Dear all,
If I would like to set up an inhomogeneous system, for example, a water-co2 interface simulation, can I just randomly create water and co2 molecules at the desired regions, and then minimize and equilibrate the system quickly with a dissipative thermostat (like langevin thermostat)? Or should I prepare a pre-equilibrated water box and pre-equilibrated co2 box at the desired sizes and then combine them together? I wonder if the two systems (with different set-ups ) after equilibration would have the same statistical properties at the end. It will be great if anyone has simulated or considered those problems could provide some answers and comments.
Thank you very much for your time.
Regards,
Pengyu
Dear all,
If I would like to set up an inhomogeneous system, for example, a water-co2
interface simulation, can I just randomly create water and co2 molecules at
the desired regions, and then minimize and equilibrate the system quickly
with a dissipative thermostat (like langevin thermostat)? Or should I
prepare a pre-equilibrated water box and pre-equilibrated co2 box at the
desired sizes and then combine them together? I wonder if the two systems
(with different set-ups ) after equilibration would have the same
statistical properties at the end. It will be great if anyone has simulated
or considered those problems could provide some answers and comments.
combining pre-equilibrated systems is one way.
you can also just alternately apply time integration fixes only to one
part of the system at a time.
axel.
p.s.: for water/co2 things depend strongly on the conditions. under
standard conditions, there is no point in equilibrating the CO2
system, since there will be only few molecules (assuming that your
system is large enough to contain any).
Thank you very much Axel for the answers. I guess that I cannot minimize and equilibrate the two liquid phases (or more than two) together in one system then if they are not pre-equilibrated. Is this due to the possibly unstable interface interactions caused by a highly unequilibrated system, since combining pre-equilibrated systems also requires an equilibration afterward (but this is possibly closer to the equilibrated state than the former)?
Sorry that I should have mentioned the system would be under the high-pressure condition and co2 can be supercritical.
Regards,
Pengyu
Thank you very much Axel for the answers. I guess that I cannot minimize and
equilibrate the two liquid phases (or more than two) together in one system
then if they are not pre-equilibrated.
you are guessing wrong. i stated the exact opposite.
Is this due to the possibly unstable
interface interactions caused by a highly unequilibrated system, since
combining pre-equilibrated systems also requires an equilibration afterward
(but this is possibly closer to the equilibrated state than the former)?
if you want to preserve the initial phase separation, you must process
each part of the system individually until the very last step.
if you don't care, you can just run.
Sorry that I should have mentioned the system would be under the
high-pressure condition and co2 can be supercritical.
FYI, supercritical != liquid.
axel.
Sorry again for the confusion. I meant that I cannot equilibrate the two phases together with time integration applied to both phases at the same time, as you said I should apply the time integration to each part at a time. However, now I know that it depends on if I would like to preserve the initial separation or not.
Thank you again for your great comments and suggestions.
Regards,
Pengyu
I should summarize Huang’s question:
What’s the best (both scientifically and computationally) approach to equilibrate a system of two phases neighboring each other, in this case, Water/H2O? either both phases at the same time or each one individually?
I should summarize Huang’s question:
What’s the best (both scientifically and computationally) approach to equilibrate a system of two phases neighboring each other, in this case, Water/CO2? either both phases at the same time or each one individually?
I should summarize Huang's question:
What's the best (both scientifically and computationally) approach to
equilibrate a system of two phases neighboring each other, in this case,
Water/CO2? either both phases at the same time or each one individually?
like in so many cases in research, the answer is: it depends
axel.