Hi, ALL
I have a question about the usage of TIP4P water model in LAMMPS.
The manual (http://lammps.sandia.gov/doc/Section_howto.html#4_8) says
the rigid bonds and angle off TIP4P water are handled by the fix shake command.
And the “fix shake” works only when the bond type and the angle type to be fix are
specified in the data file. To define the bond and the angle, a force constant is
needed. While the manual doesn’t give any numbers for these force constants for the
TIP4P water model. I am wondering that if the force constants can be an arbitrary
number if SHAKE is used to enforce the rigidity.
Thanks,
Guozhen
Hi, ALL
I have a question about the usage of TIP4P water model in LAMMPS.
The manual (http://lammps.sandia.gov/doc/Section_howto.html#4_8) says
the rigid bonds and angle off TIP4P water are handled by the fix shake
command.
And the "fix shake" works only when the bond type and the angle type to be
fix are
specified in the data file. To define the bond and the angle, a force
constant is
needed. While the manual doesn't give any numbers for these force constants
for the
TIP4P water model. I am wondering that if the force constants can be
an arbitrary
number if SHAKE is used to enforce the rigidity.
yes, they can be arbitrary, since fix shake will only read and use the
values for equilibrium distance and angle.
you may want to set them to fairly large values (10-100x larger than
typical values for equivalent flexible force field interactions), to be able
to use minimization (where fix shake is not available) and still maintain
an close approximation of the rigid structure.
cheers,
axel.
Dear Dr. Axel Kohlmeyer,
Since fix shake can only be applied to a cluster with a central atom bonded to max 4 atoms , hence fix shake cannot be applied to phenol or benzene ring. Am I right? Therefore, in order to give a close approximation of the rigid phenol ring, I have to set the force constants to 10-100 times larger?
Thank you.
Regards,
christopher
Dear Dr. Axel Kohlmeyer,
Since fix shake can only be applied to a cluster with a central atom bonded
to max 4 atoms , hence fix shake cannot be applied to phenol or benzene
ring. Am I right? Therefore, in order to give a close approximation of the
rigid phenol ring, I have to set the force constants to 10-100 times larger?
for a regular force field calculation, there would not be a need to keep
a phenyl ring rigid. the largest benefit would be in keeping any bonds
involving hydrogen atoms rigid (which is supported).
please note, that i was specifically referring to geometry optimizations,
where there is no impact from dynamics. using overly large force
constants in MD can be counter productive, since they will change
(blue shift) vibrational modes and thus require you to use a shorter
time step without any need or benefit.
the situation would be different, if you want to keep a complete molecule
rigid, for which you can then using the fix rigid style integrators.
overall, you have to provide a bit more motivation and background
for your question in order to get a qualified and more specific answer.
axel.