Special Bonds for lj in bonded molecule

LAMMPS LAMMPS Mailing List Mirror
1

Hi,

I have a question on the special bonds for LJ interactions in a molecule (atom type 1).

I am running a system with some of these bonded molecules.

My question is if i were to set “special bonds 1.0 0.0 0.0” and my LJ interaction was defined as

#LJ Potential
pair_style lj/cut 2.5

pair_coeff 1 1 0.0 1.0

pair_modify shift yes

Is the 1-2 interaction activated with a LJ potential having a well depth of 0 (atoms are bonded with a bond length smaller than the cutoff)? Does this has the same effect as setting “special bonds 0.0 0.0 0.0”?

Thanks

william

2

Hi,

I have a question on the special bonds for LJ interactions in a molecule
(atom type 1).

I am running a system with some of these bonded molecules.

please post your complete input. it is very difficult to judge what
you are asking by the fragments below. it looks like you are
completely misunderstanding the meaning of the special_bonds command.

My question is if i were to set "special bonds 1.0 0.0 0.0" and my LJ
interaction was defined as

#LJ Potential
pair_style lj/cut 2.5
pair_coeff 1 1 0.0 1.0
pair_modify shift yes
Is the 1-2 interaction activated with a LJ potential having a well depth of
0 (atoms are bonded with a bond length smaller than the cutoff)? Does this
has the same effect as setting "special bonds 0.0 0.0 0.0"?

if your LJ epsilon is 0.0, then you have *no* interaction at all.

axel.

3

Hi Axel,

Thanks for your reply. I think you have answered my question on the special bond command. Just to be doubly sure, top portion of my input script is shown below.

I have 2 atom types in the structure. Type 1 has LJ for 1-1 interaction. Type 2 is bonded harmonically. Type 1 and 2 are harmonically constrained by bonds and angles.

For my study, I sometimes want LJ for 2-2 to be switched on with different values of epilson to run different cases. For convenience, I only change the LJ epilson to different values (including 0) while keeping the “special bonds” set as 1 0 0.

My question is, by setting LJ epilson to 0, does that has the same effect as setting “special bonds” to 0 0 0 (i.e. only bonded interaction with no LJ)?

Thanks again.

william

#------------READ STRUCTURE-----------------------------------------------------------

units lj

atom_style angle

boundary f p p

read_data final.txt

group end1 molecule 2 10 12 #Fixed End 1

group mid molecule 1 3 4 5 6 7 8 #Simulation atoms

group end2 molecule 9 11 13 #Fixed End 2

#------------LJ Potentials------------------------------------------------------------------

###Potential

#####Intra-atomic Species

#LJ Potential

pair_style lj/cut 2.5

pair_coeff 1 1 1.0 1.0

pair_modify shift yes

#LJ Potential

pair_style lj/cut 2.5

pair_coeff 2 2 0.0 1.0

pair_modify shift yes

#LJ Potential

pair_style lj/cut 2.5

pair_coeff 1 2 1.0 1.0

pair_modify shift yes

#####Harmonic Potential

#between atoms in chain

bond_style harmonic

bond_coeff 1 ${HB1_S} 1.1

#between interface atom in chain and NC

bond_coeff 2 ${HB2_S} 1.1

special_bonds lj 1.0 0.0 0.0

#####Angles between atoms in Chain

angle_style hybrid harmonic cosine

angle_coeff 1 cosine ${HA1_S}

#Angles between NC and interface atom

angle_coeff 2 harmonic ${HA2_S} 109

4

Hi Axel,

Thanks for your reply. I think you have answered my question on the special bond command. Just to be doubly sure, top portion of my input script is shown below.

I have 2 atom types in the structure. Type 1 has LJ for 1-1 interaction. Type 2 is bonded harmonically. Type 1 and 2 are harmonically constrained by bonds and angles.

For my study, I sometimes want LJ for 2-2 to be switched on with different values of epilson to run different cases. For convenience, I only change the LJ epilson to different values (including 0) while keeping the “special bonds” set as 1 0 0.

My question is, by setting LJ epilson to 0, does that has the same effect as setting “special bonds” to 0 0 0 (i.e. only bonded interaction with no LJ)?

No.

Axel

5

Hi Axel,

Just to sure, so the correct way is to always set special bonds to 0 0 0 to remove the pair LJ interaction?

I tested both cases where

  1. Setting special bonds as 1 0 0 and LJ epilson = 0.0
  2. Setting special bonds as 0 0 0 regardless of epilson value.

They gave the same simulation answer (e.g. same pressure, temperature, total energy, KE, PE) through every timestep.

However, I notice that if I set LJ epilson = 0.0 and special bonds as 1 1 1, it does not produce the same answer. I guess this got to do with my structure and it was pure coincidence that earlier cases produced the same result. Any advices on this observation?

Thanks for your time,

william

6

Hi Axel,

Just to sure, so the correct way is to always set special bonds to 0 0 0 to
remove the pair LJ interaction?

emphatically no. this is a very elementary
thing of classical force fields.

the special_bonds command and setting epsilon
to 0.0 are *very* different operations. special_bonds
*only* affects pairs that *also* have a bond. setting
epsilon to 0.0 affects *all* pairs within the cutoff.

I tested both cases where

1. Setting special bonds as 1 0 0 and LJ epilson = 0.0
2. Setting special bonds as 0 0 0 regardless of epilson value.

They gave the same simulation answer (e.g. same pressure, temperature, total
energy, KE, PE) through every timestep.

that can only work if in your test case all
pairs of atoms affected by setting epsilon
to 0.0 also have a bond.

However, I notice that if I set LJ epilson = 0.0 and special bonds as 1 1 1,
it does not produce the same answer. I guess this got to do with my
structure and it was pure coincidence that earlier cases produced the same
result. Any advices on this observation?

see above.

axel.

7

Hi Axel,

Thanks for your quick reply and clarification.

I think i was not clear in my earlier explanation. Yes, i was referring to pairs of atoms that have bonds as well as LJ interactions when i use the special bonds in the 2 test cases.

Thanks again.

william