dear Lammps-Users,
I am now trying to use two cutoffs for "coul/dsf",
in the framework of "hybrid/overlay".
pair_style hybrid/overlay buck 25.0 coul/dsf 0.1 25.0
pair_coeff 1 2 buck 18003.7572 0.205205 133.5381 5.0
pair_coeff 2 2 buck 1388.7730 0.362319 175 25.0
pair_coeff 1 1 buck 0 0.205205 0 25.0
pair_coeff 1 2 coul/dsf 5.0
pair_coeff 2 2 coul/dsf 25.0
pair_coeff 1 1 coul/dsf 25.0
This gives an error
"ERROR: Incorrect args for pair coefficients (../pair_coul_dsf.cpp:184)"
at the position of
"pair_coeff 1 2 coul/dsf 5.0"
This means that I could set two different cutoffs for "buck",
but the syntax for "pair_coeff coul/dsf" is somewhat different.
yes, and that is actually documented in the manual.
I have changed the syntax as
"pair_coeff 1 2 coul/dsf 0.1 5.0"
to accept the value of alpha, but the error message is the same.
it is not a good idea to do this. please recall, that coul/dsf is an
approximation to long-range coulomb interactions, only that you don't
have an explicit kspace style.
for that to be consistent, *all* coulomb interactions must have the
*same* damping and the damping determines the cutoff. so the correct
choice here is to define one global cutoff for *all* coulomb
interactions based on the choice of alpha and then just use
pair_coeff * * coul/dsf
axel.