Relaxation of polystyrene at 300 K

Dear all,

I’m trying to obtain equilibration configuration of polystyrene at 300 K. I use following relaxation protocols: first CG energy minimization, followed by NVE relaxation, and finally heat up too 300 K using NPT ensemble. The detailed description can be found in the input. It is seen from attached “temperature.jpg” that the temperature keeps constant as 17 K during the NVE relxation, and gradually increases to 300 K and then fluctuates around 300 K during the NPT relaxation. This matches well with my expectation. While the volume of the system keeps unchanged during the NVE relaxation, however, it increases significantly during following NPT relaxation, as shown in attached “volume.jpg”. I also attach two snapshots, as initial.jpg and npt.jpg, to demonstrate the significant change in the system. Since the polystyrene system is stable during the NVE relaxation, I just can not understand why the volume changes so much in following NPT relaxation? Is there any wrong with my relaxation protocols or input? Additionally, how could I obtain the equilibration configuration of polystyrene at 300 K? Your comments will be highly appreciated!

Below you will find the input:

initial.jpg

npt.jpg

temperature.jpg

volume.jpg

Dear all,

dear <unknown>,

I'm trying to obtain equilibration configuration of polystyrene at 300 K. I
use following relaxation protocols: first CG energy minimization, followed
by NVE relaxation, and finally heat up too 300 K using NPT ensemble. The

for the record this is *not* npt ensemble, since you have one two open
surfaces.due to "m" boundary conditions. with that in mind, what is the
reason to use fix npt over just fix nvt?

detailed description can be found in the input. It is seen from attached
"temperature.jpg" that the temperature keeps constant as 17 K during the NVE
relxation, and gradually increases to 300 K and then fluctuates around 300 K
during the NPT relaxation. This matches well with my expectation. While the

in this case, you get out what you put in.

volume of the system keeps unchanged during the NVE relaxation, however, it
increases significantly during following NPT relaxation, as shown in
attached "volume.jpg". I also attach two snapshots, as initial.jpg and

volume is a bad parameter due to the way
how you have set up your system. you rather
need to determine whether you system is expanding
in y direction (with the free surfaces) or x and z.

with shrinkwrap boundaries it is quite possible, that
your system is expanding due to a single molecule
being expelled. as soon as it is outside of any
interaction cutoff, it will continue with constant
velocity (newton's 2nd law, remember?).

npt.jpg, to demonstrate the significant change in the system. Since the
polystyrene system is stable during the NVE relaxation, I just can not
understand why the volume changes so much in following NPT relaxation? Is

that 17K there is not much kinetic energy in your
system. my first guess would be that you are heating
up too fast and too far for the time step.

there any wrong with my relaxation protocols or input? Additionally, how

well, you didn't say what kind of polystyrene configuration
you are looking for, i.e. how ordered or disordered
you want it to have. due to the physics of the system,
i would not expect many conformational changes going
to happen quickly and at low temperature (and 300K *is*
still low). so if you want to keep following a simple "biosystem"
type relaxation protocol (which will mostly preserve the
initial structure, then you are on the right way, but may
need to run the temperature ramp slower and a smaller
time step and first and foremost need to find out what
exactly causes the expansion. you definitely will need
to run much longer simulations until you reach equilibrium
(you are not showing the total energy of the system.
that is something one needs to look at. also you should
continue the final configuration with nve to see, if you
can run a stable, energy conserving MD. a thermostating
integrator will hide a lot of flaws.

if you want an amorphous configuration, then you probably
want to first run the system in constant volume (p p p, or
p f p with walls) at higher temperature until the structure is
well disordered and then gradually anneal down to 300K
and then keep running until the total energy is constant,
remove the walls and switch to p m p boundaries.

could I obtain the equilibration configuration of polystyrene at 300 K? Your
comments will be highly appreciated!

it is really difficult to make comments, if you don't
tell us what you are equilibrating for. 300K doesn't
say much...

axel.