problems of harmonic bond

Hi,

May I know is there any difference in LAMMPS for the following two cases, considering a bead-spring chain molecule in the solvent?
(1) There is no bond assigned on the chain molecule at all.
(2) There are harmonic bonds assigned for connecting near beads (1-2 body), but the bond strength (K) is zero.

More specifically, is there any additional constrain added on a chain molecule once there is a bond assigned. No matter how strong it is?

The attachments are the related tests where I can reason only if the above two cases are different. System is some 4-bead chain molecules dissolved in the water. I started with the same configuration for the following two tests, where chain molecules are randomly distributed.
(1) in.ao-nb and data.ao-nb are inputs for the chain molecules assigned no bonds at all (they were built with chained, but I removed the bonds)
(2) in.ao-zb and data.ao-zb are inputs for bonded chain molecule, and the bonding energy is zero.

While running the simulation in LAMMPS, the output pressures from two cases are different (~95 for case 1 and ~93.5 for case 2). While visualizing the trajectories; in case (1), all beads are reasonably flushed away, since there is no spring at all. In case (2), I assumed it would be the same, since the bonds are weak as nothing, and all beads should be flushed away with the breaking of bonds. However, the beads in a chain are squeezed to small balls. You can visualize the attached dump files using VMD in the drawing method VDW with sphere scale 0.2.

I ran these tests because of same errors in my simulation. Systems are similar to the above, but chain molecules are created with the bonds in the equilibrium bond length. What puzzles me is that some additional forces from the program are acting on the chain molecules,causing the squeezed chain molecules just as the testing case (2) above.

Sincerely,
Ming-Tsung

data.ao-bz (497 KB)

data.ao-nb (485 KB)

dump.ao-bz.xyz (1.57 MB)

dump.ao-nb.xyz (1.54 MB)

in.ao-bz (866 Bytes)

in.ao-nb (790 Bytes)

Hi,

May I know is there any difference in LAMMPS for the following two cases,
considering a bead-spring chain molecule in the solvent?
(1) There is no bond assigned on the chain molecule at all.
(2) There are harmonic bonds assigned for connecting near beads (1-2 body),
but the bond strength (K) is zero.

More specifically, is there any additional constrain added on a chain
molecule once there is a bond assigned. No matter how strong it is?

no. but when defining a bond, the corresponding non-bonded
interactions are by default excluded. please see the documentation for
the special_bond command for details.

axel.

Thank you so much Axel,

I did notice the page before submitting questions, but I didn’t add it in my input simply because the pairwise potential I used (DPD) is not specified in the keywords (amber or charmm or dreiding or fene or lj/coul or lj or coul or angle or dihedral or extra) on the webpage (http://lammps.sandia.gov/doc/special_bonds.html). I tried using special_bond dpd, but it is an illegal input.

I also tried to use special_bonds lj, and the output is normal. However, may I know it is correct to use with pair style DPD? Could you kindly exemplify the proper command for the following case, say I have pair_style DPD, and 1-2 body harmonic bonds and 1-3 body harmonic bonds on each atom. I guess that I should use
special_bonds “keyword” 1.0 1.0 0.0
but what is my keyword here? I would appreciate your answer.

Sincerely,
Ming-Tsung

Thank you so much Axel,

I did notice the page before submitting questions, but I didn't add it in my
input simply because the pairwise potential I used (DPD) is not specified in
the keywords (amber or charmm or dreiding or fene or lj/coul or lj or coul
or angle or dihedral or extra) on the webpage
(http://lammps.sandia.gov/doc/special_bonds.html). I tried using
special_bond dpd, but it is an illegal input.

I also tried to use special_bonds lj, and the output is normal. However, may
I know it is correct to use with pair style DPD? Could you kindly exemplify
the proper command for the following case, say I have pair_style DPD, and
1-2 body harmonic bonds and 1-3 body harmonic bonds on each atom. I guess
that I should use
special_bonds "keyword" 1.0 1.0 0.0
but what is my keyword here? I would appreciate your answer.

you have read the page, but you didn't understand it.
my answer is: read it again. there is no "just click here" in research.

axel.

Dear Axel,

I apologize for the careless reading. I missed a sentence, “The LJ factors are applied to the remaining terms that the potential calculates, whether they represent LJ interactions or not.” so I didn’t realize in this command LJ means van der Waals forces, but not LJ potential as it usually does. Thank you for the suggestion.